Media Player .NET (MPDN) - D3D HQ GPU Video Renderer [v2.49.0/v1.31.0 27 Dec 2018]

[aufkrawall]

I unfortunately can't test with AMD.

Btw: I just used the updater for the installed version for the first and I'm really impressed how well designed the routine is.
Not bad, guys!

[Belphemur]

I unfortunately can't test with AMD.

Btw: I just used the updater for the installed version for the first and I'm really impressed how well designed the routine is.
Not bad, guys!

I'm happy to read this!
I redesigned it completely to make it the easiest to use and as informative as possible :)

[videonerd]

Hi Zachs, I am using the latest 64 bit version on 8.1 + nvidia 970. I notice that when I open MPDN, i get a split second spark / screen corruption on the screen. Also when I close the application, a much larger split second spark / screen corruption as it closes - any idea what may be causing this? No screen corruption during playback or otherwise though.

Thank you!

[Zachs]

Does the screen corruption go beyond MPDN window? If it does, it's almost certainly an issue outside of MPDN's control. Could be the driver or hardware.

[Zachs]

I'm happy to read this!
I redesigned it completely to make it the easiest to use and as informative as possible :)

Great job Belphemur! I too love the new updater.

BTW, MPDN now has an Extensions API version (4 currently) - I think we could use this now to match the extensions to the player. Any thoughts on how we could do this? I was thinking a text file that has the following entries on the MPDN server.

[API2]
Player=2.39.0...2.39.3
Extensions=1.14.0, 1.14.1

[API3]
Player=2.40.0
Extensions=1.15.0

[API4]
Player=2.41.0
Extensions=1.16.0

You could also use this file to check for the latest version for both the player and extensions as it should be a relatively small download, or have a server side script to extract just the latest versions from this file. The added benefit of having this file would be that when I upload a new version, I don't have to make sure I do all at once - i.e. I can upload the player, then do an extensions release, and user won't be notified of a new update until I actually update this text file. This way we guarantee nothing slips in between the time I release a new player and a new extensions version.

Users who use standalone editions could also simply lookup this file and they'll know which is the latest extensions for the player version they're currently using.

[Dlget]

i updated to 355.60.
I noticed crash when using MPC-BE.So i started testing MPDN
But same issue with MPDN.

Whenever i try fullscreen,player crashes.

[videonerd]

Does the screen corruption go beyond MPDN window? If it does, it's almost certainly an issue outside of MPDN's control. Could be the driver or hardware.

Yes it does extend beyond the window. May I know what gpu / hardware initialization happens at startup for MPDN?

[Belphemur]

Great job Belphemur! I too love the new updater.

BTW, MPDN now has an Extensions API version (4 currently) - I think we could use this now to match the extensions to the player. Any thoughts on how we could do this? I was thinking a text file that has the following entries on the MPDN server.

[API2]
Player=2.39.0...2.39.3
Extensions=1.14.0, 1.14.1

[API3]
Player=2.40.0
Extensions=1.15.0

[API4]
Player=2.41.0
Extensions=1.16.0

You could also use this file to check for the latest version for both the player and extensions as it should be a relatively small download, or have a server side script to extract just the latest versions from this file. The added benefit of having this file would be that when I upload a new version, I don't have to make sure I do all at once - i.e. I can upload the player, then do an extensions release, and user won't be notified of a new update until I actually update this text file. This way we guarantee nothing slips in between the time I release a new player and a new extensions version.

Users who use standalone editions could also simply lookup this file and they'll know which is the latest extensions for the player version they're currently using.

I have a better idea:

I can get the minimum player version required for the Extension from the changelog.

But I would need the player to advertise its compatibility with the Extensions version, this could be done very easily, modify the changelog.txt (and would make sense also):

v2.41.0 Changelog:
Works with Extension version: 1.16.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


In case of multiple use a COMMA to separate them.

v2.41.0 Changelog:
Works with Extension version: 1.16.0, 1.16.1, 1.16.2
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


Easy to parse for me, only one change for you and still the same file to change at each new version. Ho and there is no need to update retroactively the compatibility for the MPDN Changelog.

[Zachs]

I have a better idea:

I can get the minimum player version required for the Extension from the changelog.


In the future when the extensions API no longer changes so frequently, we may end up with changelogs for every version of MPDN that shows a long list of compatible MPDN extensions version...

It also doesn't cover the fact that we could release several versions of extensions without a new MPDN release.

[Zachs]

Yes it does extend beyond the window. May I know what gpu / hardware initialization happens at startup for MPDN?

Initialise Direct3D and OpenCL (if it's available).

[Zachs]

I've managed to improve startup time yet again. This time on my system, startup time (subsequent ones - first startup is always erratic for every app) has been reduced to ~250ms *with* the full MPDN extensions.

[Belphemur]

In the future when the extensions API no longer changes so frequently, we may end up with changelogs for every version of MPDN that shows a long list of compatible MPDN extensions version...

It also doesn't cover the fact that we could release several versions of extensions without a new MPDN release.

Using the API.txt file doesn't change that. At every new version of MPDN, you'll still need to update the file and to add the new version of the extensions.

We could do this differently and using the idea of Range with an hyphen:

v2.41.0 Changelog:
Works with Extension version: 1.16.0-1.17.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API

Only one number to change.

In case of several release of the extensions, there is no problem at all, since each extension release does tell for which player it's meant to work with. (The Updater will simply parse the "MPDN v2.41.0 and above." part of the body of the release, and since you put it in bold, it's even easier to be sure to parse the right one).

MindMap UseCase:
http://i.imgur.com/CKJtRkF.png

[Zachs]

Let's take the following sequence of events as a hypothetical use-case.


v2.41.0 Changelog:
Works with Extension version: 1.16.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


Release MPDN Extensions v1.17.0 (I understand it is not necessary to update the changelog, but more on that later)


v2.41.0 Changelog:
Works with Extension version: 1.16.0-1.17.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


Release MPDN v2.41.1 .2 and .3


v2.41.3 Changelog:
Works with Extension version: 1.16.0-1.17.0
Fix bug 2

v2.41.2 Changelog:
Works with Extension version: 1.16.0-1.17.0
Fix bug 1

v2.41.1 Changelog:
Works with Extension version: 1.16.0-1.17.0
Improved startup speed

v2.41.0 Changelog:
Works with Extension version: 1.16.0-1.17.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


Release MPDN Extensions 1.17.1


v2.41.3 Changelog:
Works with Extension version: 1.16.0-1.17.1
Fix bug 2

v2.41.2 Changelog:
Works with Extension version: 1.16.0-1.17.1
Fix bug 1

v2.41.1 Changelog:
Works with Extension version: 1.16.0-1.17.1
Improved startup speed

v2.41.0 Changelog:
Works with Extension version: 1.16.0-1.17.1
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


This gets messy pretty quickly. If I don't update retroactively, then it's confusing for users who look at the changelog to try and figure out which version of extensions their existing player is compatible with (e.g. they want to test an old version with its latest compatible extensions).

[Belphemur]

This gets messy pretty quickly. If I don't update retroactively, then it's confusing for users who look at the changelog to try and figure out which version of extensions their existing player is compatible with (e.g. they want to test an old version with its latest compatible extensions).

I see your point, then we just put the minimum compatible version of the extension, like you're doing for the Extension with the Player Version.


v2.42.0 Changelog:
Works with Extension from version: 1.17.0
Change of API

v2.41.3 Changelog:
Works with Extension from version: 1.16.0
Fix bug 2

v2.41.2 Changelog:
Works with Extension from version: 1.16.0
Fix bug 1

v2.41.1 Changelog:
Works with Extension from version: 1.16.0
Improved startup speed

v2.41.0 Changelog:
Works with Extension from version: 1.16.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API


This way the user are not getting confused, the version is clear and if they want to test an old version, they know from which version of Extension they can start (and check when that min version changed).

[Dlget]

i updated to 355.60.
I noticed crash when using MPC-BE.So i started testing MPDN
But same issue with MPDN.

Whenever i try fullscreen,player crashes.


I found solution.
Nvidia 3D vision is making this problem.
Disabled 3D vision & everything works fine

[madshi]

Yes, I've seen NVidia 3D vision turned on causing problems before.

[diimaan]

When I try to put the example scripts(AudioScripts) in the Extension folder, it shows a warning and the filters are not getting loaded.

I am running the 2.41.1

[Anima123]

I am not sure it's the right place to report.

When playing back with MPDN for some time, the newly installed windows 10 will stuck into a BSOD indicating VIDEO_MEMORY_MANAGEMNET_INTERNAL error, at least with fullscreen exclusive mode. It may happened when quiting the player or switching back to windows mode.

Read from another forum about this issue, someone said a game with fullscreen causes the same BSOD, I suspect that fullscreen exclusive mode would trigger it. I didn't have time to do tests to confirm that though.

[Zachs]

When I try to put the example scripts(AudioScripts) in the Extension folder, it shows a warning and the filters are not getting loaded.

I am running the 2.41.1

Make sure you're using the source code version of the extension, not the binary or the installer.

[Zachs]

I am not sure it's the right place to report.

When playing back with MPDN for some time, the newly installed windows 10 will stuck into a BSOD indicating VIDEO_MEMORY_MANAGEMNET_INTERNAL error, at least with fullscreen exclusive mode. It may happened when quiting the player or switching back to windows mode.

Read from another forum about this issue, someone said a game with fullscreen causes the same BSOD, I suspect that fullscreen exclusive mode would trigger it. I didn't have time to do tests to confirm that though.

You might want to report that in the NVIDIA driver feedback forum. I haven't encountered this on my NVS4200M though.

[edison]

Is there any way to reset all settings.

[Zachs]

Yes. Just delete the %localappsettings%\MediaPlayerDotNet folder.

[Zachs]

v2.42.0 Changelog:
Works with Extension from version: 1.17.0
Change of API

v2.41.3 Changelog:
Works with Extension from version: 1.16.0
Fix bug 2

v2.41.2 Changelog:
Works with Extension from version: 1.16.0
Fix bug 1

v2.41.1 Changelog:
Works with Extension from version: 1.16.0
Improved startup speed

v2.41.0 Changelog:
Works with Extension from version: 1.16.0
MPDN Extensions API version 4
Fixed OpenCL compatibility issue with newer Nvidia drivers
Added option to force the use of the new OpenCL interop API



This still seems to have a lot of duplicate lines of "Works with Extension from version: 1.16.0".

Would it be possible to simply use "MPDN Extensions API version 4" instead? And in the MPDN Extensions release notes, I'll add that line too. And each time the API version changes, MPDN changelog still also have that line.

[diimaan]

Make sure you're using the source code version of the extension, not the binary or the installer.

I used the Gain.cs and Silencer.cs!

[Zachs]

I used the Gain.cs and Silencer.cs!
I meant you can't mix the .cs files with the .DLL files. You have to use the entire MPDN extensions in its source form.

[Belphemur]

This still seems to have a lot of duplicate lines of "Works with Extension from version: 1.16.0".

Would it be possible to simply use "MPDN Extensions API version 4" instead? And in the MPDN Extensions release notes, I'll add that line too. And each time the API version changes, MPDN changelog still also have that line.

That line will also get duplicated everywhere into the Changelog.txt, it doesn't change that. But I agree it's a cleaner way to advertise the compatibility.

New use cases using API version:
http://i.imgur.com/IpokSJV.png

[burfadel]

I notice the version of Mediainfo in the native\x64 folder is pretty old, version 0.7.72 whereas the latest is 0.7.76. I replaced the one that comes with MPDN with the newer one, it works fine.

[diimaan]

I meant you can't mix the .cs files with the .DLL files. You have to use the entire MPDN extensions in its source form.

I have deleted all files in Extensions folder and %programdata% too. Extracted the scripts from MPDN_Extensions-master.zip file. Am I doing something wrong?

MPDN still shows extensions are incompatible. I tried a clean install as well.

[ryrynz]

Am I doing something wrong?


Delete %localappdata%\MediaPlayerDotNet that should sort it.

[Zachs]

I notice the version of Mediainfo in the native\x64 folder is pretty old, version 0.7.72 whereas the latest is 0.7.76. I replaced the one that comes with MPDN with the newer one, it works fine.

I'll get the DLLs updated for the next release.

I have deleted all files in Extensions folder and %programdata% too. Extracted the scripts from MPDN_Extensions-master.zip file. Am I doing something wrong?

MPDN still shows extensions are incompatible. I tried a clean install as well.

What version of MPDN are you running? As stated on the GitHub page, prerequisites: MPDN v2.41.0 and above.

[Zachs]

That line will also get duplicated everywhere into the Changelog.txt, it doesn't change that. But I agree it's a cleaner way to advertise the compatibility.

New use cases using API version:
http://i.imgur.com/IpokSJV.png

Ah I meant just the one line only when the API level changes. The added benefit of doing so is that the existing changelog is already written in such a way starting from API level 3.

[Belphemur]

Ah I meant just the one line only when the API level changes. The added benefit of doing so is that the existing changelog is already written in such a way starting from API level 3.

Sounds good, I'll make the updater assume the player use the same version as the current extension when the string is not detected in the changelog. We can do the same for the extension in fact.

I'll work on it in the following week.

String to be used:
In the changelog of the player --> "MPDN Extensions API version 4"
In the changelog of the extension --> "**API Version 4**"

Good for you ?

[diimaan]

I'll get the DLLs updated for the next release.



What version of MPDN are you running? As stated on the GitHub page, prerequisites: MPDN v2.41.0 and above.

My mistake I deleted from appdata folder only not programdata!

Running 2.41.1! Is there a binary for example scripts? I want to try them.

I liked the Tomsteady 32 bit floating point arithmetic normalizer with gain from PotPlayer, which actually is a winamp plugin Tomsteady ported into PotPlayer. It gives reasonable boost without distortion on DTS/AC3 tracks.

Can we do something similar with the mpdn?

[Zachs]

Sounds good, I'll make the updater assume the player use the same version as the current extension when the string is not detected in the changelog.

I'll work on it in the following week.

Thanks Belphemur!

BTW, I've reworded the changelog for MPDN and release notes for MPDN extensions to use the phrase "Extensions API Level" instead of extensions version. I'll stick with that exact phrase whenever I release a new update.

EDIT: LOL I just saw your edits.

[Zachs]

My mistake I deleted from appdata folder only not programdata!

Running 2.41.1! Is there a binary for example scripts? I want to try them.

I liked the Tomsteady 32 bit floating point arithmetic normalizer with gain from PotPlayer, which actually is a winamp plugin Tomsteady ported into PotPlayer. It gives reasonable boost without distortion on DTS/AC3 tracks.

Can we do something similar with the mpdn?

I'm sure it's possible now with Audio Scripts. Just need someone to port it to MPDN. If you know any audio expert who'd like to be MPDN's Audio Script dev, feel free to let me know! It could use some Audio Scripts...

[Zachs]

Some of my thoughts on Audio Scripts:

I'd like to be able to get MPDN to use Winamp's DSP plugins without any modifications - does anyone have any experience with Winamp DSP plugins?
Is this possible at all? And legally, is it allowed?

[nevcairiel]

I don't know if there are any official places left to download the Winamp SDK now that Winmap is dead, but there was an SDK to build DSP plugins, so you can equally just use that to use DSP plugins.
Laws cannot really prevent that.

[Zachs]

Ah good to know. I remember that too, not just for DSP plugins. For all sorts of plugins. Winamp is my inspiration for MPDN actually.
I'll try to find if I could get hold of that old SDK...

[Magik Mark]

Zachs is it possible to put controls for contrast, brightness, hue & saturation?

[foozoor]

Why not include the certificate inside the installer?

Why not add some performance presets?
I think a lot of noobs make dummy things with rendering scripts.

Where can I find the mpdn logo svg/ai file or highres png?
This is to create some wallpapers. ;)

[aufkrawall]

Some people have some weird configuration with their TV expecting interlaced input.
Would it be difficult to interlace output via render script?

[burfadel]

I'll get the DLLs updated for the next release.

Thanks for that :).

[Belphemur]

Why not include the certificate inside the installer?

Why not add some performance presets?
I think a lot of noobs make dummy things with rendering scripts.

Where can I find the mpdn logo svg/ai file or highres png?
This is to create some wallpapers. ;)

For the first question I can answer it.
Because in some system you won't be able to even run the installer if the certificate is not in the Trusted CA of the system.

Which means, the certificate need to be provided separately.

[elraffa]

hi.. I would like to know how to increase the brightness in MPDN.. or if there is any pixel shader for it... thanks

Enviado desde mi GT-S5360 usando Tapatalk 2

[aufkrawall]

Why has still no Radeon user tested if new interop works for them?

[Zachs]

Zachs is it possible to put controls for contrast, brightness, hue & saturation?

hi.. I would like to know how to increase the brightness in MPDN.. or if there is any pixel shader for it... thanks

Enviado desde mi GT-S5360 usando Tapatalk 2

Not something I'd put in MPDN itself, but ImageProcessor script can be used for such a thing (I suspect 3DLUT is a little over the top for you?). You just need to find the right hlsl file from the net. It's not something I use personally and I feel those controls are best adjusted on your display, not the source.

Why not add some performance presets?
I think a lot of noobs make dummy things with rendering scripts.

Where can I find the mpdn logo svg/ai file or highres png?
This is to create some wallpapers. ;)

Performance presets are simply .rs files (loaded via Scripted Render Chain) - you'd have one for each performance preset. Why haven't we got that yet? Because I have a long list of bugs to fix and things to implement and improve and I just haven't got time to do it. If anyone wants to contribute, please feel free to do so.

Ah there isn't any vector/highres MPDN logo file.

[x7007]

Hi,

Where do I copy the dtsdecoderdll.dll file ? is there a newer version better than 1.1.0.9 ? the file name suppose to be dtsdecoderdll.dll or dtsdecoder.dll

[huhn]

Hi,

Where do I copy the dtsdecoderdll.dll file ? is there a newer version better than 1.1.0.9 ? the file name suppose to be dtsdecoderdll.dll or dtsdecoder.dll

you don't need it anymore lavfilter plays DTS HD now for some time.

Is there a huge difference for you in GPU load, clocks and render times with prefer scalar over OpenCL?
Why has still no Radeon user tested if new interop works for them?
was out of internet I'm sorry.

my AMD card is currently not in a PCIe 3.0 x16 slot have to test this later.

i will test both things soon.

edit1:
MPDN 2.41.0.3220 (alternate weight access method test build)
prescale
nnedi3 256/256 480p -> 1080p 62% GPU ~30ms
openCL
nnedi3 256/256 480p -> 1080p 100% GPU ~42ms

prescale is worlds faster.

edit2:
MPDN 2.41.2.3353 with extension from the first post
i guess this version has the openCL fix. my GTX 760 died so i don't have a way to test this fix to make sure.
openCL
nnedi3 256/256 480p -> 1080p 100% GPU ~49ms
so it is slower than the old test version but is works. i haven't checked it visually. not a huge things for me the other version is way faster anyway.

[huhn]

Some people have some weird configuration with their TV expecting interlaced input.
Would it be difficult to interlace output via render script?

the driver should interlaced the progressive input frame.
and to be totally honest i don't things the picture will be any good if the TV needs interlaced as in input.

[aufkrawall]

Thanks for testing, huhn. "Always use new OpenCL Intertop API" option was checked?
If yes, AMD driver still seems to do copyback?

[huhn]

i can't find that option so i guess i still have the old version will try the newest version from github later. but this version is clearly slower strange...

[aufkrawall]

The option can be found in MPDN itself under video renderer -> general.
It should be there with normal final build.

Edit: Alternate weight access method for non-OpenCL is now also in final extensions.

[Zachs]

@huhn
Thanks for carrying out the NNEDI3 tests.

So the SM5.0 version is also worlds faster on AMD then. From memory, I thought the prefer vector method actually works best on AMD? Anyway this then really leaves Maxwell as the odd one out. My suspicion is NVIDIA simply haven't fully optimised the drivers for their new architecture yet.


The old test version and the latest version are exactly the same if you don't enable the new interop as far as OpenCL is concerned. I'm really at a loss to explain why the test build is so much faster.

[aufkrawall]

Maxwell is especially faster than Kepler in games that require excessive shader power.
It'd be strange if NNEDI3 SM couldn't be optimized for it in some way.

[Zachs]

Yeah my gut feeling tells me they just haven't optimised it enough yet. There's no reason for OpenCL to be faster than shader as far as my understanding goes. One is kind of a 'hard coded' version in how the hardware is used while the other gets dynamically mapped to the hardware resource by the driver. In fact it still puzzles me that NVIDIA prefers OpenCL for 128 and 256 neurons when both Intel and AMD hates it all across the board. The speedup from OpenCL to shader is almost exactly the same for both of them too.

[Zachs]

@huhn

Does Lav filter support DTS HD for both x86 and x64?
I should update the OP then...

[nevcairiel]

Does Lav filter support DTS HD for both x86 and x64?
I should update the OP then...

It does, without any external DLL.

[huhn]

@huhn
Thanks for carrying out the NNEDI3 tests.

So the SM5.0 version is also worlds faster on AMD then. From memory, I thought the prefer vector method actually works best on AMD? Anyway this then really leaves Maxwell as the odd one out. My suspicion is NVIDIA simply haven't fully optimised the drivers for their new architecture yet.

vector is way slower than pre scaler on my r9 270 is is nearly as slow as openCL in the old test build.

The old test version and the latest version are exactly the same if you don't enable the new interop as far as OpenCL is concerned. I'm really at a loss to explain why the test build is so much faster.

i have both version "installed" (the old test version is just in a folder and not installed properly) currently the main difference is that the old faster test build is anyCPU and the new test is done with the installer for x64 and the installer for extensions.

i just retested it the test build is clearly faster all other settings should be at default.

here are the results with the new interop api:

the new api uses 40 ms with the same test file. so it is faster and that quite a bit but of cause still no match to pre scaler.

and don't forget i test SD to FHD here the impact from the copyback problem with openCL should be way worse with higher resolutions so openCL doesn't stand a chance against SM 5.0

[ryrynz]

@huhn

Does Lav filter support DTS HD for both x86 and x64?
I should update the OP then...
Yup told you about that 😊

[Zachs]

Ah I thought it was WMA tracks.

[x7007]

If I have 970 GTX , should I select Always use new OpenCL Intertop API ? and OpenCL NNEDI3 extension ? ( I chose 64 neurons )

I'm using the old drivers in windows 10 the 353.49 . What will happen when I upgrade to 353.63 for example or newer . Should I still select those settings ? I have crazy good performance with this now, only 4.0-6.0 ms when checking CTRL + J.

[Zachs]

Does the OSD screen say "fallback has occurred in renderscript"? If it does, it means NNEDI3 isn't running.

The "Always use new OpenCL interop API" won't have any effects on drivers without the legacy interop support - i.e. it defaults to the new interop.

[Zachs]

vector is way slower than pre scaler on my r9 270 is is nearly as slow as openCL in the old test build.



i have both version "installed" (the old test version is just in a folder and not installed properly) currently the main difference is that the old faster test build is anyCPU and the new test is done with the installer for x64 and the installer for extensions.

i just retested it the test build is clearly faster all other settings should be at default.

here are the results with the new interop api:

the new api uses 40 ms with the same test file. so it is faster and that quite a bit but of cause still no match to pre scaler.

and don't forget i test SD to FHD here the impact from the copyback problem with openCL should be way worse with higher resolutions so openCL doesn't stand a chance against SM 5.0

Can you test with the same AnyCPU version to rule out any difference in settings?

If the new interop API is faster, it would suggest that it doesn't need copyback (i.e. latency is gone)! Not that it really matters any more, since AMD also prefers the shader version.

[huhn]

i took the Application.64.config made a copy and renamed it to Application.anyCPU.config and overwriten the old Application.anyCPU.config with it. they are now equally in speed.

in term of speed
SM5.0 pre scaler 32 ms
prefer vector 38 ms
SM 5.0 avoid branches 46 ms and dropping frames
openCL new API 40 ms
openCL normal 48+ ms frames drop like crazy

interesting is this. small code has no meaningful difference for me but prefer scaler & small code and alternate weights access method together are really slow 46ms+. while prefer vector & small code with and without alternate weights access method has about the same speed ~36 ms.

about the interop not sure SM 5.0 can be really slow as slow as slow as normal openCL so hard to judge. i may set my PCIe to 1.0 and test again. SM 5.0 shouldn't really care but openCL should lose a ton of speed if a copyback is used.

edit: +-2 ms for the same test is pretty normal the rendertimes are not really stable.

and a small news about the audio drop issue.
i have my r9 270 in my gaming PC now and just got the audio issue. i can't remember any issue on this PC before. so it can be assumed it is the card it self or the AMD driver.
i may do some better test or may not. moving PC hardware from one PC to the next is quite time consuming.

[Zachs]

I think it is safe to say the new interop gets rid of AMD copyback!

The 8ms difference is massive and nothing else would explain that difference. With NVIDIA where no copyback is involved either for new or old interop, the difference is around 0.1ms (in favour of the new interop).

Since the new interop is faster all around and doesn't harm Intel (Intel doesn't even like OpenCL NNEDI3), I'll make it use that by default in MPDN 2.42.0.

p.s. Thank you so much for testing. Your efforts are very much appreciated.

[x7007]

Is that good ?

Which version to use x86 , x64 or anycpu ? what does it mean any cpu ?

So if I update to nvidia newest drivers without legacy Interop I won't have the OpenCL and NNEDI3 support ?
http://s8.postimg.org/446k2pfz9/3333.jpg

[Zachs]

Is that good ?

So if I update to nvidia newest drivers without legacy Interop I won't have the OpenCL and NNEDI3 support ?

If you update to the latest drivers, you *will* still have OpenCL NNEDI3 support - that's the point of the new interop!

What I was trying to say was even if you don't check the new interop option, it will *still* use the new interop when the old interop isn't available. The option "forces" the use of the new interop path, but it doesn't mean the new path won't be used automatically when it has no other choice.

The warning you see below the new interop option is mainly only for old drivers that may not have the new interop path - all recent drivers have it.

The other thing you might want to do from your screen shot is revert "Max Quality" back to "Image Quality". The difference between max quality is so minute your eyes won't (at least shouldn't) be able to see it.

EDIT: And the reason you're getting such low render time is because you're not upscaling at all - you're actually downscaling.

[Zachs]

Shiandow has just committed a new feature that allows you to copy and paste your script chain. In going forward, you could now share your script chain with other users on the forum.

*Note: The latest commits require the unreleased MPDN v2.42.0 so don't try it yet.

e.g. SuperChromaRes --> Deband --> NNEDI3 256/256 --> Sinc-Blackman 2D

<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="SuperChromaRes" Guid="48fc4ac2-225d-4249-a3b3-c5a80362e268">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>2</Passes>
<Strength>0.5</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</Preset>
<Preset Name="Deband" Guid="8791dff6-9588-429e-aff1-7ed8ea136879">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>8</maxbitdepth>
<power>0.5</power>
<grain>True</grain>
</Settings>
</Script>
</Preset>
<Preset Name="NNEDI3" Guid="ff8ade8c-6487-4f3c-8c55-491b691e63a1">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.NNedi3Scaler">
<Settings>
<Neurons1>Neurons256</Neurons1>
<Neurons2>Neurons256</Neurons2>
<CodePath>ScalarMad</CodePath>
<Structured>False</Structured>
</Settings>
</Script>
</Preset>
<Preset Name="EWA Scaler" Guid="753b685e-de61-49c1-94fa-d2d462c39356">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerScaler">
<Settings>
<TapCount>Eight</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincBlackman" />
</Settings>
</Script>
</Preset>
</ListOfPreset>


Copy-and-pasting the code above into a blank script chain will allow you to duplicate my example script chain exactly.

[huhn]

I think it is safe to say the new interop gets rid of AMD copyback!

The 8ms difference is massive and nothing else would explain that difference. With NVIDIA where no copyback is involved either for new or old interop, the difference is around 0.1ms (in favour of the new interop).

Since the new interop is faster all around and doesn't harm Intel (Intel doesn't even like OpenCL NNEDI3), I'll make it use that by default in MPDN 2.42.0.

p.s. Thank you so much for testing. Your efforts are very much appreciated.

the test with PCIe 1.0 will show it for sure.

i use 256 neuron so the 8 ms just for better interop is very very huge the calculation shouldn't change what so ever and 256 is very slow. so yes it is likely not using an copy back operation.

what extension is used this time? i don't remember a extension all GPU share.

[Zachs]

No no. There isn't a common one - it's the same as the old interop where they have their own extensions. When I compared NVIDIA to AMD's difference in new and old interop speed, I didn't mean they both share the same API.

Anyway it'll still be good to find out if the copyback is truly eliminated - even if no AMD user will use the OpenCL version...

[huhn]

PCIe 1.0:
SM 5.0 prefer scaler 36 ms
openCL 53 ms
openCL new API 40 ms

numbers are very very unstable. but openCL new api is as fast as always so it can be assumed it isn't using a copyback.
pretty interesting that SM 5.0 got slower.

what so ever on my card prefer scaler is boss.

[Zachs]

Fantastic!

It's fully confirmed then - the new OpenCL interop doesn't have the copyback problem!

Thanks again huhn.

[x7007]

If you update to the latest drivers, you *will* still have OpenCL NNEDI3 support - that's the point of the new interop!

What I was trying to say was even if you don't check the new interop option, it will *still* use the new interop when the old interop isn't available. The option "forces" the use of the new interop path, but it doesn't mean the new path won't be used automatically when it has no other choice.

The warning you see below the new interop option is mainly only for old drivers that may not have the new interop path - all recent drivers have it.

The other thing you might want to do from your screen shot is revert "Max Quality" back to "Image Quality". The difference between max quality is so minute your eyes won't (at least shouldn't) be able to see it.

EDIT: And the reason you're getting such low render time is because you're not upscaling at all - you're actually downscaling.

So I don't understand, how can I upscale ? I should see Upscaling NNEDI3 right on the CTRL + J . why it's not active ?


My settings now are

Renderer Script - OpenCL NNEDI3 64-64
Luma Upscaler - Jinc 4 Taps , unselected Anti-Ringing
Chroma Upscaler - Same ^
Luma Downscaler - Same ^
Chroma Downscaler - SoftCubic Softness 75


Should I select Rely on DWM to handle the Vsync in windowed mode ?

[huhn]

this renderer doesn't show the used scaler in the OSD.

most default setting are default because they are the best so yes leave them as they are.

to use nnedi3 you need an file that needs upscaling a 1080p file on a 1080p screen isn't upscaled. chroma is a different story.

[ryrynz]

this renderer doesn't show the used scaler in the OSD.


Yeah, I've hit Zach up about this once before and I've wished it there many times since, been meaning to mention it again.. It's a pretty useful bit of info seen in madVR's OSD.

[Zachs]

Well the problem is MPDN doesn't know what scaler is actually being used in the render script.

[huhn]

and to be honest is that something really needed in a OSD?

[Zachs]

BTW, this could be done via player extensions. So please go ahead and add a new request on GitHub.

[aufkrawall]

Fantastic!

It's fully confirmed then - the new OpenCL interop doesn't have the copyback problem!

Thanks again huhn.
Kudos for your coder skills. :devil:

[burfadel]

What is the order of the way the filters are applied? Do the Luma and Chroma upscalers/downscalers work first, then the render script, or does the render script work first, then the upscalers/downscalers?

I would be thinking SuperRes etc would be better applied before upscaling or downscaling, but then I have a custom luma greyscale random noise shader that adds a bit of warmth to the picture (as in making it look perceptively better). This shader would be best applied last.

[Zachs]

It depends on what renderscript you're using - you could entirely or partially replace MPDN's scalers, or use MPDN's scalers for first phase chroma only as well as the final scaling to target size etc.

In your case, if you want to use SuperRes with a noise shader at the end, you'd have something like "SuperRes (NNEDI3) --> Resizer (100% target size) --> noise shader".

[x7007]

So how do I know if I use NNEDI3 and it works and I get the same quality from Madvr with potplayer ? If I have low Ms , that means it doesn't work or it has better performance than the normal OpenCL api ? I'm confused. I want to use this player but I can't be sure about some of the settings compare to the normal Madvr.


Which version to use x86 , x64 or anycpu ? what does it mean any cpu ?
Should I select Rely on DWM to handle the Vsync in windowed mode ?

[aufkrawall]

So how do I know if I use NNEDI3 and it works and I get the same quality from Madvr with potplayer ? If I have low Ms , that means it doesn't work or it has better performance than the normal OpenCL api ?

When an upscaling render script is selected without conditionals, it will always be used if upscaling is needed.
Even if target rectangle is just one pixel more than native resolution, it is used. Then you should also see render time increasing, in case of NNEDI3 by a lot.
If you want to use render scripts in a smarter way, you have to use conditional render scripts (see Wiki for MPDN extensions at Github).


I'm confused. I want to use this player but I can't be sure about some of the settings compare to the normal Madvr.

For chroma scaling, I suggest Sinc-Blackman AR with 6 or 8 taps.
Spline 4 taps is very similar to madVR Catmull-Rom (or is even equal), according to Zachs. So you might want to use it for downscaling.


what does it mean any cpu ?

I don't know.


Should I select Rely on DWM to handle the Vsync in windowed mode ?
Option is unchecked by default and this way, playback is totally reliable here. I suppose it shouldn't be enabled, unless there are problems.

[x7007]

When an upscaling render script is selected without conditionals, it will always be used if upscaling is needed.
Even if target rectangle is just one pixel more than native resolution, it is used. Then you should also see render time increasing, in case of NNEDI3 by a lot.
If you want to use render scripts in a smarter way, you have to use conditional render scripts (see Wiki for MPDN extensions at Github).


For chroma scaling, I suggest Sinc-Blackman AR with 6 or 8 taps.
Spline 4 taps is very similar to madVR Catmull-Rom (or is even equal), according to Zachs. So you might want to use it for downscaling.


I don't know.


Option is unchecked by default and this way, playback is totally reliable here. I suppose it shouldn't be enabled, unless there are problems.


Which movie will be upscaled ? 720p only on 1080p monitor resolution ? or 1080p on 2560x1440 monitor resolution ? a 1080p movie won't upscale on 1920x1080p monitor ?

When using Potplayer + Madvr + NNEDI3 it always upscaled my 1080p and 720p. Something I am missing here, I just don't why OpenCL NNEDI3 doesn't work because of the low ms.

[huhn]

nnedi3 is an image doubling scaler so no it is not used on a 1080p source on a 1080p screen.

are you sure you know what you want to achieve?

madVR can use nnedi3 for chroma but this is so placebo for movies...

and never forget if you don't see a difference you are just wasting power nothing else.

[ryrynz]

So I don't understand, how can I upscale ? I should see Upscaling NNEDI3 right on the CTRL + J . why it's not active ?

Your source is 1920 x 1080 and your display resolution is smaller because you're not running in fullscreen.
Waste of power using NNEDI3 on full HD content on a full HD screen or lower.


Should I select Rely on DWM to handle the Vsync in windowed mode ?

Nah, leave that one alone.

[Zachs]

I don't know if there are any official places left to download the Winamp SDK now that Winmap is dead, but there was an SDK to build DSP plugins, so you can equally just use that to use DSP plugins.
Laws cannot really prevent that.

Ah alas Winamp plugins are 32-bit only.

[x7007]

Your source is 1920 x 1080 and your display resolution is smaller because you're not running in fullscreen.
Waste of power using NNEDI3 on full HD content on a full HD screen or lower.



Nah, leave that one alone.

I watch Exclusive full screen, I just took the screenshot to show the settings.

So all you guys are saying NNEDI3 is not useful for 1080p even on Exclusive Full Screen mode , but only 720p ?

and if I have 2560x1440 resolution then upscaling will work because it will upscale a 1080p movie to the higher resolution .

Is that correct ?

What I want to understand hopefully is the MPDN versions.

What is the difference between x86 or x64 vs AnyCpu version ?

What should I use for the Chroma and Luma Upscaling and Downscaling settings ? I couldn't find such suggestions in the thread. do you guys have any suggestions ?

[Zachs]

Yes that is correct.

As for the question in regards to the different editions,
AnyCPU version will run on x86 and x64 OSes in its native mode (i.e. 32-bit process on 32-bit OS, 64-bit process on 64-bit OS).
x86 will run on both x86 and x64 Oses too but will run in WOW64 mode under x64 (as 32-bit process).
x64 can only run on x64 OSes.

[x7007]

Yes that is correct.

As for the question in regards to the different editions,
AnyCPU version will run on x86 and x64 OSes in its native mode (i.e. 32-bit process on 32-bit OS, 64-bit process on 64-bit OS).
x86 will run on both x86 and x64 Oses too but will run in WOW64 mode under x64 (as 32-bit process).
x64 can only run on x64 OSes.

So if I can't use NNEDI3 even in Exclusive full screen 1080p bluray movies MKV with 1080p resolution monitor.

Which Chroma and Luma Upscaling should I use ?

[ryrynz]

So if I can't use NNEDI3 even in Exclusive full screen 1080p bluray movies MKV with 1080p resolution monitor.


You can, it'll just upscale then downscale to your display resolution and you won't gain anything significant picture quality wise from it by doing so.


Which Chroma and Luma Upscaling should I use ?
Entirely up to you, depends on your source, display settings and personal preference of course. My recommendation would be to try out the EWA scalers for both chroma and luma.

[Zachs]

So if I can't use NNEDI3 even in Exclusive full screen 1080p bluray movies MKV with 1080p resolution monitor.

An honest question - do you know what NNEDI3 does?

[Dlget]

:p
Upscaling is not even used if you watch 1080p video in 1080p display.
so why bother about setting??? those settings are useful only for lower resolution video.

[ryrynz]

:p
Upscaling is not even used if you watch 1080p video in 1080p display.

Don't know why I thought it'd upscale regardless.. but NEDI has the option.
Hey Zach, what is the required upscale for MPDN to use NNEDI3? Doesn't appear to be 2x as I'm having it activate on a 1280 x 694 file > 1920 x 1200.

[Zachs]

It's always upscaled 2x. Then downscaled implicitly to target size using MPDN luma scaler.

[ryrynz]

I'll put that another way, how much of an upscale is required for NNEDI3 to activate?
Being a 2x upscale and without any options to specify otherwise you'd either assume it always activates or only activate when 2x upscaling is required to reach target size.

madVR has the options 'only if scaling factor is 2.0x/1.5x/1.2x or larger' or 'always if upscaling is needed'.

[aufkrawall]

As I already said, it doubles resolution if target rectangle is just one pixel more than source resolution (so always doubled if any upscaling is needed).
If you want to simulate madVR options, you have to use conditional render scripts.

e.g. choose script group, add script chain and in it add conditionals like:
If SourceSize.Width <= 1280 and SourceSize.Height <= 720 and NeedsUpscaling Use NNEDI3
You can also simulate madVR's sharpen after each doubling option with sharpen shaders (can be added into image processor render script). Instead of just NNEDI3 in the example above, you choose render script and in it you first run NNEDI3 doubling and afterwards image processor.
http://abload.de/img/1leaab.png (http://abload.de/image.php?img=1leaab.png)

http://abload.de/img/2swskg.png (http://abload.de/image.php?img=2swskg.png)

[ryrynz]

As I already said, it doubles resolution if target rectangle is just one pixel more than source resolution (so always doubled if any upscaling is needed).
If you want to simulate madVR options, you have to use conditional render scripts.


Missed that. Yeah figured, haven't looked into them much. Cheers.

[huhn]

it is at always but that's why this exist https://github.com/zachsaw/MPDN_Extensions/wiki/The-Conditional-Render-Script
but not sure if this can stop the quadrupling...

a lot of settings can't be used effectively without render script and/or conditional and it is relative complicated to use all this stuff.

for example getting nnedi3 only scaling when 2x or more scaling is needed is quite some work and needs so extra knowledge.

the Conditional system is extrem powerful but not that easy to use.

[aufkrawall]

I agree that some conditional options should be ported into the UI, probably not much work for the devs.
But afterall, conditionals are basically simple Excel when -> then functions. Not rocket science. :)

[elraffa]

Hi...If I going to play 1080p movies in 1080p monitor, what settings should I use to sharpen image quality? My pc specs are: I5 - gtx770 - 12gb ram - Windows 10 64bits

[aufkrawall]

You can use sharpen shaders (e.g. AdaptiveSharpen, FineSharp or Lumasharpen) via image processor render script.
Chroma scaling will hardly have much influence with filmed content, try Sinc-Blackman 8 taps AR.

[elraffa]

You can use sharpen shaders (e.g. AdaptiveSharpen, FineSharp or Lumasharpen) via image processor render script.
Chroma scaling will hardly have much influence with filmed content, try Sinc-Blackman 8 taps AR.

Thanks for reply.. Script chain like the image, it´s ok?
http://i.imgur.com/x0v2Axp.jpg

[Eyldebrandt]

Hey.

The only issue I currently have is annoying.
I suspect this is related to dotnet.

On many mkv files (maybe 70%), which all are BD Remux, MPDN gives me the followed error message when i try to play them :

Object reference not set to an instance of an object

If my translation is correct.
For the information, all the files have been realised with the exact same process and the same software.
And some are plays, and others are not.

And some days ago, they were played easily, and today no longer.

[Eyldebrandt]

Thanks for reply.. Script chain like the image, it´s ok?


I think AS on hlsl claim two scripts, pass one and pass two.
The hlsl file of pass two is the one where you can adjust the strength of the AS.

[Shiandow]

Hey.

The only issue I currently have is annoying.
I suspect this is related to dotnet.

On many mkv files (maybe 70%), which all are BD Remux, MPDN gives me the followed error message when i try to play them :

Object reference not set to an instance of an object

If my translation is correct.
For the information, all the files have been realised with the exact same process and the same software.
And some are plays, and others are not.

And some days ago, they were played easily, and today no longer.

That doesn't sound like dotnet, but rather a bug somewhere in the c# code (either in the extensions or in MPDN itself). If you go to General-Debugging you can enable a more advanced error dialog which contains more information. Especially the Stack Trace would be helpful.

[Eyldebrandt]

Here is what I have :

===================================

An unexpected error 'System.NullReferenceException' has occurred.

------------------------------
Error Type = System.NullReferenceException
Error Message = La référence d'objet n'est pas définie à une instance d'un objet.
Error Source = MediaPlayerDotNet
Error Site = Void OpenMedia(System.String, Boolean, Boolean, Boolean)
Error occurred = à MediaPlayerDotNet.MainForm.OpenMedia(String , Boolean , Boolean , Boolean )
à MediaPlayerDotNet.MainForm.f.A()
à a.R.A()
OS Version = Microsoft Windows NT 6.3.9600.0
Total Physical Memory = 31,9 GB
Available Physical Memory = 26,6 GB
Date Time = 27/08/2015 01:25:10
User Name = TITAN\chris
Application Startup Path = C:\HTPC\MPDN
Application Executable Path = C:\HTPC\MPDN\MediaPlayerDotNet.exe
CurrentDirectory = E:\Blu-ray\Child 44 (Daniel Espinosa - 2015) - BDRX - AVC - DTS-HD Master Audio 5.1 - Vost
SystemDirectory = C:\WINDOWS\system32
Ran As Admin = False
HasShutdownStarted = False
MachineName = TITAN
ProcessorCount = 8
LogicalDrives = C:\, D:\, E:\, F:\, G:\, H:\, I:\, J:\, K:\, L:\
EnvironmentVariables = COMPUTERNAME = TITAN; USERPROFILE = C:\Users\chris; HOMEPATH = \Users\chris; LOCALAPPDATA = C:\Users\chris\AppData\Local; JD2_HOME = C:\Users\chris\AppData\Local\JDownloader v2.0; PSModulePath = C:\WINDOWS\system32\WindowsPowerShell\v1.0\Modules\; PROCESSOR_ARCHITECTURE = AMD64; Path = C:\Program Files (x86)\Intel\iCLS Client\;C:\Program Files\Intel\iCLS Client\;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\WINDOWS\System32\WindowsPowerShell\v1.0\;C:\Program Files\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files\Intel\Intel(R) Management Engine Components\IPT;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\IPT;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common; CommonProgramFiles(x86) = C:\Program Files (x86)\Common Files; ProgramFiles(x86) = C:\Program Files (x86); PROCESSOR_LEVEL = 6; LOGONSERVER = \\MicrosoftAccount; PATHEXT = .COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH;.MSC; HOMEDRIVE = C:; SystemRoot = C:\WINDOWS; SESSIONNAME = Console; ALLUSERSPROFILE = C:\ProgramData; PUBLIC = C:\Users\Public; FPS_BROWSER_APP_PROFILE_STRING = Internet Explorer; APPDATA = C:\Users\chris\AppData\Roaming; PROCESSOR_REVISION = 3a09; USERNAME = chris; CommonProgramW6432 = C:\Program Files\Common Files; TEMP = C:\Users\chris\AppData\Local\Temp; CommonProgramFiles = C:\Program Files\Common Files; OS = Windows_NT; USERDOMAIN_ROAMINGPROFILE = TITAN; PROCESSOR_IDENTIFIER = Intel64 Family 6 Model 58 Stepping 9, GenuineIntel; ComSpec = C:\WINDOWS\system32\cmd.exe; SystemDrive = C:; FPS_BROWSER_USER_PROFILE_STRING = Default; ProgramFiles = C:\Program Files; NUMBER_OF_PROCESSORS = 8; TMP = C:\Users\chris\AppData\Local\Temp; ProgramData = C:\ProgramData; ProgramW6432 = C:\Program Files; windir = C:\WINDOWS; USERDOMAIN = TITAN
RuntimeVersion = 4.0.30319.42000
Thread Count = 34
Handle Count = 848
VM Size = 105,3 GB
Peak VM Size = 105,4 GB
Working Set Size = 162 MB
Max Working Set Size = 1,3 MB
Min Working Set Size = 200 KB
Modules = MediaPlayerDotNet.exe, ntdll.dll, MSCOREE.DLL, KERNEL32.dll, KERNELBASE.dll, avcuf64.dll, ADVAPI32.dll, msvcrt.dll, sechost.dll, RPCRT4.dll, mscoreei.dll, SHLWAPI.dll, combase.dll, GDI32.dll, USER32.dll, IMM32.DLL, MSCTF.dll, kernel.appcore.dll, VERSION.dll, clr.dll, MSVCR120_CLR0400.dll, mscorlib.ni.dll, ole32.dll, bcryptPrimitives.dll, uxtheme.dll, ADvdDiscHlp64.dll, System.ni.dll, CRYPTSP.dll, bcrypt.dll, rsaenh.dll, CRYPTBASE.dll, System.Core.ni.dll, MediaPlayerDotNet.ni.exe, System.Drawing.ni.dll, System.Windows.Forms.ni.dll, Mpdn.Core.ni.dll, comctl32.dll, dwmapi.dll, System.Configuration.ni.dll, System.Xml.ni.dll, gdiplus.dll, DWrite.dll, WindowsCodecs.dll, CommandLine.ni.dll, clrjit.dll, OLEAUT32.dll, shell32.dll, windows.storage.dll, shcore.dll, powrprof.dll, profapi.dll, YAXLib.ni.dll, System.Xml.Linq.ni.dll, clbcatq.dll, dataexchange.dll, d2d1.dll, d3d11.dll, dcomp.dll, dxgi.dll, twinapi.appcore.dll, USERENV.dll, MediaInfoDotNet.ni.dll, VideoFrameServicesNative.dll, MediaInfo.dll, SharpDX.Direct3D9.ni.dll, SharpDX.ni.dll, d3d9.dll, nvd3dumx.dll, nvspcap64.dll, SETUPAPI.dll, CFGMGR32.dll, PSAPI.DLL, WINHTTP.dll, nvapi64.dll, SspiCli.dll, SharpDX.DXGI.ni.dll, D3D9NativeServices.dll, AVRT.dll, dxva2.dll, WINMM.dll, WTSAPI32.dll, WINMMBASE.dll, DEVOBJ.dll, opencl.dll, nvSCPAPI64.dll, WINSTA.dll, WINTRUST.dll, MSASN1.dll, CRYPT32.dll, SharpDX.Direct3D11.ni.dll, nvwgf2umx.dll, nvopencl.dll, WS2_32.dll, NSI.dll, Dx11Font.dll, FW1FontWrapper.dll, CSScriptLibrary.ni.dll, DirectShowLib-2005.ni.dll, WindowsBase.ni.dll, PresentationCore.ni.dll, explorerframe.dll, rasapi32.dll, rasman.dll, rtutils.dll, mswsock.dll, ondemandconnroutehelper.dll, IPHLPAPI.DLL, WINNSI.DLL, dhcpcsvc6.DLL, quartz.dll, dhcpcsvc.DLL, LAVSplitter.ax, avformat-lav-56.dll, avutil-lav-54.dll, avcodec-lav-56.dll, libbluray.dll, avresample-lav-2.dll, DNSAPI.dll, rasadhlp.dll, fwpuclnt.dll, qcap.dll, sxs.dll, XySubFilter.dll, COMDLG32.dll, WINSPOOL.DRV, WININET.dll, LAVVideo.ax, swscale-lav-3.dll, avfilter-lav-5.dll, nvcuda.dll, secur32.dll, schannel.DLL, mskeyprotect.dll, ncrypt.dll, NTASN1.dll, ncryptsslp.dll, gpapi.dll, System.ComponentModel.DataAnnotations.ni.dll, Microsoft.ExceptionMessageBox.ni.dll, Microsoft.VisualBasic.ni.dll, System.Runtime.Serialization.ni.dll, diasymreader.dll

[elraffa]

I think AS on hlsl claim two scripts, pass one and pass two.
The hlsl file of pass two is the one where you can adjust the strength of the AS.

Yes, that´s right. Also SuperRes like the image, increse a bit the sharpen.
http://i.imgur.com/cWds9Lz.jpg

[Zachs]

Here it is :

TITLE: MediaPlayerDotNet Error
------------------------------

An unexpected error 'System.NullReferenceException' has occurred.

------------------------------
ADDITIONAL INFORMATION:

La référence d'objet n'est pas définie à une instance d'un objet. (MediaPlayerDotNet)

------------------------------
BUTTONS:

&Ignore
&Abort
------------------------------


Could you get the stack trace please? There's a little button on the lower left of the error dialog box that will show the detail and in there you'll find the stack trace. Make sure Debug Dialog is enabled.

EDIT: Ah just saw your edits. Thanks, I'll have a look.

[Zachs]

Hey.

The only issue I currently have is annoying.
I suspect this is related to dotnet.

On many mkv files (maybe 70%), which all are BD Remux, MPDN gives me the followed error message when i try to play them :

Object reference not set to an instance of an object

If my translation is correct.
For the information, all the files have been realised with the exact same process and the same software.
And some are plays, and others are not.

And some days ago, they were played easily, and today no longer.

Very strange. I had a look at the stack trace but I can't seem to figure out what's causing it. So a few questions.

1) Can you try reinstalling MPDN to see if it fixes it?
2) Are you running MPDN extensions? From the debug info I think it's not running but if it is, could you test MPDN without its extensions?
3) Can you try clearing the app settings to see if it fixes the problem? You can delete the whole %localappdata%\MediaPlayerDotNet folder. But make sure you do a backup of it first in case it's not related to that problem.

[Eyldebrandt]

Very strange. I had a look at the stack trace but I can't seem to figure out what's causing it. So a few questions.

1) Can you try reinstalling MPDN to see if it fixes it?
2) Are you running MPDN extensions? From the debug info I think it's not running but if it is, could you test MPDN without its extensions?
3) Can you try clearing the app settings to see if it fixes the problem? You can delete the whole %localappdata%\MediaPlayerDotNet folder. But make sure you do a backup of it first in case it's not related to that problem.


1) Done twice, change nothing, unfortunately
2)maybe it's a lead, because extensions are running, and works well with the .mkv I can play.
Anyway, issue's occured even without the extensions.
3)will try this right now.
Edit : change nothing.

Thx for the support.

[Zachs]

OK can you try running the AnyCPU edition (download and unzip the file) and see if it happens there as well?
And do you happen to have a short sample of your MKV handy so I could try replicate the problem?

EDIT: BTW does the issue only occur when you try to open those specific MKV files? Or does it happen randomly?

[Eyldebrandt]

OK can you try running the AnyCPU edition (download and unzip the file) and see if it happens there as well?
And do you happen to have a short sample of your MKV handy so I could try replicate the problem?

EDIT: BTW does the issue only occur when you try to open those specific MKV files? Or does it happen randomly?

Just downloaded to try.
Well, that concern hundreds of +25GB files, well, they're untouched Blu-rays.
And yes and no for the random type. Well, at the moment, that occur on the sames files, avg 70% of all of them. But yesterday, some files that I cannot play are played now, and others I was able to play are no longer played right now.

[Eyldebrandt]

Same result with anycpu edition, on the exact sames files.

[Zachs]

Hmm. OK I'll need to isolate this problem further - Can you try with this old MPDN version 2.39.3 (http://mpdn.zachsaw.com/Old%20Releases/2.39.3/)? Just grab the AnyCPU edition and run it without extensions.

[Eyldebrandt]

With this build, I still have the issue, but report is different.
Here it is

===================================

An unexpected error 'Mpdn.VideoPlayer.DsGraphException' has occurred.

------------------------------
Error Type = Mpdn.VideoPlayer.DsGraphException
Error Message = Failed to render file 'E:\Blu-ray\Child 44 (Daniel Espinosa - 2015) - BDRX - AVC - DTS-HD Master Audio 5.1 - Vost\Child 44 (Daniel Espinosa - 2015) - BDRX - AVC - DTS-HD Master Audio 5.1 - Vost.mkv'

Details:
Video graph failed to render

Hint:
Make sure video filters are installed and configured properly
Error Source = MediaPlayerDotNet
Error Site = Void .ctor(System.String, D)
Error occurred = à d.H..ctor(String , D )
à B.H.A(String , B )
à MediaPlayerDotNet.MainForm.c(String )
à MediaPlayerDotNet.MainForm.A(String , Boolean , Boolean , Boolean )
à MediaPlayerDotNet.MainForm.OpenMedia(String , Boolean , Boolean , Boolean )
à MediaPlayerDotNet.MainForm.P(Object , EventArgs )
à System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
à System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
à System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
à System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
à System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
à System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
à System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
à System.Windows.Forms.Control.WndProc(Message& m)
à System.Windows.Forms.ToolStrip.WndProc(Message& m)
à System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
à System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)
OS Version = Microsoft Windows NT 6.3.9600.0
Total Physical Memory = 31,9 GB
Available Physical Memory = 26,6 GB
Date Time = 27/08/2015 02:24:08
User Name = TITAN\chris
Application Startup Path = G:\Apps\MPDN\MediaPlayerDotNet_AnyCPU_2_39_3_3251
Application Executable Path = G:\Apps\MPDN\MediaPlayerDotNet_AnyCPU_2_39_3_3251\MediaPlayerDotNet.exe
CurrentDirectory = G:\Apps\MPDN\MediaPlayerDotNet_AnyCPU_2_39_3_3251
SystemDirectory = C:\WINDOWS\system32
Ran As Admin = True
HasShutdownStarted = False
MachineName = TITAN
ProcessorCount = 8
LogicalDrives = C:\, D:\, E:\, F:\, G:\, H:\, I:\, J:\, K:\
EnvironmentVariables = COMPUTERNAME = TITAN; USERPROFILE = C:\Users\chris; HOMEPATH = \Users\chris; LOCALAPPDATA = C:\Users\chris\AppData\Local; JD2_HOME = C:\Users\chris\AppData\Local\JDownloader v2.0; PSModulePath = C:\WINDOWS\system32\WindowsPowerShell\v1.0\Modules\; PROCESSOR_ARCHITECTURE = AMD64; Path = C:\Program Files (x86)\Intel\iCLS Client\;C:\Program Files\Intel\iCLS Client\;C:\WINDOWS\system32;C:\WINDOWS;C:\WINDOWS\System32\Wbem;C:\WINDOWS\System32\WindowsPowerShell\v1.0\;C:\Program Files\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files\Intel\Intel(R) Management Engine Components\IPT;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\DAL;C:\Program Files (x86)\Intel\Intel(R) Management Engine Components\IPT;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common; CommonProgramFiles(x86) = C:\Program Files (x86)\Common Files; ProgramFiles(x86) = C:\Program Files (x86); PROCESSOR_LEVEL = 6; LOGONSERVER = \\MicrosoftAccount; PATHEXT = .COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH;.MSC; HOMEDRIVE = C:; SystemRoot = C:\WINDOWS; ALLUSERSPROFILE = C:\ProgramData; PUBLIC = C:\Users\Public; APPDATA = C:\Users\chris\AppData\Roaming; PROCESSOR_REVISION = 3a09; USERNAME = chris; CommonProgramW6432 = C:\Program Files\Common Files; CommonProgramFiles = C:\Program Files\Common Files; OS = Windows_NT; USERDOMAIN_ROAMINGPROFILE = TITAN; PROCESSOR_IDENTIFIER = Intel64 Family 6 Model 58 Stepping 9, GenuineIntel; ComSpec = C:\WINDOWS\system32\cmd.exe; SystemDrive = C:; TEMP = C:\Users\chris\AppData\Local\Temp; ProgramFiles = C:\Program Files; NUMBER_OF_PROCESSORS = 8; TMP = C:\Users\chris\AppData\Local\Temp; ProgramData = C:\ProgramData; ProgramW6432 = C:\Program Files; windir = C:\WINDOWS; USERDOMAIN = TITAN; __COMPAT_LAYER = DetectorsWin8 Installer
RuntimeVersion = 4.0.30319.42000
Thread Count = 18
Handle Count = 555
VM Size = 53 GB
Peak VM Size = 53 GB
Working Set Size = 124,2 MB
Max Working Set Size = 1,3 MB
Min Working Set Size = 200 KB
Modules = MediaPlayerDotNet.exe, ntdll.dll, MSCOREE.DLL, KERNEL32.dll, KERNELBASE.dll, avcuf64.dll, apphelp.dll, ADVAPI32.dll, msvcrt.dll, sechost.dll, RPCRT4.dll, mscoreei.dll, SHLWAPI.dll, combase.dll, GDI32.dll, USER32.dll, IMM32.DLL, MSCTF.dll, kernel.appcore.dll, VERSION.dll, clr.dll, MSVCR120_CLR0400.dll, mscorlib.ni.dll, ole32.dll, bcryptPrimitives.dll, uxtheme.dll, clrjit.dll, OLEAUT32.dll, System.ni.dll, System.Core.ni.dll, System.Drawing.ni.dll, System.Windows.Forms.ni.dll, cryptsp.dll, bcrypt.dll, rsaenh.dll, CRYPTBASE.dll, DirectShowLib-2005.ni.dll, comctl32.dll, dwmapi.dll, System.Configuration.ni.dll, System.Xml.ni.dll, gdiplus.dll, DWrite.dll, WindowsCodecs.dll, CommandLine.ni.dll, shell32.dll, windows.storage.dll, shcore.dll, powrprof.dll, profapi.dll, System.Xml.Linq.ni.dll, clbcatq.dll, dataexchange.dll, d2d1.dll, d3d11.dll, dcomp.dll, dxgi.dll, twinapi.appcore.dll, USERENV.dll, SharpDX.ni.dll, VideoFrameServicesNative.dll, MediaInfo.dll, SharpDX.Direct3D9.ni.dll, d3d9.dll, nvd3dumx.dll, nvspcap64.dll, SETUPAPI.dll, CFGMGR32.dll, PSAPI.DLL, WINHTTP.dll, nvapi64.dll, SspiCli.dll, SharpDX.DXGI.ni.dll, SharpDX.Direct3D11.ni.dll, D3D9NativeServices.dll, AVRT.dll, dxva2.dll, WINMM.dll, WTSAPI32.dll, WINMMBASE.dll, DEVOBJ.dll, opencl.dll, nvSCPAPI64.dll, WINSTA.dll, WINTRUST.dll, MSASN1.dll, CRYPT32.dll, nvwgf2umx.dll, nvopencl.dll, WS2_32.dll, NSI.dll, d3dx9_43.dll, usp10.dll, tiptsf.dll, quartz.dll, LAVSplitter.ax, avformat-lav-56.dll, avutil-lav-54.dll, avcodec-lav-56.dll, libbluray.dll, avresample-lav-2.dll, qcap.dll, sxs.dll, XySubFilter.dll, COMDLG32.dll, WINSPOOL.DRV, WININET.dll, LAVVideo.ax, swscale-lav-3.dll, avfilter-lav-5.dll, devenum.dll, ntmarta.dll, msdmo.dll, MMDevApi.dll, PROPSYS.dll, wdmaud.drv, ksuser.dll, AUDIOSES.DLL, wintypes.dll, msacm32.drv, MSACM32.dll, midimap.dll, dsound.dll, Microsoft.ExceptionMessageBox.ni.dll, Microsoft.VisualBasic.ni.dll, System.Management.ni.dll, diasymreader.dll


===================================

Failed to render file 'E:\Blu-ray\Child 44 (Daniel Espinosa - 2015) - BDRX - AVC - DTS-HD Master Audio 5.1 - Vost\Child 44 (Daniel Espinosa - 2015) - BDRX - AVC - DTS-HD Master Audio 5.1 - Vost.mkv'

Details:
Video graph failed to render

Hint:
Make sure video filters are installed and configured properly (MediaPlayerDotNet)

------------------------------
Program Location:

à d.H..ctor(String , D )
à B.H.A(String , B )
à MediaPlayerDotNet.MainForm.c(String )
à MediaPlayerDotNet.MainForm.A(String , Boolean , Boolean , Boolean )
à MediaPlayerDotNet.MainForm.OpenMedia(String , Boolean , Boolean , Boolean )
à MediaPlayerDotNet.MainForm.P(Object , EventArgs )
à System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
à System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
à System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
à System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
à System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
à System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
à System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
à System.Windows.Forms.Control.WndProc(Message& m)
à System.Windows.Forms.ToolStrip.WndProc(Message& m)
à System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
à System.Windows.Forms.NativeWindow.Callback(IntPtr hWnd, Int32 msg, IntPtr wparam, IntPtr lparam)

===================================

Élément introuvable. (Exception de HRESULT : 0x80070490) (MediaPlayerDotNet)

------------------------------
Program Location:

à d.H.A(String , a )
à d.H..ctor(String , D )


Maybe it's an issue relative to the PC itself, nuh ?

[Zachs]

No no this is a different issue - it simply means you haven't got LAV filters installed. Did you uninstall your x64 edition before running the AnyCPU edition? If you did, you'll need to reinstall the x64 edition before running the AnyCPU edition again. Uninstalling MPDN causes LAV filters to get uninstalled too.

[Eyldebrandt]

Nope, he's still installed.
Really strange issue.

[Zachs]

Nope, he's still installed.
Really strange issue.

Just to confirm, can you extract all of the files to a new empty folder?

[Eyldebrandt]

Well, I've installed the portable version on an other HDD, but yes, I can and I have.

[Zachs]

Can you play any files at all with this old AnyCPU version?

[Eyldebrandt]

Can play well the same ones I can play with classic install + extensions, and cannot plays the same ones.

This is what I really can't understand.
All the files have been encoded with the exact same software, on the sames conditions.
I figured it out if I can't play no one, but there, it's a mistery.

And I'm pretty sure than tomorrow or in a couple of days, files I can't play will works, and some I can play right now will no longer. XD

[Zachs]

Well with the old version it's complaining that LAV can't open the files for some reason.
Perhaps if you try using something like graph edit (I use GraphStudioNext) and manually use the chain LAV Source Splitter --> LAV Video Decoder --> Video Renderer to check if LAV filter is working at all.

[Eyldebrandt]

LAV is working.
I uninstall the MPDN integrated LAV version and install the classic LAV x64, for science.

Result is identiqual.
Files played by the classic MPDN are played by the old one, and those who aren't, aren't with the old one.

So, Maybe it's an issue from LAV with MPDN.
Funny thing is if I try to play the files "corrupted" with MPC + madVR + LAV, they're played normally.

I understand nothing :D

[x7007]

An honest question - do you know what NNEDI3 does?

I just know when using potplayer and watching 1080p/720p Remux or Bluray MKV quality the image was really noticeable compare to JINC and others I tried. I just see the different with my TV. Philips 7007, it has IPS panel as far as I know and the quality is better than 50% of the normal pc monitors. the only issue it has is no support for 0-255 RGB only 16-235 Limited. but I notice and would like to keep using NNEDI3. My friend to compare do too saw a different between what we used JINC or the Lazarvos or something, so it can't just be my eyes. he has Dell 22 or something monitor the one with the good panel. All I want is to know MPDN is using NNEDI3 so I can start watching movies with it, but it doesn't look like it works, I always get 6.0ms render time. I am using 353.49 is that's matter for the problem.


:p
Upscaling is not even used if you watch 1080p video in 1080p display.
so why bother about setting??? those settings are useful only for lower resolution video.

I just notice the different , my friend too.

[Eyldebrandt]

I found the problem -_-
That was.... XySubfilter -_-
The Fuck ? -_-

When i disable him, everything works fine.

Edit : the exact problem was : when I set Xysubfilters on "never load", MPDN fails to play large amount of files.

Well, that is not an answer for the issue on the old MPDN.

[Zachs]

I just know when using potplayer and watching 1080p/720p Remux or Bluray MKV quality the image was really noticeable compare to JINC and others I tried. I just see the different with my TV. Philips 7007, it has IPS panel as far as I know and the quality is better than 50% of the normal pc monitors. the only issue it has is no support for 0-255 RGB only 16-235 Limited. but I notice and would like to keep using NNEDI3. My friend to compare do too saw a different between what we used JINC or the Lazarvos or something, so it can't just be my eyes. he has Dell 22 or something monitor the one with the good panel. All I want is to know MPDN is using NNEDI3 so I can start watching movies with it, but it doesn't look like it works, I always get 6.0ms render time. I am using 353.49 is that's matter for the problem.

FWIW madVR *does not* use NNEDI3 when rendering 1080p on 1080p screen for luma as well. Are you talking about chroma upscaling with NNEDI3?

[Zachs]

I found the problem -_-
That was.... XySubfilter -_-
The Fuck ? -_-

When i disable him, everything works fine.

Is that the same problem for the latest MPDN version too?

EDIT: Ah found the problem. It would appear by selecting "Do not load" subtitles in XySubFilter, it doesn't just skip the loading part - it simply doesn't want to connect to anything in the graph! So it's not really a "Do not load" subtitles option - it's actually a "Do not use" XySubFilter option! I'll have to make a workaround to ignore the error when that happens.

[Eyldebrandt]

Yep.
Everything works fine now.

At least, you do not wasting your time too much, thx a lot for your kind and comprehensive support.

[Zachs]

No thank *you* for helping me debug this!
If it affects you, chances are it'll affect others too.

[elraffa]

Hi Zachs..I´ve posted earlier and want to know if my settings are fine for a sharper image quality. Is there any other settings for it?

[Zachs]

If you're viewing 1080p on 1080p screen, SuperRes won't have any effects. If you're upscaling, you may want to turn down the SuperRes strength (if you're running 3 pass, you should use a much lower strength e.g. 0.5) and leave super-xBR settings default (1.0 and 1.0) with Use third pass enabled.

But otherwise how sharp you want it really comes down to personal choice.

[elraffa]

If you're viewing 1080p on 1080p screen, SuperRes won't have any effects. If you're upscaling, you may want to turn down the SuperRes strength (if you're running 3 pass, you should use a much lower strength e.g. 0.5) and leave super-xBR settings default (1.0 and 1.0) with Use third pass enabled.

But otherwise how sharp you want it really comes down to personal choice.
Thanks Zachs..

[Zachs]

Just a quick note about MPDN AudioScripts.

AudioScripts require that you have OpenCL runtimes installed for your CPU as it can currently only use the Intel iGPU or the CPU. If you want to use audio scripts, you need to make sure you have these runtimes installed on your system.

OpenCL CPU runtimes,
Intel:
https://software.intel.com/en-us/articles/opencl-drivers
AMD:
http://developer.amd.com/tools-and-sdks/opencl-zone/amd-accelerated-parallel-processing-app-sdk/

Note that you can use Intel OpenCL runtimes on AMD CPUs too but performance isn't great whereas using AMD runtimes on Intel CPUs can give greater performance than Intel's own runtimes.

[x7007]

FWIW madVR *does not* use NNEDI3 when rendering 1080p on 1080p screen for luma as well. Are you talking about chroma upscaling with NNEDI3?


I think so

[Zachs]

I think so

MPDN doesn't have a NNEDI3 chroma scaler yet but it's really simple to add - just that no one had the time to get to it yet. And like a lot of other experts have said, NNEDI3 on chroma is just a waste of power. Chroma by nature has less high contrast edges than Luma (see this picture (https://upload.wikimedia.org/wikipedia/commons/2/29/Barn-yuv.png)), so an edge detection algo like NNEDI3 isn't going to be much good. In fact, if you wanted a proper NNEDI3 chroma scaler, you'll need to retrain the neural networks to work on chroma only. And this has been discussed before, tritical (who created NNEDI3) has not released his NN training software.

A better alternative would be to use SuperChromaRes, which our resident expert / developer Shiandow has mentioned that it is much more suited to chroma scaling than NNEDI3. Sinc-Blackman EWA 8-tap with AR also works very well as a chroma scaler as it is very sharp.

NNEDI3 isn't the be all and end all scaler that you might think it is.

[Zachs]

For those of you asking for something similar to TomSteady's normalizer with gain, I've made a similar script. It's called Dynamic Range Compressor.

The parameters in the config dialog is almost similar to this (http://manual.audacityteam.org/o/man/compressor.html) so you should be able to figure out most of it.

This is currently only available on GitHub and can be used with MPDN test build 3375 (http://mpdn.zachsaw.com/Test%20Builds/3375/).

Cheers.

[lanzorg]

Hey Eyldebrandt!

- Are you going to create a MPDN tutorial like you did with Avisynth?
- Could you share your current config?
- What about to add you expertise to create presets directly inside MPDN?

Hey MPDN developers!

- Are you going to create something like Kodi DSPlayer?
- Does MPDN work with youtube-dl like with VLC or mpv?
- Does MPDN work with hybrid gpu? The previous releases didn't recognize the powerful one.

[Zachs]

- Are you going to create something like Kodi DSPlayer?
- Does MPDN work with youtube-dl like with VLC or mpv?

Not sure what Kodi DSPlayer is. So I think the answer to that one is a "no".

I've not used youtube-dl before but I suspect it downloads youtube videos into a file? If so then as long as LAV Filter can open it, MPDN will play it.

[lanzorg]

Not sure what Kodi DSPlayer is. So I think the answer to that one is a "no".

I've not used youtube-dl before but I suspect it downloads youtube videos into a file? If so then as long as LAV Filter can open it, MPDN will play it.

Kodi DSPlayer is a Kodi internal player that works with madVR as video renderer.
Everything is integrated in Kodi just like the default player.
http://forum.kodi.tv/showthread.php?tid=223175

I use youtube-dl for streaming not downloading, it support a tons of websites.
With VLC and mpv, it works very well but with MPC-HC that uses LAV Filters it doesn't.

[aufkrawall]

A better alternative would be to use SuperChromaRes, which our resident expert / developer Shiandow has mentioned that it is much more suited to chroma scaling than NNEDI3.
I have to disagree, SuperChromaRes can brighten up colors.
And with NNEDI3 for chroma, you wouldn't have to run an AR filter.
Chroma can also have hard structures, at least with artificial content (which is not fully unrealistic to appear "in the wild", like pixel art).
I'd really appreciate it if NNEDI3 for chroma would get some higher priority. :)

[huhn]

you know it needs a correction scale thanks to mpeg2 chroma position which makes nnedi3 for chroma even more questionable.

[elraffa]

how can I stretch the image to remove black bands?
http://i.imgur.com/ZG0vulG.jpg

[huhn]

do you really want to change the AR of the image?

[elraffa]

do you really want to change the AR of the image?

I thought that enabling AR 4x3 option would eliminate the black bands, but still the same..

[aufkrawall]

you know it needs a correction scale thanks to mpeg2 chroma position which makes nnedi3 for chroma even more questionable.
Some people prefer it and with MPDN becoming more popular, there will always be people asking.
Especially if there is no "MPDN options explained" thread and if good chroma scalers like Sinc-Blackman are quite hidden in the UI.
It's not a nonsense feature.

[Anima123]

Some people prefer it and with MPDN becoming more popular, there will always be people asking.
Especially if there is no "MPDN options explained" thread and if good chroma scalers like Sinc-Blackman are quite hidden in the UI.
It's not a nonsense feature.

Are you talking about the Sinc-Blackman configure in the MPDN's chroma upscaling options, or the Sinc-Blackman in the script chain options as EWA scaler?

[Zachs]

I thought that enabling AR 4x3 option would eliminate the black bands, but still the same..

Set "Video Letter Boxing" to "No letter boxing (Clip video)"

I have to disagree, SuperChromaRes can brighten up colors.
And with NNEDI3 for chroma, you wouldn't have to run an AR filter.
Chroma can also have hard structures, at least with artificial content (which is not fully unrealistic to appear "in the wild", like pixel art).
I'd really appreciate it if NNEDI3 for chroma would get some higher priority. :)

It's got a high-ish priority. But there are other even higher priority things that needed to be done first.

Are you talking about the Sinc-Blackman configure in the MPDN's chroma upscaling options, or the Sinc-Blackman in the script chain options as EWA scaler?

Sinc-Blackman EWA is the sharpest of all the scalers so far, so it's a good chroma scaler. There's an outstanding bug currently that makes the EWA chroma scaler fallback to MPDN's internal chroma scaler if your Luma doesn't need upscaling (i.e. the bug is it's checking for luma size when it's meant to check for chroma size).

[elraffa]

Set "Video Letter Boxing" to "No letter boxing (Clip video)"
Not working .. black bands are still.. In potplayer I can manually stretch the image with numpad 8 key. Is there anything similar in MPDN?

[huhn]

than they are hard coded

[Zachs]

Yeah the "black band" is actually part of your video!

[elraffa]

than they are hard coded

Yeah the "black band" is actually part of your video!

:/ Thanks..

[x7007]

Thanks for reply.. Script chain like the image, it´s ok?
http://i.imgur.com/x0v2Axp.jpg



What did you set in Deband and Chroma Upscaler ? does EWA and all the rest that you set there has effect on 1080p and 720p ?

If I choose the scripted, what happens in the MPDN Luma,Chroma,Downscaling,Upscaling options ?

[Braum]

Hi ! :)

First of all thank you a lot Zachs for this awesome media player !

I have a question about Conditional render script and Script Group.

I've made a Script Group with three conditional render scripts for High/Medium/Low encoding quality and the problem is they don't seem to work.

So here's the conditional statements :

SourceSize.Width == 1920 and SourceSize.Height == 1080. use Script Chain
SourceSize.Width < 1900 and SourceSize.Height < 1080 and SourceSize.Width >= 920 and SourceSize.Height >= 720. use Script Chain
SourceSize.Width <= 1280 and SourceSize.Height <= 720. use Script Chain

Exemple :
http://reho.st/self/84573b60b730faf34ddcf166c07af73727c4218b.png

When I play a movie in 3 different encoding quality the conditional render scripts doesn't auto select like it should.
So for the moment I use the Hotkey and it does the trick.

Did I do something wrong ? How do I get this to work ?

:thanks:

[Zachs]

If I choose the scripted, what happens in the MPDN Luma,Chroma,Downscaling,Upscaling options ?

Just a little comment on that screenshot. If you want to sharpen at target size (i.e. post scale), you should add a resizer to make it "resize to 100% of target size" before adding the adaptive sharpen scripts. In the case where 1080p is viewed on 1080p screens, this is not necessary since post scale size is the same as pre scale size.

Now to answer your question. In that screenshot, MPDN's internal scalers are completely bypassed, because the only thing that would've been active anyway would be the chroma upscaler. However, this has been replaced by the EWA chroma scaler.

For lower res video (e.g. 720p) on 1080p screens, the resizer script as mentioned above would still use MPDN's luma upscaling option to scale from 720p to 1080p, unless of course if you choose to use NNEDI3, then it would use MPDN's luma downscaling option to scale from 1440p to 1080p. In those cases, the "luma" scaling options double as "image" scalers.

[Zachs]

Hi ! :)

First of all thank you a lot Zachs for this awesome media player !

I have a question about Conditional render script and Script Group.

I've made a Script Group with three conditional render scripts for High/Medium/Low encoding quality and the problem is they don't seem to work.

So here's the conditional statements :

SourceSize.Width == 1920 and SourceSize.Height == 1080. use Script Chain
SourceSize.Width < 1900 and SourceSize.Height < 1080 and SourceSize.Width >= 920 and SourceSize.Height >= 720. use Script Chain
SourceSize.Width <= 1280 and SourceSize.Height <= 720. use Script Chain

Exemple :
http://reho.st/self/84573b60b730faf34ddcf166c07af73727c4218b.png

When I play a movie in 3 different encoding quality the conditional render scripts doesn't auto select like it should.
So for the moment I use the Hotkey and it does the trick.

Did I do something wrong ? How do I get this to work ?

:thanks:

You need to put the conditionals in a "Script Chain", not script group (a script group is to enable hotkey toggling), and make sure you have the script chain selected as the active render script.

Your last conditional conflicts with your second one, so you'll need to correct that as well. The second one becomes active when width is between 920 and 1900 and height is between 720 and 1080. However, the last one also gets activated when the width is between 920 and 1280 and height is 720.

[Braum]

You need to put the conditionals in a "Script Chain", not script group (a script group is to enable hotkey toggling), and make sure you have the script chain selected as the active render script.

Your last conditional conflicts with your second one, so you'll need to correct that as well. The second one becomes active when width is between 920 and 1900 and height is between 720 and 1080. However, the last one also gets activated when the width is between 920 and 1280 and height is 720.

Ah my bad ! Indeed I've also made an error with my conditional !

Thank you a lot it works like a charm now ! :D

When using a sweetfx shader, do we need to use Linear&Gamma Light converter script ?

Like this :

ConvertToLinearLight
Shader
ConvertToGammaLight

[Zachs]

No there's no need to.

[Braum]

Okay http://reho.st/gif/65c89cd03a632899b5df556da5d65f11ca058dc2.gif

[elraffa]

What did you set in Deband and Chroma Upscaler ? does EWA and all the rest that you set there has effect on 1080p and 720p ?
In deband are the default options. In EWA chroma scaler I set Sinc-blackman - 8 taps - AR enabled - strenght 1.
Yes, all the scripts render that I set has effect in 1080p and 720p.

[diimaan]

With the test version MediaPlayerDotNet_AnyCPU_2_42_0_3375

I get the error "Video graph failed to render"

Hint
Make sure video filters are installed and configured properly.

I have LAV Filters 0.65 x64 version installed. Not the MPDN installer. LAV filters from the separate thread.

Also DynamicRangeCompressor extension fails to load in the test version.

[diimaan]

Updated to this version MediaPlayerDotNet_AnyCPU_2_42_0_3380

And still the same results.

But the installer version works fine on playback...

[Magik Mark]

zachs what is the functional difference between script chain & script group? Which scenario is best suited for each ?

[x7007]

In deband are the default options. In EWA chroma upscaler I set Sinc-blackman - 8 taps - AR enabled - strenght 1.
Yes, all the scripts render that I set has effect in 1080p and 720p.
You forgot the ChromaScaler, can you tell me the settings to this too .
If you select Video Renderer Script then the options Chroma,Luma Scaler - Downscaling are disabled and the script has the priority ? In shortcut, in NONE the normal MPDN Scaler,Downscaling works and if renderer selected then they are disabled.


I don't what settings to use then in the Chorma,Luma Upscaling and the Downscaling , EWA , Debanding.


couple of questions.

How come x86 doesn't manage to work CUVID ? it freezes when you try to open the movie. and you need to End the software from Task manager.
x64 works fine with CUVID

How can I use or enable Chroma NNEDI3 + AR + Debanding like MadVr ?

Does anyone uses FluidMotion with nvidia card ? it works the same like the Amd control panel option?

For some reason I have 2 Lav Filter Splitter tray icons, and checking the filters in MPDN it shows 2 Lav Filter Source, any idea why this could happen ?

[Zachs]

How come x86 doesn't manage to work CUVID ? it freezes when you try to open the movie. and you need to End the software from Task manager.
x64 works fine with CUVID

How can I use or enable Chroma NNEDI3 + AR + Debanding like MadVr ?

Does anyone uses FluidMotion with nvidia card ? it works the same like the Amd control panel option?

For some reason I have 2 Lav Filter Splitter tray icons, and checking the filters in MPDN it shows 2 Lav Filter Source, any idea why this could happen ?

LAV Filter with CUVID works fine for me on both x86 and x64... anyone else having this problem?

Chroma NNEDI3: read the posts on the previous page.

FluidMotion is not AMD's FluidMotion.

LAV Filter Splitter will show up twice because MPDN uses two separate graphs (use search next time please).

[Zachs]

zachs what is the functional difference between script chain & script group? Which scenario is best suited for each ?

Does no one ever read the thread before asking questions, even just one page back? http://forum.doom9.org/showthread.php?p=1736232#post1736232

[Zachs]

Updated to this version MediaPlayerDotNet_AnyCPU_2_42_0_3380

And still the same results.

But the installer version works fine on playback...

Did you extract all the files in the zip archive?

[elraffa]

You forgot the ChromaScaler, can you tell me the settings to this too .
Those settings are EWA chroma scaler from script render.
If you select Video Renderer Script then the options Chroma,Luma Scaler - Downscaling are disabled and the script has the priority ? In shortcut, in NONE the normal MPDN Scaler,Downscaling works and if renderer selected then they are disabled.
This was replied by Zachs in post #3156..If you use script render for both scalers, MPDN´s internal scaler will not work. The scalers will work depending on resolution of the movie and resolution of the monitor.

[aufkrawall]

Btw: You don't have to define Sinc-Blackman for chroma via render script if you want to use it all the time anyway.
Then it's enough to simply set it in main MPDN's chroma scaler settings (under "custom").

[diimaan]

Did you extract all the files in the zip archive?

I just extracted the entire zip to folder and tried to run from the extracted location. It does not see the LAV Filter install. I installed the x86 and x64, x86 alone, x64 alone all after a fresh install.

[Zachs]

Do you get the same problem with the released AnyCPU edition?

[diimaan]

Do you get the same problem with the released AnyCPU edition?

I am using the AnyCPU release only.

[Braum]

I'm trying to find a way to have a clean audio gain because I can't bring my DAC/AMP where I'll live. So one of the things I've tried is your exemple of Audio Script : Gain.cs ( https://github.com/zachsaw/MPDN_Extensions/blob/600ef8f1b20bdfc6dcf58e2c01fa68edfd47cdb1/Examples/AudioScripts/Gain.cs)

Alas the script failed to load :

http://reho.st/self/b6c22e1f9f3bb3520ab287ce13bea19804d04293.png


Maybe I've done something wrong ?

Is there a way to have an audio gain script ?

I was using a VST/DSP plugin before but it's not compatible with MPDN yet.

[Zachs]

I am using the AnyCPU release only.
So you're saying the AnyCPU release version is working but not the test builds?

[Zachs]

I'm trying to find a way to have a clean audio gain because I can't bring my DAC/AMP where I'll live. So one of the things I've tried is your exemple of Audio Script : Gain.cs ( https://github.com/zachsaw/MPDN_Extensions/blob/600ef8f1b20bdfc6dcf58e2c01fa68edfd47cdb1/Examples/AudioScripts/Gain.cs)

Alas the script failed to load :

http://reho.st/self/b6c22e1f9f3bb3520ab287ce13bea19804d04293.png


Maybe I've done something wrong ?

Is there a way to have an audio gain script ?

I was using a VST/DSP plugin before but it's not compatible with MPDN yet.
The gain example is an example written to show developers how they can write their own audio scripts. It clips anything that is over amplified. Are you sure you know what you want?

[Zachs]

Btw: You don't have to define Sinc-Blackman for chroma via render script if you want to use it all the time anyway.
Then it's enough to simply set it in main MPDN's chroma scaler settings (under "custom").
Actually just to clarify, the EWA version is different to the MPDN internal scaler. They use the same kernel function but they'll give you different results.

[Braum]

The gain example is an example written to show developers how they can write their own audio scripts. It clips anything that is over amplified. Are you sure you know what you want?
Well if it clip then no, my bad.
Actually I think the low audio ouput problem come from the downmix of DTS 7.1 to Stereo.

Do you see a solution ? Or a clue/lead ?

:thanks:

[huhn]

if your output is low think about DRC. AC3 filter can do thsi and it should work totally fine with MPDN.

DRC is bad for audio quality just to make that clear the other way would be a audio system with a proper amp.

[Braum]

if your output is low think about DRC. AC3 filter can do thsi and it should work totally fine with MPDN.

DRC is bad for audio quality

DRC isn't what I'm looking for as it reduce dynamic range (which I want to conserve).
just to make that clear the other way would be a audio system with a proper amp
Indeed, so I'll need to keep my DAC/AMP with me.

Thank you for your answer !

[huhn]

is not possible to boost audio digital without lowering the dynamic range.

different headphones and USB DAC can easily have a higher max volume and there are mobile headphone amps...

[Braum]

is not possible to boost audio digital without lowering the dynamic range.

different headphones and USB DAC can easily have a higher max volume and there are mobile headphone amps...

I got a Burson HA160DS

Measurements:
Input impedance: 36.5 KOhms
Frequency response: ± 1 dB 0 – 20Khz
Signal to noise ratio: 110dB
THD: <0.12% at 150mW , 0.06% at 100mW
Channel separation: <54dB
Output power: 250mW (less than 1% distortion)
Output impedance: phones out 5.6 Ohms
Power dissipation: >20W, internal, regulated power supply

With an orthodynamic (high impedance) headphone (ZMF x Vibro).

With DTS (downmix to stereo) and high impedance headphone I need to put the volume near to the max. The signal is clean but the DAC/AMP is very hot and during summer that's no good ^^

[huhn]

a normal sound card isn't made to feed headphones with more than 32 OHM. and that can't be fixed with digital processing.
these headphones are simply not made to run without AMP or an really good soundcard.

the strange thing is that these are just listed with 50 ohm so they shouldn't get in trouble except for mobile devices. for mobile use a headphone should be at 16-32 ohm or even lower.
on the other hand even my 16 ohm t5p gets into trouble with an source device like an 3ds.

what so ever nothing a software should be used for if audio quality matters.

btw. it would be nice if you could can MPDN with an video file and check the latency with latencymon: http://www.resplendence.com/latencymon

i would really like to know if i'm the only one with issues thanks to my DACs.

[Zachs]

if your output is low think about DRC. AC3 filter can do thsi and it should work totally fine with MPDN.

DRC is bad for audio quality just to make that clear the other way would be a audio system with a proper amp.

Oh btw the next version of MPDN will have a DRC audio script too with pretty extensive settings that you can tune to your liking.

[diimaan]

So you're saying the AnyCPU release version is working but not the test builds?

I am sorry for not being clear. The release edition 64 bit(installer version) runs fine.

But when I try the Latest test version(zip files from here (http://mpdn.zachsaw.com/Test%20Builds/3380/)) of AnyCPU/x86/x64 versions all were not able to find the LAV filters.

While trying test versions, I uninstalled the release version and cleared the folder from %appdata%\local\mpdn.
Then Installed the LAV filters separately.

Still the test versions (zip versions) were not able to locate the LAV filters. I tried installing and uninstalling various versions of LAV filters. (0.65/0.64/0.63) etc.

[Anime Viewer]

I updated to Windows 10 this morning, and I was pleasantly surprised to find that that when I launched a video in MPDN fullscreen I no longer have high present times when using my Optimus Nvidia GTX 680M as the gpu. My present times are now 0.30ms. MPDN coding seems to have something to do with it, for when I tested with MPC-HC and madVR I still get 9ms present times.

[aufkrawall]

I rather suppose the reported low rendering time is far off from any reality.

[huhn]

he is talking about present times.

there where always pretty high present times issue with optimus system.
if they are high enough to cause problems is another story.

[aufkrawall]

Whoops, I wanted to recheck if it was about present times, but forgot it. :)

[Braum]

a normal sound card isn't made to feed headphones with more than 32 OHM. and that can't be fixed with digital processing.
these headphones are simply not made to run without AMP or an really good soundcard.

That's why I got a Burson HA160DS
http://reho.st/self/ecce65b87a58146ab4e7ebc3c16f5413fdbb4908.jpg (http://reho.st/view/self/ecce65b87a58146ab4e7ebc3c16f5413fdbb4908.jpg)

the strange thing is that these are just listed with 50 ohm so they shouldn't get in trouble except for mobile devices. for mobile use a headphone should be at 16-32 ohm or even lower.
on the other hand even my 16 ohm t5p gets into trouble with an source device like an 3ds.

what so ever nothing a software should be used for if audio quality matters..

When I listen to music my DAC/AMP is at low volume but when I Watch movies (mainly DTS audio) I need to put the volume near to the maximum.

btw. it would be nice if you could can MPDN with an video file and check the latency with latencymon: http://www.resplendence.com/latencymon

i would really like to know if i'm the only one with issues thanks to my DACs.

Will do, I'm moving to a new place so I'll check it later this week. ;)

[huhn]

That's why I got a Burson HA160DS
http://reho.st/self/ecce65b87a58146ab4e7ebc3c16f5413fdbb4908.jpg (http://reho.st/view/self/ecce65b87a58146ab4e7ebc3c16f5413fdbb4908.jpg)

and i'm happy that i didn't need my headphone AMP at all.


When I listen to music my DAC/AMP is at low volume but when I Watch movies (mainly DTS audio) I need to put the volume near to the maximum.

that normal music is DRC to death all ready on the disc/dl/streaming while a movies isn't DRC or at least not as hard.

i usually watch movies at 30-45 % while i hear music at 10-20%

Will do, I'm moving to a new place so I'll check it later this week. ;)

thanks

[aufkrawall]

:thanks: for making render script chroma scalers compatible with NNEDI3, it really looks better in cartoons with resolution doubling now.

[Zachs]

No problem. That was always in the plan. And once we sort out the basic chroma scaling framework, NNEDI3 chroma will follow.

[aufkrawall]

I will test it. :cool:
Will both also be possible with OpenCL NNEDI3?

Btw: Maybe it will make sense to completely replace main MPDN scalers by render script ones in future, or at least remove some of them?
I could imagine that new users might gonna panic if they see all those options. :)
Of course MPDN is mainly for advanced users, but there are really quite some buttons now. :D

[Zachs]

Ah well to cater for those non-advanced users, I'll keep the internal scalers there just so they could use it if render script is a little too much for them.

EDIT: Yes the OpenCL and shader version will work the same. Just haven't had the time to bring the chroma scaler across to the OpenCL version.

[ryrynz]

Ah well to cater for those non-advanced users, I'll keep the internal scalers there just so they could use it if render script is a little too much for them.


And for people not using extensions..or would that somehow work?

[Zachs]

And for people not using extensions..or would that somehow work?

Yes it would be needed for those who're not using extensions too.

[Zachs]

Time flies and it's been a year since MPDN was first released. Happy Birthday MPDN!

Just wanted to take the opportunity to thank all the hard work from the devs and all the testers in helping to make MPDN a great media player that it is today. Thank you!

MPDN v2.42.0 has been released with two key features you've been asking for: Custom chroma scaler in NNEDI3 (both shader and OpenCL) and Dynamic Range Compressor AudioScript.

It remains our top priority to bring you the following features:
Chaining of AudioScripts
NNEDI3 Chroma Scaler

Cheers!

[Anima123]

Congratulations all the developers who dedicated your enthusiasm into this great project.

Your attitudes are cautious yet open, solved one issue and another either implicitly or explicitly. That's what we called professional.

[Anima123]

I don't really understand the new NNEDI3 Chroma Scaler, what it do and when to trigger it.

Previously we can use render script chain to invoke EWA Chroma Scaler, what's the difference between that and the new NNEDI3 Chroma Scaler? And why it's still be activated when playing a 1080p video with native target resolution and the rendering time is high?

[Zachs]

EWA Chroma Scaler in the render script scales the initial chroma to luma size. NNEDI3's custom chroma scaler then doubles the chroma alongside the luma which is being doubled with NNEDI3. Previously it is not possible to use any custom chroma scaler alongside NNEDI3 (it used to scale with MPDN's internal scalers only).

[Anima123]

It might be a false alarm.

The only thing that I can't figure out is why playing back a low-resolution file, which formerly takes far less rendering time than now.

[Zachs]

I have to disagree, SuperChromaRes can brighten up colors.
And with NNEDI3 for chroma, you wouldn't have to run an AR filter.
Chroma can also have hard structures, at least with artificial content (which is not fully unrealistic to appear "in the wild", like pixel art).
I'd really appreciate it if NNEDI3 for chroma would get some higher priority. :)

I've just committed NNEDI3 chroma scaler (shader version only for now). If you like you can test it out with the latest source on GitHub.

[Zachs]

It might be a false alarm.

The only thing that I can't figure out is why playing back a low-resolution file, which formerly takes far less rendering time than now.

Are you using the same settings as before?

[Anima123]

Are you using the same settings as before?

Solved. It seems that the new version need a clean installation.

[burfadel]

The first post links to extensions version v1.17.0, however I downloaded and installed both the new MPDN and the extensions, and the update check came up with extensions version v1.17.1. I noticed this a little while ago, I thought I'd give time in case it happened to be between the time of the update and the post being updated :).

[Zachs]

The first post links to extensions version v1.17.0, however I downloaded and installed both the new MPDN and the extensions, and the update check came up with extensions version v1.17.1. I noticed this a little while ago, I thought I'd give time in case it happened to be between the time of the update and the post being updated :).

Ah thanks for that. Totally forgot to update the first post.

[LDD9O]

Comodo reporting that there is a trojan in one of your files. Can you submit it for analysis or write them an email/ticket.

[aufkrawall]

I've just committed NNEDI3 chroma scaler (shader version only for now). If you like you can test it out with the latest source on GitHub.
Thanks. Is a new MPDN test build required for this? With 2.42 I get an error that it couldn't load extensions when using Github version.

[Zachs]

Hmm I'll have a look. BTW I'm still working on it so the GitHub master might not work properly.

[Zachs]

I've just committed what I believe should be the working version of the NNEDI3 chroma scalers for both shader and OpenCL.
I also verified that it runs without a problem with MPDN 2.42.0.3380.

[Zachs]

Comodo reporting that there is a trojan in one of your files. Can you submit it for analysis or write them an email/ticket.

Can you let me know what the full false-positive name of that trojan is? And which files did it report the false-positive?

I'm trying to submit it using this link: https://www.comodo.com/home/internet-security/submit.php

[aufkrawall]

What kind of crazy service is this? :D

But somehow it still refuses with this error :confused: :
http://abload.de/img/erroru5suh.jpg (http://abload.de/image.php?img=erroru5suh.jpg)
Deleting MPDN's config doesn't help.

[Zachs]

Can you try deleting the extensions cache by issuing a flush cache command line argument?

[Zachs]

And make sure you're not unzipping it into a non empty extensions folder.

[aufkrawall]

Everything's in the right folder and I did flushcache, doesn't help.
Somehow non-binary scripts here stopped working with the recent changes.
Two days ago it was just working fine.

[Zachs]

That's strange. I just tested it again, this time with the zip file downloaded from GitHub directly - it still works fine...

EDIT: I see the problem now. For some reason with the AnyCPU edition it works fine...

[Zachs]

Right. Can you try test build 3382 to see if it fixes the problem?

[aufkrawall]

Fixed. :)
Will test NNEDI3 chroma soon, I'll first have to do some gardening. :o

[aufkrawall]

It seems to be working well, except if MPDN's internal luma scalers are used at the same time. Then aliasing is introduced.
However, it works fine in conjunction with any kind of render script luma upscaling.

NNEDI3 for chroma looks as good as I have hoped: black lines crossing red mostly don't show any aliasing and are clearly defined. There is no nasty ringing boosted with low quality sources either.
:cool:

[elraffa]

If I want to use NNEDI3 chroma scaler, how do I configure script render?
http://i.imgur.com/to08rlo.jpg

[aufkrawall]

It is not yet officially released in a stable version, you have to grab the latest render scripts from Github.
But I rather recommend to wait until it is in final binary, usually things get there very quickly.

[Shiandow]

By the way, this was already announced before, but with the latest extensions you can now copy settings from the Script Chain dialog to clipboard. It stores these settings as text, so you can then paste them into a doom9 comment, and someone else could copy this text and paste it into MPDN as a RenderScript. For technical reasons this only works in the Script Chain and Script Group dialogs, though.

This makes it significantly easier to share your settings with someone else. Which, with the complexity MPDN allows, could save a lot of time (I think MPDN might be the only video player with a Turing complete settings dialog).

For example, here is a script which combines a Script Group and a Deband script to make a debanding script which can be toggled on and off by pressing "Ctrl+Shift+D":


<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="Debanding" Guid="2eedac80-a1d3-4ad9-a8e6-027464debb1c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptGroup.ScriptGroupScript">
<Settings>
<SelectedIndex>1</SelectedIndex>
<Hotkey>Ctrl+Shift+D</Hotkey>
<Options>
<Preset Name="Off" Guid="aecfa54b-7ac8-4f37-a2b5-ad68a482c9f0">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</Preset>
<Preset Name="On" Guid="5508d6a2-9ec8-4b0d-8c11-22f2c96d406e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>8</maxbitdepth>
<power>0.5</power>
<grain>False</grain>
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
</ListOfPreset>

[aufkrawall]

Neat. I also noticed that you can do copy&paste now in the settings dialogs themselves, which is also a cool thing (e.g. to clone script chains or groups).

[Anima123]

It seems to be working well, except if MPDN's internal luma scalers are used at the same time. Then aliasing is introduced.
However, it works fine in conjunction with any kind of render script luma upscaling.

NNEDI3 for chroma looks as good as I have hoped: black lines crossing red mostly don't show any aliasing and are clearly defined. There is no nasty ringing boosted with low quality sources either.
:cool:

Do I have to add a resizer at the end of Script chain to use luma upscaling?

What's your configuration for NNEDI3 chroma scaling, 16 or 32 neurons?

[Zachs]

It seems to be working well, except if MPDN's internal luma scalers are used at the same time. Then aliasing is introduced.

That's because there's a bug in the framework that causes it to stretch the resulting NNEDI3 chroma. I'll see if I can fix it, if not I'll get Shiandow to have a look.

EDIT: This should be fixed with the latest commits, but make sure you don't use NNEDI3 chroma with non 4:2:0 sources yet. I'll need to check with Shiandow to see how we're going to handle it in the framework.

EDIT2: Non 4:2:0 sources should now work.

[Zachs]

I think MPDN might be the only video player with a Turing complete settings dialog.

In case anyone's wondering what "Turing complete" means: https://en.wikipedia.org/wiki/Turing_completeness

[Zachs]

And here's another decent script I like to use:


<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="SuperChromaRes" Guid="16227e78-1997-460b-bd78-f76b332237e7">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</Preset>
<Preset Name="Debanding" Guid="2eedac80-a1d3-4ad9-a8e6-027464debb1c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptGroup.ScriptGroupScript">
<Settings>
<SelectedIndex>1</SelectedIndex>
<Hotkey>Ctrl+Shift+D</Hotkey>
<Options>
<Preset Name="Off" Guid="aecfa54b-7ac8-4f37-a2b5-ad68a482c9f0">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</Preset>
<Preset Name="On" Guid="5508d6a2-9ec8-4b0d-8c11-22f2c96d406e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>8</maxbitdepth>
<power>0.5</power>
<grain>False</grain>
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
<Preset Name="NNEDI3" Guid="bae19540-a4d6-49af-ad00-194a19de9c0e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.NNedi3Scaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons32</Neurons2>
<CodePath>UnrolledVectorDot</CodePath>
<Structured>False</Structured>
<ChromaScalers>
<ChromaScalerPreset Name="None" Guid="6aa7ce83-40c6-4af3-932e-442fad3449bc">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="ChromaScaler" Guid="52c90742-3348-41b2-917d-70b77246db3c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Chroma.ChromaScaler">
<Settings>
<B>0.3333333</B>
<C>0.3333333</C>
<Preset>MitchellNetravali</Preset>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="EWA Chroma Scaler" Guid="770eb8ed-bede-4c35-a15c-a115693f0abe">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Eight</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincBlackman" />
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="SuperChromaRes" Guid="926e48ac-220e-4d17-9d28-64028740bb74">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</ChromaScalerPreset>
</ChromaScalers>
<ChromaScalerGuid>d93e8c6f-1a4c-40d2-913a-3773c00d1541</ChromaScalerGuid>
</Settings>
</Script>
</Preset>
<Preset Name="EWA Scaler" Guid="d8eeb145-35fe-461a-b3dd-29f99f68240a">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerScaler">
<Settings>
<TapCount>Eight</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincBlackman" />
</Settings>
</Script>
</Preset>
</ListOfPreset>


EDIT: There's a bug that if you clear the list and paste again you'll get an error. The workaround for now is clear then hit OK to close the dialog and reopen before you paste again. This has been fixed in the GitHub repository.

[Zachs]

If I want to use NNEDI3 chroma scaler, how do I configure script render?

NNEDI3 Chroma Scaler is now available.

Usage example:


<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="NNEDI3 Chroma Doubler" Guid="a9367f14-7dd3-4b69-8e4f-9edee4a7db8e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.Chroma.NNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons32</Neurons2>
<CodePath>UnrolledVectorDot</CodePath>
<Structured>False</Structured>
</Settings>
</Script>
</Preset>
<Preset Name="NNEDI3" Guid="434cca2b-f1c2-4ef9-ada8-a78c1e437fc8">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.NNedi3Scaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons32</Neurons2>
<CodePath>UnrolledVectorDot</CodePath>
<Structured>False</Structured>
<ChromaScalers>
<ChromaScalerPreset Name="None" Guid="0715cc9b-43c5-4890-96cc-6e24efb3ed61">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="ChromaScaler" Guid="8d6c6371-ea0d-4c5b-b789-6f7516983408">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Chroma.ChromaScaler">
<Settings>
<B>0.3333333</B>
<C>0.3333333</C>
<Preset>MitchellNetravali</Preset>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="EWA Chroma Scaler" Guid="309ae2b6-a5d2-43f0-9bca-fd4f633a4b63">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Four</TapCount>
<AntiRingingEnabled>False</AntiRingingEnabled>
<AntiRingingStrength>0.85</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScaler+JincScaler" />
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="NNEDI3 Chroma Doubler" Guid="976ea251-25dd-4eef-a5a6-58e4244c954f">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.Chroma.NNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons32</Neurons2>
<CodePath>UnrolledVectorDot</CodePath>
<Structured>False</Structured>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="OpenCL NNEDI3 Chroma Doubler" Guid="dc47c5e5-abc6-4f01-b211-f28f056e2eca">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.Chroma.OclNNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="SuperChromaRes" Guid="e4cc356c-5e8d-4cb3-94e6-e1650f75e3b0">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</ChromaScalerPreset>
</ChromaScalers>
<ChromaScalerGuid>994c176f-ab9f-47e0-81fe-dc20609a40c2</ChromaScalerGuid>
</Settings>
</Script>
</Preset>
</ListOfPreset>

[Shiandow]

By the way, this is a somewhat minor issue, but would it be possible to make the OSD work when no videos are being played?

[Zachs]

By the way, this is a somewhat minor issue, but would it be possible to make the OSD work when no videos are being played?

I'll see what I can do but I remember it would involve a fair bit of changes the last time I looked at it.

[Zachs]

Similar to Render Script Chain, Audio Chain allows you to chain multiple audio scripts for audio processing.

Example (DRC --> Reclock):

<ListOfIExtensionUi xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<DynamicRangeCompressor yaxlib:realtype="Mpdn.Extensions.AudioScripts.Mpdn.DynamicRangeCompressor">
<Settings>
<ThresholddB>-15</ThresholddB>
<Ratio>3</Ratio>
<MakeupGaindB>3</MakeupGaindB>
<AttackMs>100</AttackMs>
<ReleaseMs>200</ReleaseMs>
</Settings>
</DynamicRangeCompressor>
<Reclock yaxlib:realtype="Mpdn.Extensions.AudioScripts.Mpdn.Reclock">
<Settings />
</Reclock>
</ListOfIExtensionUi>

[Zachs]

@Shiandow BTW there seems to be a bug in NNEDI3 chroma scaler when used as NNEDI3's chroma scaler. It would seem the chroma scaler is active anyway even when we're downscaling (i.e. NNEDI3 Luma itself is inactive).
Anyway I'm going call it a night!

[Zachs]

Just a quick heads up. There's a bug in the release where audio scripts won't run for some reason - it's complaining that Cudafy.NET cannot be resolved. It's late here so I'll try to fix it some other time. In short, you can't use audio scripts yet.

[Shiandow]

@Shiandow BTW there seems to be a bug in NNEDI3 chroma scaler when used as NNEDI3's chroma scaler. It would seem the chroma scaler is active anyway even when we're downscaling (i.e. NNEDI3 Luma itself is inactive).
Anyway I'm going call it a night!

That's odd, as far as I can tell it shouldn't even reach that part of the code. In fact, I don't seem to be able to reproduce the problem.

Oh well, we'll figure it out tomorrow.

[aufkrawall]

Do I have to add a resizer at the end of Script chain to use luma upscaling?

No, it will automatically use chosen upscalers if upscaling is needed.


What's your configuration for NNEDI3 chroma scaling, 16 or 32 neurons?
16 neurons should be enough. Maybe it can make sense to use 32 neurons for first pass if you're watching lots of cartoons, but in most cases 16 should be sufficient.
And even if there are artifacts due to low count of neurons, they are probably harder to spot than with luma.
16 neurons are already more expensive than 8 taps of EWA scalers.

[Dlget]

And here's another decent script I like to use:

As i can't use that setting for all sources [sd rendering time was 10-15ms but hd content took nearly 35-40ms & same for 10bit anime 35-40ms]
So i wanted to make different script groups & toggle between them & use best suited script group for individual content[instead of using condition,i wanted to select groups manually by hotkey].
I created a script group using your script.But when i toggle,i see no change in rendering time.[i hoped it'll increase].


<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="Max" Guid="4bea4c95-b0c7-451d-9486-fc53892ff16f">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptGroup.ScriptGroupScript">
<Settings>
<SelectedIndex>1</SelectedIndex>
<Hotkey>Ctrl+Shift+D1</Hotkey>
<Options>
<Preset Name="Off" Guid="e6bb7235-c6fa-4685-9167-328881496139">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</Preset>
<Preset Name="On" Guid="3115834f-ba83-4be2-9066-2e29f93d19b7">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptGroup.ScriptGroupScript">
<Settings>
<SelectedIndex>1</SelectedIndex>
<Hotkey></Hotkey>
<Options>
<Preset Name="SuperChromaRes" Guid="16227e78-1997-460b-bd78-f76b332237e7">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</Preset>
<Preset Name="Debanding" Guid="2eedac80-a1d3-4ad9-a8e6-027464debb1c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptGroup.ScriptGroupScript">
<Settings>
<SelectedIndex>-1</SelectedIndex>
<Hotkey>Ctrl+Shift+D</Hotkey>
<Options>
<Preset Name="Off" Guid="aecfa54b-7ac8-4f37-a2b5-ad68a482c9f0">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</Preset>
<Preset Name="On" Guid="5508d6a2-9ec8-4b0d-8c11-22f2c96d406e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>8</maxbitdepth>
<power>0.5</power>
<grain>False</grain>
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
<Preset Name="NNEDI3" Guid="bae19540-a4d6-49af-ad00-194a19de9c0e">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.NNedi3Scaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons32</Neurons2>
<CodePath>UnrolledVectorDot</CodePath>
<Structured>False</Structured>
<ChromaScalers>
<ChromaScalerPreset Name="None" Guid="6aa7ce83-40c6-4af3-932e-442fad3449bc">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="ChromaScaler" Guid="52c90742-3348-41b2-917d-70b77246db3c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Chroma.ChromaScaler">
<Settings>
<B>0.3333333</B>
<C>0.3333333</C>
<Preset>MitchellNetravali</Preset>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="EWA Chroma Scaler" Guid="770eb8ed-bede-4c35-a15c-a115693f0abe">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Eight</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincBlackman" />
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="SuperChromaRes" Guid="926e48ac-220e-4d17-9d28-64028740bb74">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</ChromaScalerPreset>
</ChromaScalers>
<ChromaScalerGuid>d93e8c6f-1a4c-40d2-913a-3773c00d1541</ChromaScalerGuid>
</Settings>
</Script>
</Preset>
<Preset Name="EWA Scaler" Guid="d8eeb145-35fe-461a-b3dd-29f99f68240a">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerScaler">
<Settings>
<TapCount>Eight</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincBlackman" />
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
</ListOfPreset>

In OSD we can't see which scripts are active or which upscaling algo is used.It should be implemented so that we can know that selected script is working[like in madVR]

[Shiandow]

As i can't use that setting for all sources [sd rendering time was 10-15ms but hd content took nearly 35-40ms & same for 10bit anime 35-40ms]
So i wanted to make different script groups & toggle between them & use best suited script group for individual content[instead of using condition,i wanted to select groups manually by hotkey].
I created a script group using your script.But when i toggle,i see no change in rendering time.[i hoped it'll increase].


[...]



For whatever reason the "EWA" scaler seems to be disabling NNEDI3 somehow. That shouldn't happen. I also noticed some minor bugs that lowered the performance a bit.

Edit: Oh nvm, I misdiagnosed the problem.

You're using a Script Group instead of a Script Chain, which means that it doesn't do chroma scaling, then debanding, then NNEDI3, then EWA. But it only does one of those.

If you copy paste the things in your "On" preset to a Script Chain then it should work.

[aufkrawall]

MPDN is somehow wasting lots of GPU power when doing downscaling.
I'm downscaling 4k 60fps down to WQHD with just bilinear scaling for chroma and luma, and GPU usage is at 70% with full clocks.
In madVR with the same settings (I checked them twice, no dithering etc., same queue lengths...), GPU usage is only at 30% with lower clocks.
So something weird must be going on.
Sample:
http://distribution.bbb3d.renderfarming.net/video/mp4/bbb_sunflower_2160p_60fps_normal.mp4

With the latest MPDN versions, I also noticed that video input colorimetric often sets itself to PC BT.601 and improve chroma reconstruction turns itself off. I suppose this randomly happens when changing other settings.

[Dlget]

Thanks for fast reply.Now it's working fine.Time to make more Script chain & test them.
still script group & script chain is still bit confusing

[Shiandow]

With the latest MPDN versions, I also noticed that video input colorimetric often sets itself to PC BT.601 and improve chroma reconstruction turns itself off. I suppose this randomly happens when changing other settings.

Hmm, sounds like issue #142 (https://github.com/zachsaw/MPDN_Extensions/issues/142)might still not be completely solved after all. Although, did you check if the settings returned to the proper values when the playback has ended?

[aufkrawall]

Haven't watched for this yet, but I suppose the settings stayed wrong between several sessions.
I think it also occured when I was playing around with render scripts or at least MPDN scalers.

Edit: Yes, settings stay wrong between different MPDN process sessions.

[Shiandow]

Haven't watched for this yet, but I suppose the settings stayed wrong between several sessions.
I think it also occured when I was playing around with render scripts or at least MPDN scalers.

Edit: Yes, settings stay wrong between different MPDN process sessions.

Yeah, I think I just saw something similar. No clue how it happened though. If you've found any way to reproduce it that isn't entirely random then that would help a lot.

I fear that it could be random though, caused by some weird race condition...

[Zachs]

MPDN is somehow wasting lots of GPU power when doing downscaling.
I'm downscaling 4k 60fps down to WQHD with just bilinear scaling for chroma and luma, and GPU usage is at 70% with full clocks.
In madVR with the same settings (I checked them twice, no dithering etc., same queue lengths...), GPU usage is only at 30% with lower clocks.
So something weird must be going on.
Sample:
http://distribution.bbb3d.renderfarming.net/video/mp4/bbb_sunflower_2160p_60fps_normal.mp4

MPDN uses a box filter when you do bilinear downscaling. Hardware bilinear is similar to MPDN's nearest neighbour. Not sure if madVR does the same thing now but the last I checked it didn't so they are completely different scalars.

[Zachs]

Hmm, sounds like issue #142 (https://github.com/zachsaw/MPDN_Extensions/issues/142)might still not be completely solved after all. Although, did you check if the settings returned to the proper values when the playback has ended?
I did notice this too. I think I know what is causing it. MPDN seems to crash occasionally when you close it not it doesn't show up as a crash. And it had to happen in between the settings commit event pairs.

[aufkrawall]

MPDN uses a box filter when you do bilinear downscaling. Hardware bilinear is similar to MPDN's nearest neighbour. Not sure if madVR does the same thing now but the last I checked it didn't so they are completely different scalars.
When I choose Lanczos with 4 taps for both chroma and luma in madVR, I only have 55% GPU usage.
So there simply must be something going wrong in MPDN's downscaling.
Even with nearest neighbor I get 70% usage in MPDN...

[Zachs]

What do you get with MPDN's 4 tap downscaler?

Edit: do you have render script chroma scaler active?

[aufkrawall]

I forgot to disable frame blending and dithering in madVR.
New results:

Lanczos 4 taps w.o. AR for luma & chroma:
madVR: 46% gpu usage@1126Mhz
MPDN: 75%@1421Mhz

Of course render scripts are completely disabled. Image quality mode is "image quality".
When doing upscaling or no luma scaling, MPDN and madVR pretty much have the same GPU load amount. Just with downscaling there is this huge difference.

[Zachs]

Something is definitely wrong. 4 taps and 2 taps can't only give you a difference of 5%. Also with nearest neighbour you should be getting way less since you're not downscaling much at all. Only at higher downscaling factors it would behave differently to madVR hardware bilinear. BTW you haven't got improve chroma enabled have you?

I'll take a look anyway.

[aufkrawall]

BTW you haven't got improve chroma enabled have you?

I already checked this to be sure, it was disabled for the numbers above.

Will do some more checks and report if I find anything noteworthy.

[Zachs]

Can you also try downscaling other sources? With different resolutions?

[aufkrawall]

Will try to get a 5k source from YT...

[Zachs]

What about just 1080p to half the size?

[aufkrawall]

5120x2700 30fps@WQHD
Lanczos 4 taps (same settings as above)
MPDN: 50%1126Mhz (~always)
madVR 50%@671Mhz (or 25%@1126Mhz, driver sometimes decides differently with low GPU load)

bicubic (sharpness 50% in madVR and 100% in MPDN look pretty much identical):
MPDN: 50%@1126Mhz
madVR: 43%@587Mhz

Again, just like with the 4k 60fps clip, nearest neighbor gives the same GPU load in MPDN as Lanczos4.
With 4k 60fps it stays always at 75% GPU load, with the 5120x2700 30fps clip it is always at 50% GPU load.

Very odd, it is much too high too.

[aufkrawall]

1080p30@50%:
lanczos 4
MPDN: 31%@135Mhz
madVR: 31%@135Mhz

Seems like either the strange behavior only occurs when the resolution is high or the precision of the numbers is too low when there is too low load.

I'm getting lots of dropped frames in MPDN with the 4k 60fps video when I just enable deband filter while doing lanczos4 downscaling to WQHD. This is no problem in madVR, no droppped frames there with deband filter (of course not using any quality trade-off setting).

[Zachs]

That's just really strange. It's the same algorithm that does 50% scaling. It really shouldn't matter which resolution it started off with. Is the target size the same for both renderers?

[aufkrawall]

Yes, of course. Apart from that 1080p 50% scaling, I always scaled to my native WQHD resolution, which results in the same target rectangle.
I also tried FSE and old windowed mode path, doesn't make a difference.

[Zachs]

I'll see what I can do. It's hard to test for it since I don't have a high resolution display.

[Anima123]

aufkrawall, does a clean installation, or clearing up of application data help?

[aufkrawall]

I'll see what I can do. It's hard to test for it since I don't have a high resolution display.
Maybe you can simulate it with Nvidia Dynamic Super Resolution?
But you probably even don't need a high render resolution, I'll try if I can reproduce it with 1080p display resolution.

I think the main odd thing is that there is no difference in GPU load between lanczos, bilinear and nearest neighbor.

Anima123: I reinstall MPDN very frequently as I often try around with test builds and GitHub render scripts.
For this reason I just completely reinstalled it today. I will try it again, but I think it's very unlikely that it will make any difference.

[aufkrawall]

Issue also happens when downscaling to 1080p (e.g. the Big Buck Bunny video from above) or with totally clean installation of MPDN without render scripts.

[Zachs]

OK I'll have a look.

[aufkrawall]

Please compare lanczos vs nearest neighbor in terms of performance (GPU load , clock etc.).
The way NN is implemented in MPDN it should still be quite faster than lanczos, right?

[Zachs]

That's right. It should be massively faster. Especially when you're downscaling just 50%.

[aufkrawall]

I tested again 1080p30 -> 50%, this time without DXVA2 CB (seems to have a little overhead on GPU load).
With madVR, GPU usage drops from 28% to 18% when switching from lanczos4 to NN.
With MPDN, there is again that strange behavior that there is almost no difference between lanczos4 and NN, both are around ~26%.

Probably this is already a symptom by the same bug that's to blame for the high GPU load when downscaling from very high resolution.
But since it's hardly noticeable (at least with fast GPUs), maybe it's really better to focus on that 4k 60fps video, which is much more performance critical.

[x7007]

What do I get from this ? Does it gonna play using NNEDI3 for all 1080p and 720p ? is Sinc-Blackman better than Jinc ?
The renderer time goes from 6.85ms Windowed mode to 22.33ms FSE DX11 Exclusive mode . what does it mean ?
I've added Deband too after


http://i.imgur.com/KfnJIuQ.png

[ryrynz]

What do I get from this ? Does it gonna play using NNEDI3 for all 1080p and 720p ? is Sinc-Blackman better than Jinc ?


NNEDI3 will be used for all resolutions that require upscaling to target size. You have three chroma resizes there.. drop Chromascaler or EWA depdending on what you prefer. Sinc-Blackman seems to be popular, you might prefer it.

[Zachs]

Could you guys give the audio scripts on GitHub a try to see if it works on your machine now please?
It should work now.

[Zachs]

Hmm, sounds like issue #142 (https://github.com/zachsaw/MPDN_Extensions/issues/142)might still not be completely solved after all. Although, did you check if the settings returned to the proper values when the playback has ended?
Can you try it now? I think I found the problem. The variables for the original settings just weren't initialised.

[ryrynz]

Can you try it now? I think I found the problem. The variables for the original settings just weren't initialised.

Working fine for me.

[aufkrawall]

Is there a way to allow render scripts to only run once?
I defined various conditionals with different chroma upscalers and my goal is to run deband:
-after chroma is upscaled to source resolution
-to run it only once to save performance and detail loss
-if luma is downscaled, deband should run after this to save performance

All three points should be fulfilled together and I would like to renounce to stick to a hotkey solution (which I find rather ugly).

Any ideas? :)

My current script chain:
http://www69.zippyshare.com/v/yO1TwR8q/file.html

[BetA13]

i decided to try mpdn again after it looks more friendly now :)
also with the scripts..

looks nice now..

i just have one problem..
im using an 30hz hdtv that can do 1080i.
i need to use nvidia resize so i can see teh whole image. (no sides cut off)

now, when i use mpdn and go into fullscreen, theres a lot of stuff missing on teh sides, its like the player ignores teh resolution set by ncvidia CP...

i tried different things now, but didnt help, its allways stuff missing somehow..

is this a known problem?
here are some infos:

win 7 64bit, latest NV Drivers.
gtx670

settings:

this is teh resizing stuff so it fits all, that i have to do:
http://abload.de/thumb/2txuoa.png (http://abload.de/image.php?img=2txuoa.png)

and this i do after, setting it like in teh screen, and teh desktop fits perfect. but player not :(
http://abload.de/thumb/1rnuun.png (http://abload.de/image.php?img=1rnuun.png)



greetz

[Shiandow]

Is there a way to allow render scripts to only run once?
I defined various conditionals with different chroma upscalers and my goal is to run deband:
-after chroma is upscaled to source resolution
-to run it only once to save performance and detail loss
-if luma is downscaled, deband should run after this to save performance

All three points should be fulfilled together and I would like to renounce to stick to a hotkey solution (which I find rather ugly).

Any ideas? :)

My current script chain:
http://www69.zippyshare.com/v/yO1TwR8q/file.html

How about something like:


Chroma Scaling
Resize (but only if downscaling)
Debanding

The "Resizer" script should have an option to scale only if the target size is smaller then the video size. Or you could use a condtional to do pretty much the same thing.

As an aside, I think the Resizer might need to be improved a bit. It has quite a lot of options, but I'm pretty sure most haven't ever been used. And I think conditionals make most of them unnecessary. Maybe it could be made scriptable, then we could just keep the sensible options and remove the rest.

[aufkrawall]

I've really no idea how to use the resizer render script and unfortunately, I can't find any documentation about it. :o

My problem is that I have conditionals for different source resolutions and that they can affect each other.
So if I'd add deband for each of it (to have it run after chroma scaling), I end up with multiple deband passes for low source resolutions.

[antonyfrn]

Hi I'm trying to move over to MPDN and need a bit of help setting up a script like I use in MadVr. Or if anyone has any suggestion on something better to use?

if (srcHeight < 361) and (deintFps < 31) "Flash"
elseif (srcHeight < 405) and (deintFps < 31) "<SD 30fps"
elseif (srcHeight < 481) and (deintFps < 31) "SD 30fps"
elseif (srcHeight < 714) and (deintFps > 59) "720p60 (cropped)"
elseif (srcHeight < 714) and (srcWidth > 1279) "720p24 (cropped)"
elseif (srcHeight < 541) and (deintFps < 26) "1/2 HD"
elseif (srcHeight < 541) and (deintFps < 31) "1/2 HD 30fps"
elseif (srcHeight < 721) and (deintFps < 26) "720p24"
elseif (srcHeight < 721) and (deintFps < 31) "720p30"
elseif (srcHeight < 721) and (deintFps < 51) "720p50"
elseif (srcHeight < 721) and (deintFps > 51) "720p60"
elseif (srcHeight < 1040) and (srcWidth > 1919) "1080p24 (cropped)"
elseif (deintFps < 26) "1080p24"
else "1080i30"

[Braum]

Hi I'm trying to move over to MPDN and need a bit of help setting up a script like I use in MadVr. Or if anyone has any suggestion on something better to use?

if (srcHeight < 361) and (deintFps < 31) "Flash"
elseif (srcHeight < 405) and (deintFps < 31) "<SD 30fps"
elseif (srcHeight < 481) and (deintFps < 31) "SD 30fps"
elseif (srcHeight < 714) and (deintFps > 59) "720p60 (cropped)"
elseif (srcHeight < 714) and (srcWidth > 1279) "720p24 (cropped)"
elseif (srcHeight < 541) and (deintFps < 26) "1/2 HD"
elseif (srcHeight < 541) and (deintFps < 31) "1/2 HD 30fps"
elseif (srcHeight < 721) and (deintFps < 26) "720p24"
elseif (srcHeight < 721) and (deintFps < 31) "720p30"
elseif (srcHeight < 721) and (deintFps < 51) "720p50"
elseif (srcHeight < 721) and (deintFps > 51) "720p60"
elseif (srcHeight < 1040) and (srcWidth > 1919) "1080p24 (cropped)"
elseif (deintFps < 26) "1080p24"
else "1080i30"
You need to use the Script Chain and within it use the Conditional function :

http://reho.st/self/7df07f301f0b39b05a76d1a0bb447df405ce1ae1.png

[antonyfrn]

You need to use the Script Chain and within it use the Conditional function :




When i go to create one of these scripts i get the follow error.

Error in render script ('Conditional'):
Expected ')'

Any idea?


If SourceSize.Width >=1400 and SourceSize.Height >=1080, use 'Script Chain'
If SourceSize.Width <1400 and SourceSize.Height <1080 and SourceSize.Width >=920 and SourceSize.Height >=720, use 'Script Chain'
If SourceSize.Width < 920 and SourceSize.Height <720, use 'Script Chain'

[aufkrawall]

Maybe it will be more obvious what is wrong if you copy&paste the real render script.

[antonyfrn]

Here you go.

<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="Script Chain" Guid="8e0703a7-8d8d-4c31-9d6b-b03074c22cc7">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptChain.ScriptChainScript">
<Settings>
<Options>
<Preset Name="High" Guid="d1e0f2be-2263-45c4-bd0a-014814d55401">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.Conditional.ConditionalUi">
<Settings>
<Condition />
<Preset />
</Settings>
</Script>
</Preset>
<Preset Name="Medium" Guid="7353c2f3-74ae-4bb0-b75d-daed30b0d070">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.Conditional.ConditionalUi">
<Settings>
<Condition />
<Preset />
</Settings>
</Script>
</Preset>
<Preset Name="Low" Guid="ed168d4e-50e0-44c7-802b-e74a7099d9dc">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.Conditional.ConditionalUi">
<Settings>
<Condition />
<Preset />
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
</ListOfPreset>

[aufkrawall]

Did you put something by mistake in the conditional language box that doesn't belong there, like "if" etc.?

[eX``]

Hello everyone,

First post despite a close follow-up =)

What kind of crazy service is this? :D

But somehow it still refuses with this error :confused: :
http://abload.de/img/erroru5suh.jpg (http://abload.de/image.php?img=erroru5suh.jpg)
Deleting MPDN's config doesn't help.

I'd like to bump this message, because since the 2.40 I can't launch MPDN with extensions without going through this message.

In other words, no extensions gives a message "You are missing a lot of features not using the super mega great extensions, click here etc etc...". When I have the extensions folder filled, I have the error message.
Besides, I tried with .exe MPDN + .exe extensions (both x86 & x64), standalone x64 both, anyCPU + extensions, I even tried standalone + source. Always trying 2-3 times each, always cleaning (uninstall, ccleaner, reg, ...) inbetween, but nothing seem to be working and I cannot figure out where the problem stands.

If you have any idea what could be the problem, I am all ears =)

FYI, I repaired .net 4 and nothing changed. Then reinstalled, same issue. I am running win7 x64. The only thing I did not do because I never encountered any error is repairing rundll32.

Thank you very much and see you around :thanks:

[aufkrawall]

IF MPDN itself is running, there probably hardly is any problem with any Windows/.NET component.
By cleaning, you also mean deleting the MPDN config folder @user\appdata\local?

[eX``]

Yup, un-installing, cleaning reg, sweeping local, going ccleaner.

[Zachs]

Yup, un-installing, cleaning reg, sweeping local, going ccleaner.

Can you enable debug dialog and give me the stack trace please?

[eX``]

This is the sort version, let me know if you need more details (I'll edit)
TITLE: CSScriptLibrary Error
------------------------------

An unexpected error 'csscript.CompilerException' has occurred.

------------------------------
ADDITIONAL INFORMATION:

(0,0): error CS2007: Unrecognized option: '/highentropyva'
(CSScriptLibrary)

------------------------------
BUTTONS:

&Ignore
&Abort
------------------------------

[Zachs]

This is the sort version, let me know if you need more details (I'll edit)

OK I know what's causing the problem. Cudafy (included recently in the extensions for audio scripts) drags in Mono.CSharp which takes over Microsoft.CSharp when compiling the scripts. I've made the script compiler compatible with Mono for the next release. And the reason it happens on some machines but not others is because it depends on the order the extensions are loaded (i.e. if audio scripts are loaded last, it'll be fine).

[aufkrawall]

Did you also have time to look into the downscaling issue? :)

[elraffa]

Hi zachs.. could you include Nvidia's Dynamic Super Resolution (DSR) feature in MPDN? It force to resize the video to very high resolutions (like 4K) even if you have a lower resolution monitor (1920x1080).

[aufkrawall]

MPDN can already do this, best done with NNEDI3 or Super-xbr.
DSR has very poor downscaling quality.

[Zachs]

Looking into the downscaling issue. I've found several oddities with it so far. One involves using the wrong shader altogether which may have caused the problem. Not 100% sure yet.

[aufkrawall]

Thanks, glad it isn't something weirdly hidden.

[elraffa]

MPDN can already do this, best done with NNEDI3 or Super-xbr.
DSR has very poor downscaling quality.

could you give me a script' example to do this?

[aufkrawall]

When you run NNEDI3 or super-xbr, the resolution is always doubled and then downscaled, if necessary.
You can run it in a row to quadruple etc. resolution.

However, it doesn't really make sense to scale to a much higher resolution than the display resolution is, because
1. the scaling is expensive
2. more scaling artifacts are introduced.

May I ask what effect of DSR you like it has on video?
If it's the blur, there might be a simple shader that achieves the same result.

[Zachs]

I tested again 1080p30 -> 50%, this time without DXVA2 CB (seems to have a little overhead on GPU load).
With madVR, GPU usage drops from 28% to 18% when switching from lanczos4 to NN.
With MPDN, there is again that strange behavior that there is almost no difference between lanczos4 and NN, both are around ~26%.

Probably this is already a symptom by the same bug that's to blame for the high GPU load when downscaling from very high resolution.
But since it's hardly noticeable (at least with fast GPUs), maybe it's really better to focus on that 4k 60fps video, which is much more performance critical.

I tried 1080p24 -> 50% but I see an increase in GPU load from 33% to 59% going from Bilinear to Lanczos4. Can you do a screen shot with debug OSD?

The problem I mentioned earlier was actually a separate bug and has been fixed for the next release.

[Shiandow]

Hi zachs.. could you include Nvidia's Dynamic Super Resolution (DSR) feature in MPDN? It force to resize the video to very high resolutions (like 4K) even if you have a lower resolution monitor (1920x1080).

That feature makes zero sense when rendering videos. With videos the difficult part is that you start at a low resolution and need to end at a possibly higher resolution (or lower but that's easier). With games you start at a theoretically infinite resolution but you need to lower this to a finite resolution, since it would take infinitely long to render otherwise.

These processes are completely different.

You can force MPDN to upscale to a resolution which is far too high, and then scale it all down again, but there's no point in doing so. With some algorithms like NNEDI3 you're forced to scale the image slightly too much, but it would have been better if that wasn't the case. It makes no sense to force this behaviour.

[elraffa]

When you run NNEDI3 or super-xbr, the resolution is always doubled and then downscaled, if necessary.
You can run it in a row to quadruple etc. resolution.

However, it doesn't really make sense to scale to a much higher resolution than the display resolution is, because
1. the scaling is expensive
2. more scaling artifacts are introduced.

May I ask what effect of DSR you like it has on video?
If it's the blur, there might be a simple shader that achieves the same result.

I change the resolution in nvidia control panel to 4k in my 1080p screen and the image is more clear, when I apply adaptive sharpen, the image has less artifacts and noise..

[aufkrawall]

I tried 1080p24 -> 50% but I see an increase in GPU load from 33% to 59% going from Bilinear to Lanczos4. Can you do a screen shot with debug OSD?

NN downscaling:
http://www.ld-host.de/uploads/thumbnails/d430387adb24a10ba9f710c8c5a10035.png (http://www.ld-host.de/show/d430387adb24a10ba9f710c8c5a10035.png)

Lanczos 4:
http://www.ld-host.de/uploads/thumbnails/63d1509c20b21263319920b1a48d5c75.png (http://www.ld-host.de/show/63d1509c20b21263319920b1a48d5c75.png)

FM and dithering are enabled, but that didn't make any difference in the past examples.

[Zachs]

I can't seem to be able to replicate the issue. I'm still seeing a massive difference between Lanczos4 and bilinear / nearest.
Have you got an Intel GPU to test? I'd be interested to see if it happens there as well. I get consistently lower usage on both the Intel and NVIDIA GPUs I've tested.
Just to be thorough, could you perhaps clear your app settings, replicate the issue, and then send me the config files?
I'll try to replicate it on all the GPUs I have access to (I have 8 that I usually test on).

[foxyshadis]

Time flies and it's been a year since MPDN was first released. Happy Birthday MPDN!

Just wanted to take the opportunity to thank all the hard work from the devs and all the testers in helping to make MPDN a great media player that it is today. Thank you!

MPDN v2.42.0 has been released with two key features you've been asking for: Custom chroma scaler in NNEDI3 (both shader and OpenCL) and Dynamic Range Compressor AudioScript.

It remains our top priority to bring you the following features:
Chaining of AudioScripts
NNEDI3 Chroma Scaler

Cheers!

Missed this last week; congrats on making it so far. It's gone from a bit of a joke at first (a new barebones player? really?) to my only player. Looking forward to the next year!

The new DRC is welcome, this finally makes playing comfortable whether it's quiet or really noisy out. Reading ReplayGain tags would be a great integration, as well as having the volume control be able to boost well beyond 100% in order to take maximum advantage of compression/limiting. I wonder if lowering attack/decay the more volume is boosted makes sense, turning it more into a limiter the louder it gets.

[Zachs]

Just a quick heads up. There's a bug in the release where audio scripts won't run for some reason - it's complaining that Cudafy.NET cannot be resolved. It's late here so I'll try to fix it some other time. In short, you can't use audio scripts yet.

MPDN Extensions 1.18.1 should have this fixed.

[Zachs]

Missed this last week; congrats on making it so far. It's gone from a bit of a joke at first (a new barebones player? really?) to my only player. Looking forward to the next year!

The new DRC is welcome, this finally makes playing comfortable whether it's quiet or really noisy out. Reading ReplayGain tags would be a great integration, as well as having the volume control be able to boost well beyond 100% in order to take maximum advantage of compression/limiting. I wonder if lowering attack/decay the more volume is boosted makes sense, turning it more into a limiter the louder it gets.

Thanks! :) This is a great forum to make something magical happen.

The default settings isn't the best - I actually use a 0.1ms/0.2ms attack/decay myself. Anyway if more users could play around and recommend better defaults, I'm open to changing it.

As for ReplayGain tags, care to explain a little more?

Not sure how we'd be able to boost beyond 100% though?

[Braum]

Hi ! I mostly watch Bluray (1920x1080) on a 1680x1050 screen (I'm planning on buying a 21/9 29" 4K display) and I use this render script :

Render Script :

OpenCL NNEDI3 128/32
Deband
Image Processing (Finesharp shader http://forum.doom9.org/showthread.php?t=171346, custom settings)

Default scaler :

Luma Downscaler Sinc-Jinc 16 tap AR
Chroma Downscaler Sinc-Blackman 16 tap AR


Is there a way to have the downscaling part within the render script ? I didn't find a way but maybe I've missed something.

Do you see improvement ? Or flaws ?

Thanks ! :)

[ryrynz]

Just a heads up for anyone upgrading extensions from 1.18 to 1.18.1, check the input colormetric and improve chroma reconstruction options afterwards.
The upgrade won't set these back to defaults and 1.18 likely has changed those values, so set them accordingly afterwards.

It's gone from a bit of a joke at first (a new barebones player? really?) to my only player. Looking forward to the next year!


Yeah it fleshed out super quick.. The core was strong and added to in a rapid fashion. The extensions framework was added a few devs came on board and then BOOM.

[aufkrawall]

I can't seem to be able to replicate the issue. I'm still seeing a massive difference between Lanczos4 and bilinear / nearest.
Have you got an Intel GPU to test?

Unfortunately, no.


Just to be thorough, could you perhaps clear your app settings, replicate the issue, and then send me the config files?
I'll try to replicate it on all the GPUs I have access to (I have 8 that I usually test on).
http://www37.zippyshare.com/v/SMtciqMk/file.html
Have a GPU load of 70% with it with the 4k 60fps video, while just nearest neighbor is used.
With Lanczos 16 taps it's 76%.

[eX``]

OK I know what's causing the problem. Cudafy (included recently in the extensions for audio scripts) drags in Mono.CSharp which takes over Microsoft.CSharp when compiling the scripts. I've made the script compiler compatible with Mono for the next release. And the reason it happens on some machines but not others is because it depends on the order the extensions are loaded (i.e. if audio scripts are loaded last, it'll be fine).

Thank you. Will be able to massively bang my head on the wall now, with all those scripts :p

[Zachs]

Unfortunately, no.


http://www37.zippyshare.com/v/SMtciqMk/file.html
Have a GPU load of 70% with it with the 4k 60fps video, while just nearest neighbor is used.
With Lanczos 16 taps it's 76%.

I think the downscaler shader code just doesn't run very well on Maxwell. It's written in a generic way to support custom linear scalers but for the built in scalers I'll see if I could optimise it differently.

[aufkrawall]

I might be able to try with a GK110 GPU.

[Zachs]

I might be able to try with a GK110 GPU.

Can you give test build 3400 a try?

EDIT: Nevermind. I see this on my 560GTX too. My NVS4200M behaves differently though. Not sure what to make of it... And using bilinear downscaling, my 560GTX is only twice as fast as my Intel HD 3000 when it's usually a few order of magnitudes faster.

EDIT2: Also tested a really old version where MPDN still had a hardware bilinear downscaler - it did it there too!

EDIT3: Found the problem but haven't found out how to fix it. 90% of the time is spent uploading texture to the GPU!

EDIT4: Figured out what is causing the excessive time spent on texture upload. Without this problem, Render Time has gone down from 17ms to 5.5ms with bilinear! With Lanczos4, it's now just 6.5ms on my 560GTX.

[aufkrawall]

Fixed with 2.42.3. Now I get normal 35% GPU usage with Lanczos4 downscaling (and chroma upscaling), which is less than half of the previous values. :)
Nearest neighbor is still slightly more expensive than in madVR, but this should be normal since it's an implementation with better quality which is run by the shaders instead of TMUs.

[Zachs]

Nearest neighbour is actually similar to the old MPDN hardware bilinear up to a certain downscale factor. Beyond that, it's actually much better in quality than the old bilinear. The old nearest neighbour (via TMUs) is gone. So in short, even NN is superior to the old TMU bilinear.

[Zachs]

Just in case anyone's wondering where 2.42.3 is - it's just a test build at this time and it's not actually released yet.

[aufkrawall]

Ok.
I noticed that MPDN chooses BT.2020 for any 4k video. Shouldn't it only be chosen if the video has a 10/12 bit color space?

[Zachs]

You're right! I'll fix that.

[aufkrawall]

Hm, I just noticed madVR uses BT.709 for 10 bit video as well.

[Zachs]

Hmm I can't find any definitive answers as to what default values we should use for UHD contents. Some suggest it's only for 10-/12-bit video only.

EDIT: I think the specs definitely point to BT2020 being defined for 10-/12-bit only so I'll change that. As for the defaults, I think I'm comfortable with assuming all 10-/12-bit UHD contents are BT2020 after reading more articles.

[Zachs]

Fixed with 2.42.3.

The aforementioned 2.42.3 has now been officially released.

[Braum]

In my Script Chain if I place the Debanding before the Chroma upscaler the Render Time is lower, why ?

Deband before Chroma Upscaler
http://img11.hostingpics.net/pics/832264441.png

Deband after Chroma Upscaler
http://img11.hostingpics.net/pics/165806672.png


Windows 10 64Bits

MPDN 64 updated
MPDN extensions updated


<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="Conditional" Guid="b079f42b-bc12-45e0-b541-be2f562b869d">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.Conditional.ConditionalUi">
<Settings>
<Condition>SourceSize.Width &gt;= 1400 and SourceSize.Height &gt;= 1080</Condition>
<Preset Name="Script Chain" Guid="a49ce34c-3a7e-4caa-a8cf-d8c5eaa00582">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ScriptChain.ScriptChainScript">
<Settings>
<Options>
<Preset Name="OpenCL NNEDI3 Chroma Doubler" Guid="77778766-9c6a-4d6b-bec5-00e73e078646">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.Chroma.OclNNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons128</Neurons1>
<Neurons2>Neurons32</Neurons2>
</Settings>
</Script>
</Preset>
<Preset Name="Deband" Guid="116252b7-bc90-4f7e-9f97-2c81a9e0244f">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>8</maxbitdepth>
<power>0.5</power>
<grain>True</grain>
</Settings>
</Script>
</Preset>
<Preset Name="Image Processor" Guid="0722edcd-e735-4e2c-880f-b103cf876a4a">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ImageProcessor.ImageProcessorScript">
<Settings>
<ShaderFileNames>
<String>SweetFX\ToYUV.hlsl</String>
<String>SweetFX\RemoveGrain11.hlsl</String>
<String>SweetFX\RemoveGrain4.hlsl</String>
<String>SweetFX\FineSharpA.hlsl</String>
<String>SweetFX\FineSharpB.hlsl</String>
<String>SweetFX\FineSharpC.hlsl</String>
<String>SweetFX\ToRGB.hlsl</String>
</ShaderFileNames>
<CompatibilityMode>True</CompatibilityMode>
</Settings>
</Script>
</Preset>
</Options>
</Settings>
</Script>
</Preset>
</Settings>
</Script>
</Preset>
</ListOfPreset>

[Shiandow]

In my Script Chain if I place the Debanding before the Chroma upscaler the Render Time is lower, why ?


Chroma scalers won't do anything if they're not the first script. At least not with the way it's set up now.

[Braum]

Chroma scalers won't do anything if they're not the first script. At least not with the way it's set up now.

Okay, I see, thank you ! :)

[Zachs]

@Shiandow
I think the latest changes with regards to chroma scaling optimization might've caused a bug. When using EWA Chroma Scaler alone, the final image doesn't seem to get converted back to RGB.

EDIT: This is odd. Contents up to 1080p has no problem with the EWA Chroma Scaler. Only the UHD sources will trigger the issue.

EDIT2: NNEDI3 chroma scalers are fine. ChromaScaler also has the same issue.

EDIT3: Found the problem. It would seem we need to change the chroma fallback to the following

(new ChromaFilter(lumaInput, chromaInput, null, targetSize, chromaOffset)).ConvertToYuv();


EDIT4: I've pushed the changes to GitHub. Could you take a look please?

[aufkrawall]

Chroma scalers won't do anything if they're not the first script. At least not with the way it's set up now.
Can deband render script be changed to that it's "hardcoded" to be allowed to only run one?
Then it wouldn't be such a hassle when you have plenty of conditionals for doubling and quadrupling, which will make the deband filter run multiple times (which is terrible).
Maybe it should be also "hardcoded" that the deband filter will only run after downscaling, if source resolution is downscaled?

[Dlget]

There is setting for luma & croma upscaling & downscaling in video renderer sub-menu in option.
There is also some upscalers in script chain.(confused if both work or only one)
Can you show active scalers (image & croma) in stats like MadVR so we can know what are active & what are not working although we thought it should work(may be for wrong config).
so many setting for this player.so we need more detailed status

Please consider these features for your next build

what about removing border from minimal UI?? may be

[aufkrawall]

Severe bug: colors look like crazy for 1080p video (unscaled) when using deband filter:
http://abload.de/thumb/debandtkyxc.png (http://abload.de/image.php?img=debandtkyxc.png)
Likely also other formats are affected.

[Dlget]

Severe bug: colors look like crazy for 1080p video (unscaled) when using deband filter:.

i also got same problem while using debanding.
in widowed mode crazy colored video is played while in FSE mode no video while sound is playing.

status says renderer script error

[Shiandow]

Severe bug: colors look like crazy for 1080p video (unscaled) when using deband filter:
http://abload.de/thumb/debandtkyxc.png (http://abload.de/image.php?img=debandtkyxc.png)
Likely also other formats are affected.

i also got same problem while using debanding.
in widowed mode crazy colored video is played while in FSE mode no video while sound is playing.

status says renderer script error

Is that with the latest extensions on Github? Because I think I know what's wrong in that case. The last release (v1.18.2) should be fine.

Can deband render script be changed to that it's "hardcoded" to be allowed to only run one?
Then it wouldn't be such a hassle when you have plenty of conditionals for doubling and quadrupling, which will make the deband filter run multiple times (which is terrible).
Maybe it should be also "hardcoded" that the deband filter will only run after downscaling, if source resolution is downscaled?

No. I don't like the idea of renderscripts changing their behaviour depending on which other renderscripts are active. That would open the way for some incredibly annoying bugs, or worse undefined (undefinable) behaviour.

FWIW I am planning to modify the "deband" filter to make it work before any chroma scaling happens. That would make it possible to just begin with debanding and perform any other processing afterwards.

There's already a way to ensure debanding happens after downscaling, you simply need to downscale before any debanding happens. This can be done using the Resize script, although the UI isn't too clear at the moment, there's certainly some room for improvement there.

[Funkmiser]

I can't seem to get the MPDN Extension examples to work with any of the recent versions of MPDN.

I tried dragging the KeyRemap.cs and HelloWorld.cs examples to the PlayerExtensions folder but anytime I try to run the player I get an error pop up. I tried it on the x64 install, x86 portable, and anycpu portable files.

Also would it be possible to change the filetype description from say 'MPDN MP4 Video' to simply 'MPDN Video' using an extension?

http://i.imgur.com/1bhcvaI.png

[aufkrawall]

Is that with the latest extensions on Github? Because I think I know what's wrong in that case. The last release (v1.18.2) should be fine.

It occurs with both versions.

[Shiandow]

It occurs with both versions.

Ah I managed to reproduce the problem. Should be fixed soon.

Edit: Or I have discovered an unrelated but similar problem, not entirely sure yet.

[Zachs]

Ah I managed to reproduce the problem. Should be fixed soon.

Edit: Or I have discovered an unrelated but similar problem, not entirely sure yet.

Should be fixed now with the latest commit.

[Zachs]

There is setting for luma & croma upscaling & downscaling in video renderer sub-menu in option.
There is also some upscalers in script chain.(confused if both work or only one)
Can you show active scalers (image & croma) in stats like MadVR so we can know what are active & what are not working although we thought it should work(may be for wrong config).
so many setting for this player.so we need more detailed status

Please consider these features for your next build

what about removing border from minimal UI?? may be

I've created a new feature request on GitHub for showing active scalers since it's one of the highly requested features - https://github.com/zachsaw/MPDN_Extensions/issues/157

Minimal UI border is up to Windows - for example, Windows 10 only shows a thin border at the top while Windows 8.1 and below show border all around.

[aufkrawall]

Should be fixed now with the latest commit.
It's still broken with OpenCL NNEDI3 image doubling.

Edit: Ok, it's not deband. It's OpenCL NNEDI3 itself.

[Zachs]

It's still broken with OpenCL NNEDI3 image doubling.

Ah that's a different bug.

Thanks for testing and reporting though!

[Zachs]

It was just a remnant of the changes we made to ChromaFilter - should be fixed now with the latest commit.

[aufkrawall]

Yes, fixed.

However, it would be nice of deband filter could be made more useable with other render scripts not too far away.
I either have it to fallback to MPDN's internal chroma scalers (bad for quality) or I have to run it after upscaling and before downscaling (bad performance).
At least I can't manage to run it in a better way.

[Zachs]

Hmm what are you trying to achieve with deband? I think deband should always be run at source resolution after chroma upscaling since further scaling could introduce artifacts that it isn't designed to deband.

[aufkrawall]

Ha, I think I got it working (also with Resizer render script):
*deleted*, outdated

Too many symbols to post it here as a text.
Can you take a look and see if you think something looks fishy?
I have to run deband after downscaling because else it eats up too much performance.

Edit: Ah, I have it run twice for 720p -> WQHD. Have to take a loog again.

Edit 2: But now it should be done:
http://www45.zippyshare.com/v/CYEZv6BO/file.html

Dam, that was hard. :D

[Dlget]

FSE mode is turned on or off immediately.
It i right click or click option in FSE mode,it turns off immediately & after changing my setting it turns on immediately.
It would be better if we add a delay like 2-3s

Edit:Got it now

[Zachs]

FSE mode is turned on or off immediately.
It i right click or click option in FSE mode,it turns off immediately & after changing my setting it turns on immediately.
It would be better if we add a delay like 2-3s

Change via option settings.

[Zachs]

Ha, I think I got it working (also with Resizer render script):
*deleted*, outdated

Too many symbols to post it here as a text.
Can you take a look and see if you think something looks fishy?
I have to run deband after downscaling because else it eats up too much performance.

Edit: Ah, I have it run twice for 720p -> WQHD. Have to take a loog again.

Edit 2: But now it should be done:
http://www45.zippyshare.com/v/CYEZv6BO/file.html

Dam, that was hard. :D

You'll find it easier if you did something like this instead (and drop the height comparison altogether):


If SourceSize.Width < 1280 and NeedsUpscaling, use ...
If SourceSize.Width >= 1280 and SourceSize.Width < 1920 and NeedsUpscaling, use ...
If SourceSize.Width >= 1920 and NeedsUpscaling, use ...
If SourceSize.Width > 2560 and NeedsDownscaling, use ...
etc.


Note that the If statement doesn't have an `else` following it so your 3rd line for example will be true along with the first line line if your source is < 1280x720, which isn't what you want.

The trick to make it easier to understand is to make sure each condition is unique and won't cause any overlaps. It's possible to create a chain that has overlaps too that is very useful but one step at a time.

[Zachs]

Something like this: http://pastebin.com/31jHxtHQ

[Zachs]

I can't seem to get the MPDN Extension examples to work with any of the recent versions of MPDN.

The way you've done it wouldn't have worked regardless of which version of MPDN. The extensions have always required its framework - you can't just put one file in there and expect it to magically work.

If you want to develop extensions, you need to start by downloading VS2015 (community edition is free).

[Zachs]

Can you show active scalers (image & croma) in stats like MadVR so we can know what are active & what are not working although we thought it should work(may be for wrong config).
so many setting for this player.so we need more detailed status

Latest GitHub commits have preliminary support for this if you want to try it out (Ctrl+K to show script chain OSD).

[Funkmiser]

The way you've done it wouldn't have worked regardless of which version of MPDN. The extensions have always required its framework - you can't just put one file in there and expect it to magically work.

If you want to develop extensions, you need to start by downloading VS2015 (community edition is free).

I've tried that as well. I dragged the Extensions folder from the source zip to the MPDN folder and ran MPDN to make sure the extensions worked and showed up in the listbox, then dragged the KeyRemap.cs file from the Examples folder in the source zip to the PlayerExtensions folder and after running MPDN it still gave me the error.

[Zachs]

What was the detailed error message (you need to enable debug dialog)?

[Funkmiser]

What was the detailed error message (you need to enable debug dialog)?

I noticed that when getting this error message the Extension listbox in the options menu is completely blank.

TITLE: CSScriptLibrary Error
------------------------------

An unexpected error 'csscript.CompilerException' has occurred.

------------------------------
ADDITIONAL INFORMATION:

c:\Users\Cody\Desktop\MediaPlayerDotNet_x86_2_42_3_3406\Extensions\PlayerExtensions\KeyRemap.cs(24,18): error CS0535:
'Mpdn.Extensions.PlayerExtensions.KeyRemap' does not implement interface member 'Mpdn.IExtensionUi.Version'
(CSScriptLibrary)

------------------------------
BUTTONS:

&Ignore
&Abort
------------------------------


Also the update dialog for upgrading the extensions from v1.17.1 to v1.18.3 started popping up when testing without KeyRemap.cs even though I was using the 1.18.3 source zip.

EDIT:
Okay, so if I replace 'KeyRemap : IPlayerExtension' with 'KeyRemap : PlayerExtension'; add 'using Mpdn.Extensions.Framework;' to the top; and follow the dialog suggestions by adding the override keyword to 'ExtensionUiDescriptor Descriptor', 'Initialize()', 'Destroy()', 'IList<Verb> Verbs', 'ShowConfigDialog()' and adding the new keyword to 'HasConfigDialog()' it will actually not error when starting MPDN and KeyRemap will show up in the extension listbox. I tried the scrollwheel which according to the code should move the media position and that seems to work, but as I know nothing about coding the extensions it's possible that it is simply waiting to blow up in my face.

[Zachs]

Ah sorry. It turns out that the two player extension examples haven't actually been updated!
I'll update it when I get the time.

[aufkrawall]

Update checker is always offering me 2.42.4, although it is already installed.

[ryrynz]

Update checker is always offering me 2.42.4, although it is already installed.

Working fine here.

[Belphemur]

Update checker is always offering me 2.42.4, although it is already installed.

What version do you see when you go Help->About ?

What happens when you do Help->Check for Updates... ?

[aufkrawall]

it shows 2.42.4.3407 and check for updates exactly offers me this version.
Hm, maybe it didn't like that I temporary switched from binary extensions to GitHub scripts.

Gonna try a clean install.
Edit: Clean install fixed it.

[aufkrawall]

There's black flickering in windowed fullscreen mode when dragging the seekbar. It isn't there in normal windowed mode.
Edit: Also there in FSE.

[Zachs]

Not seeing it here...

[Anima123]

MPDN tend to crash when trying to play most of .wmv files. How can I help to help you with this issue?

BTW, the LAVFilter's settings are with no problem, this has already been figured out by me long time ago.

[aufkrawall]

Not seeing it here...
That unfortunately doesn't make it disappear here, though. ;)

Another report: MPDN fails to open png files that have % in their names. MPC HC opens these files with the same external LAV filters version.

[Zachs]

MPDN tend to crash when trying to play most of .wmv files. How can I help to help you with this issue?

BTW, the LAVFilter's settings are with no problem, this has already been figured out by me long time ago.

As usual: stack trace please.

That unfortunately doesn't make it disappear here, though. ;)

Another report: MPDN fails to open png files that have % in their names. MPC HC opens these files with the same external LAV filters version.

I'm still not seeing the flickering issue on any of my systems. Do you get this issue running a clean install on default settings or only with specific settings?

I'll fix the problem with the ampersand in filename.

[Anima123]

Only got this: The program '[796] MediaPlayerDotNet.exe' has exited with code -1073741819 (0xc0000005) 'Access violation'.

BTW, how can I get the full stack trace, I only got two options when crashed: Debug or Close the Program?

[Zachs]

Can you show active scalers (image & croma) in stats like MadVR so we can know what are active & what are not working although we thought it should work(may be for wrong config).
so many setting for this player.so we need more detailed status

Please consider these features for your next build

This is now available with the latest MPDN Extensions (toggle via Ctrl+K).

[Zachs]

Only got this: The program '[796] MediaPlayerDotNet.exe' has exited with code -1073741819 (0xc0000005) 'Access violation'.

BTW, how can I get the full stack trace, I only got two options when crashed: Debug or Close the Program?

Well you can't get a stack trace if the crash happens outside of .NET. However you should normally get the faulting module and address where the access violation has occurred. This is normally available under "EventViewer" unless you've disabled it.

[Anima123]

Here it is:
Faulting application name: MediaPlayerDotNet.exe, version: 2.42.4.3407, time stamp: 0x55f171bd
Faulting module name: AudioTransformFilter.ax, version: 0.0.0.0, time stamp: 0x55cc167a
Exception code: 0xc0000005
Fault offset: 0x0000000000002959
Faulting process id: 0x31c
Faulting application start time: 0x01d0eea18a68f0da
Faulting application path: C:\Program Files\MPDN\MediaPlayerDotNet.exe
Faulting module path: C:\Program Files\MPDN\Native\x64\AudioTransformFilter.ax
Report Id: eec95a44-5a94-11e5-8266-f81654b0da87
Faulting package full name:
Faulting package-relative application ID:

And from Windows Error Reporting:
Fault bucket 120319892798, type 4
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: MediaPlayerDotNet.exe
P2: 2.42.4.3407
P3: 55f171bd
P4: AudioTransformFilter.ax
P5: 0.0.0.0
P6: 55cc167a
P7: c0000005
P8: 0000000000002959
P9:
P10:

Attached files:
C:\Users\Lear\AppData\Local\Temp\WERCF2C.tmp.WERInternalMetadata.xml
C:\Users\Lear\AppData\Local\Temp\WERD6FD.tmp.appcompat.txt

These files may be available here:
C:\Users\Lear\AppData\Local\Microsoft\Windows\WER\ReportArchive\AppCrash_MediaPlayerDotNe_114050af8e1d1d9d3c7fc4a223b3598a745bba0_9d6426c3_064bd769

Analysis symbol:
Rechecking for solution: 0
Report Id: 45360978-5a98-11e5-8266-f81654b0da87
Report Status: 1
Hashed bucket: 1bb40f7d2e3e4018a6035fd691fdc583

[Zachs]

Does this happen with every wmv files or just some? I'd like to get hold of one to fix the problem.

[Anima123]

Just some of them crashes the player.

I will see if I can offer part of those files for debug purpose.

[Zachs]

That would be good because none of the wmv files I have can trigger the crash.

[Anima123]

I found the wmv files from http://download.wavetlan.com/SVV/Media/HTTP/http-wmv.htm will trigger the crash on my system.

Edit: Zachs, I just sent you a part of a problematic wmv file sample via PM.

[Zachs]

Thanks! I'll take a look.

[Dlget]

I got 60HZ display.
Can you recommend sane render time for 23,24 & 30fps content.(although most content i uses are 23fps)
so i'll try max setting without frame drop

Thanks for making Scale chain OSD painter extension.
I notices if i got to full-screen immediately from windowed mode after opening file,OSD is not shown correctly.I need to again go back to windowed mode then F11.

Edit:My display also support 23Hz mode.Should i use display mode changer?
coz at 23Hz, texts GUI of apps(OS,all apps) seems low quality.(is it because PC produces contents at higher FPS)
But video seems same quality.(?is it).

[ryrynz]

Can we get the scale chain OSD background color the same as the internal OSD?
I'd also like to see this OSD possibly painted below the CTRL internal OSD if it was showing.

Nice to see this come in, looks like there was a bit of work involved there.

[Zachs]

No I like it the way it is :) Also it is not possible to make it go under the player OSD.

[ryrynz]

I though consistent visual elements were a good thing. :P

[aufkrawall]

Nice new feature.
But it fails to get this particular comprehensive render script information:
http://abload.de/img/16asi8.png (http://abload.de/image.php?img=16asi8.png)
I guess there are either too many render scripts used for it at once or it has a problem with overlapping conditionals.

[Shiandow]

Nice new feature.
But it fails to get this particular comprehensive render script information:
http://abload.de/img/16asi8.png (http://abload.de/image.php?img=16asi8.png)
I guess there are either too many render scripts used for it at once or it has a problem with overlapping conditionals.

Neither of those seem likely, but what were you expecting it to display?

[aufkrawall]

A lot is happening before Spline downscaling, but it doesn't show it:
NNEDI3 OpenCL chroma
NNEDI3 OpenCL luma (+NNEDI3 chroma scaler)
image processor
Deband
super-xbr
image processor

[Shiandow]

A lot is happening before Spline downscaling, but it doesn't show it:
NNEDI3 OpenCL chroma
NNEDI3 OpenCL luma (+NNEDI3 chroma scaler)
image processor
Deband
super-xbr
image processor

Ah, I found the problem. The Conditional scripts weren't passing through the status correctly.

You might see the 'correct' status after changing the size back and forth.

[aufkrawall]

Yes, I noticed this. Just with (windowed) fullscreen you can't make use of this workaround. :)

[Zachs]

Another report: MPDN fails to open png files that have % in their names. MPC HC opens these files with the same external LAV filters version.

This one is firmly LAV Source Splitter's fault. I'll change MPDN so it no longer uses the source splitter - it'll just be LAV Splitter from now on.

[Zachs]

I though consistent visual elements were a good thing. :P

It's a debug OSD. I wouldn't worry too much about it. It gives you a bit more clarity with the slight colour difference when you have the OSDs stacked too.

[Zachs]

I found the wmv files from http://download.wavetlan.com/SVV/Media/HTTP/http-wmv.htm will trigger the crash on my system.

Edit: Zachs, I just sent you a part of a problematic wmv file sample via PM.

Fixed in the next release.

[Zachs]

I got 60HZ display.
Can you recommend sane render time for 23,24 & 30fps content.(although most content i uses are 23fps)
so i'll try max setting without frame drop

Thanks for making Scale chain OSD painter extension.
I notices if i got to full-screen immediately from windowed mode after opening file,OSD is not shown correctly.I need to again go back to windowed mode then F11.

Edit:My display also support 23Hz mode.Should i use display mode changer?
coz at 23Hz, texts GUI of apps(OS,all apps) seems low quality.(is it because PC produces contents at higher FPS)
But video seems same quality.(?is it).

I can't reproduce the problem. I tried going full screen immediately after opening a file but the OSD shows fine.

If 23Hz gives you low quality fonts etc, you'd expect it's the same for video so I'd probably avoid it if I were you. I suspect your display isn't really capable of 23Hz mode.

As for "sane" render time, the player OSD shows you in the form of "video interval". As long as you don't get too close to it you should be fine.

[aufkrawall]

This one is firmly LAV Source Splitter's fault. I'll change MPDN so it no longer uses the source splitter - it'll just be LAV Splitter from now on.
Sounds like an easy fix. :D
Btw: The issue with wrong colors for some pngs I reported some months ago is fixed.

[Dlget]

I can't reproduce the problem. I tried going full screen immediately after opening a file but the OSD shows fine.

Shiandow Already noticed that here (http://forum.doom9.org/showpost.php?p=1738587&postcount=3371)

As for "sane" render time, the player OSD shows you in the form of "video interval". As long as you don't get too close to it you should be fine.
Thanks.It seems that 41.71ms for 23fps ,41.67ms for 24fps.
So as long as rendering time is below 33ms i'm good.

Ohhh wait,i can even mention frame rate in condition inside script chain.Wow


As for my display it's a cheap one.AOC e2352Phz.
Hmmm.may be it doesn't support 23Hz.

[aufkrawall]

Something like this: http://pastebin.com/31jHxtHQ
Thanks. I removed height conditions from my script and I think I'll leave it this way.

[Zachs]

Shiandow Already noticed that here (http://forum.doom9.org/showpost.php?p=1738587&postcount=3371)


If it's the same bug you were reporting then it's been fixed in the latest source code on GitHub.

EDIT: This has made it to the latest release.

[Zachs]

I found the wmv files from http://download.wavetlan.com/SVV/Media/HTTP/http-wmv.htm will trigger the crash on my system.

Edit: Zachs, I just sent you a part of a problematic wmv file sample via PM.

Can you let me know if it's fixed on all of your problematic files?

[Zachs]

MPDN v2.43.0 brings about a new feature that allows player extensions to add support for custom source filters. I've added a Youtube source provider that uses the 3DYD Youtube Source Filter http://ys.3dyd.com/home/ to stream video from all (youtube, vimeo, etc.) the sources it supports.

You'll need to install the source filter seperately for it to work.

For maximum quality, you should set the 3DYD Youtube Source Filter to "Enable DASH Streams" and "Best Quality" for resolution.

[Anima123]

Can you let me know if it's fixed on all of your problematic files?

Just tried some with the latest version, it all renders normally. Thank you for fixing this issue so quickly.

[huhn]

MPDN v2.43.0 brings about a new feature that allows player extensions to add support for custom source filters. I've added a Youtube source provider that uses the 3DYD Youtube Source Filter http://ys.3dyd.com/home/ to stream video from all (youtube, vimeo, etc.) the sources it supports.

You'll need to install the source filter seperately for it to work.

Note however that during testing I found a problem with the latest version of the third party source filter - it hangs (not just MPDN) if you use the default settings. You need to set "Automatically download subtitles" to "Never".


why not livestreamer?

https://github.com/chrippa/livestreamer/tree/master

works with lavfilter so no new source filter needed.

[lanzorg]

why not livestreamer?

https://github.com/chrippa/livestreamer/tree/master

works with lavfilter so no new source filter needed.

And why not youtube-dl?
It supports a lot of streaming websites.
https://rg3.github.io/youtube-dl/supportedsites.html

[Zachs]

If they are compatible with lav filter already then there's no need to support it explicitly is there?

[huhn]

If they are compatible with lav filter already then there's no need to support it explicitly is there?

it works fine with mpc-hc but not your player.

[nevcairiel]

It needs to support forwarding stdin to LAV for these things to work. ie. mpc-hc just detects having "-" passed as a filename, and then just tells LAV to open "pipe:0", which is the stdin pipe.

[Zachs]

This has to be using lav source splitter though right?

[nevcairiel]

Obviously. :)
Need to tell it a filename, which only the source can.

[Zachs]

BTW is it possible to connect lav source splitter to lav splitter?

[nevcairiel]

No. LAV Source is not a file reader, its a full demuxer. Why would you ever want to do that anyway.

[Zachs]

Because MPDN always connects LAV splitter to the source filter.

[ryrynz]

You'll need to install the source filter seperately for it to work.

Note however that during testing I found a problem with the latest version of the third party source filter - it hangs (not just MPDN) if you use the default settings. You need to set "Automatically download subtitles" to "Never".

Do you know if the author aware of this issue? *Edit* NVM I alerted them it to them anyway.

BTW Is it possible to bundle this in with the extensions as well? If not perhaps have the required download info in the config for it.

What are your thoughts on detecting a file location or URL when launching File -> Open and pasting automatically if one is found?

[aufkrawall]

There's a bug regarding displaying the correct used chroma scaler in the render script OSD when using NNEDI3 image doubling: It always shows the selected chroma scaler in MPDN main application, even when it isn't used at all (checked it with nearest neighbor which would introduce terrible aliasing if it was actually used).

[Zachs]

BTW is the update checker working for any of you? It's not prompting me to update to this latest version for some reason.

[aufkrawall]

Yep, me neither.

[Zachs]

There's a bug regarding displaying the correct used chroma scaler in the render script OSD when using NNEDI3 image doubling: It always shows the selected chroma scaler in MPDN main application, even when it isn't used at all (checked it with nearest neighbor which would introduce terrible aliasing if it was actually used).

Uh not sure I understand what you're saying.

When I use NNEDI3 by itself...


Chroma: <MPDN chroma scaler>
NNedi3 ...
<MPDN luma scaler>


Is that not what you're seeing?

[ryrynz]

BTW is the update checker working for any of you? It's not prompting me to update to this latest version for some reason.

build 3420 is still listed as 2.43.0 here in the about, is that right?

Not getting 2.42.4 updater seeing 2.43.0 at all.

[aufkrawall]

Uh not sure I understand what you're saying.

When I use NNEDI3 by itself...


Chroma: <MPDN chroma scaler>
NNedi3 ...
<MPDN luma scaler>


Is that not what you're seeing?
I'm seeing this:
http://abload.de/img/1ktshx.jpg (http://abload.de/image.php?img=1ktshx.jpg)
But the displayed chroma scaler for NNEDI3 is incorrect, In the NNEDI3 render script NNEDI3 is also defined for scaling up chroma after NNEDI3 luma scaling. Nearest neighbor isn't used at all. It just displays it because it is selected in MPDN, but render script makes it impossible that it is used (and according to quality, nearest neighbor definitely isn't used).

[Zachs]

What did you select as the chroma scaler in the OclNNedi3 config dialog?

EDIT: NVM. I reproduced the problem here as well.

[Zachs]

build 3420 is still listed as 2.43.0 here in the about, is that right?

Did it prompt you to update when you were running 2.42.4 though?

[ryrynz]

Did it prompt you to update when you were running 2.42.4 though?

Nope. I was actually watching you upload it and installed it when it was done.

I have installed 2.42.4 and 2.42.3 with extensions 1.19 & 1.18 now to test and neither detect 2.43.0 I get "you have the latest release"

[Zachs]

Same here.

@Belphemur I have a feeling the MPDN Extensions API level check is preventing it from coming through?

[huhn]

What did you select as the chroma scaler in the OclNNedi3 config dialog?

EDIT: NVM. I reproduced the problem here as well.

is the problem from the fact that technically any file scaled with nnedi3 is first upsacel with NN and than treated as an interlaced frame and nnedi3 is used on it two times?

[aufkrawall]

It's most likely just a simple read-out bug.

[Belphemur]

Same here.

@Belphemur I have a feeling the MPDN Extensions API level check is preventing it from coming through?

Yep the logic applied when both Extension API changed in the same time ... was flawed. I corrected everything: https://github.com/zachsaw/MPDN_Extensions/pull/165

The problem was more deep than expected:
https://github.com/zachsaw/MPDN_Extensions/issues/164

[Dlget]

Offtopic:
Which things has significance for rendering time.

I'm using 2 video of same width,same fps.

one is 5500kbps bitrate with rendering time 35ms.
2nd one is 3000kbps bitrate with 46ms rendering time.

so mentioning bitrate,fps,width as condition is not enough to use particular script chain.(as 2nd video gave me dropped frame with same setting)
May be we need something more parameter in conditional chain.

EDIT:Can "Format profile High@L4.1 , High@L3.1 , Main@L2.1 etc" can significantly affect rendering time???

[aufkrawall]

You have to consider GPU stats like GPU usage and clock as well.
Driver doesn't always make the same decision which clock to use when there is low GPU load. If it chooses a lower clock, rendering times will go up.

[Zachs]

is the problem from the fact that technically any file scaled with nnedi3 is first upsacel with NN and than treated as an interlaced frame and nnedi3 is used on it two times?

Nah it's just a bug in the status reporting system.

[eX``]

Hi there =)

I'd like to report a crash I'm having regularly when trying to make comparison on fixed scenes. When I "CTRL+G" and enter a value, I have the following error message: Cannot seek with a negative value. Parameter name: usec. FYI, I also have it if I just click "GO" on the actual value, i.e. without actually writing any value but letting the one already called.

[Zachs]

I'll investigate. Cheers.

[ryrynz]

Received a reply regarding the Youtube Source.



I've tested MPDNx64 in several ways (including default settings) and I don't see any hangs. Though subtitles I also don't see. But according to log they definitely downloaded and downloaded without hang (for test used this clip with pt-BR subtitles). So it need more info to reproduce this issue.

All in all, need check more deeper. The way MPDN works (i.e. two graphs and thus two source filters within the same process) is not expected and thus not tested. Nice to know that it works.

Other (more important imho) thing relates to providing DASH streams. Highest quality on Youtube available only trough them and at current time, as far as I can check, they cannot be played in MPDN (at least video+audio, only audio is possible). DASH streams are disabled by default, to enable need check 'Quality -> Enable DASH streams' in filter properties, Options tab. Then at opening hold Shift key to make appear qualities dialog box and check out all available qualities (or select different variant in Quality -> Manual quality selection). My favorite example: DASH disabled (http://ys.3dyd.com/home/dash_disabled.png), DASH enabled (http://ys.3dyd.com/home/dash_enabled.png) (for this clip (https://www.youtube.com/watch?v=doZYHhF6YXI)). Of course usually there are less qualities but idea is the same.

In case of DASH streams each media stream (video and audio) provided by separated file. In this case filter creates two MEDIATYPE_Stream output pins, one for each file. And the fact that audio comes from second stream pin seems unexpected by MPDN. But anyway, this approach is not very good because each filter instance will download whole video+audio pair separately i.e. will be downloaded twice more data than needed. Btw, this also applies to playing regular qualities, MPDN uses two filter instances so the same file will be downloaded twice.

The better way would be to play video-only stream by one source filter and audio-only by other. But if manually select video-only stream for first filter instance at opening, usage of second filter will not be initiated because expected that it will give the same quality which does not contain audio. Also in current filter state, this cannot be done automatically: which quality to select currently can be configured only through filter properties i.e. it is the same for all filter instances. Need provide some configuration interface which then can be used by MPDN extension (YouTubeSourceProvider?). Need to discuss if this is interesting.

Of course, another variant, for which filter was originally projected, is to use one filter instance for both streams - no waste of data in this way, will be downloaded only data which will be used. But I don't know whether one source filter instance is allowed in MPDN. Also audio stream still will come from second pin which seems not supported at this moment.

My apologies if some of my assumptions about MPDN are wrong, I've meet it just now.

More info about pins. There can be media file data pins (MEDIATYPE_Stream) and subtitles pins (MEDIATYPE_Subtitle, MEDIASUBTYPE_UTF8, FORMAT_SubtitleInfo). Filter contains at least one media pin. In future can be more than two media pins (I want to add free streams selection in manual quality selection dialog box). More useful reason - this will be need for multi-angle (http://www.theverge.com/2015/2/4/7978427/youtube-experimenting-with-multiple-angle-videos) video clips if they will become widely used. Also I don't know why Youtube still haven't multiple audio tracks (e.g. for different languages), quite obvious idea and if it will be available, it also will need separated file streams for each track.

Subtitles are downloading at opening. Each subtitles track has its own pin. Max number of subtitles that I've ever seen is 58 (http://ys.3dyd.com/home/ss.png) (in this clip (http://www.youtube.com/watch?v=vfW5e6jVsMs)).

--
Regards, 3dyd

[ryrynz]

v2.43.1 Changelog:
Fixed URL drag-drop not working


I wondered why that wasn't working..I thought it hadn't been enabled. Thoughts on the auto copy paste?

Belphemur has fixed the updater, I get prompted to 2.43.1 now.

Zach could you have the installer check to see if the other filters are installed in the system and skip the installation of them?

[Zachs]

EDIT:Can "Format profile High@L4.1 , High@L3.1 , Main@L2.1 etc" can significantly affect rendering time???

With software decode it shouldn't affect render time as long as the input format and size are the same.

[Zachs]

It's most likely just a simple read-out bug.

Should be fixed with latest github. Can you check and see if there's anything else I've missed?

[edison]

how can I repeat play the current playing file?

[ryrynz]

how can I repeat play the current playing file?

Use the playlist. Select the repeat playlist icon (two arrows making a circle)
This will repeat everything in the playlist. If you have other files there you'll need to remove them.

[lanzorg]

I have strange issues with MPDN and the leshcat's amd/intel hybrid drivers. MPDN can't find my laptop amd graphic card, only the intel one. And if I set MPDN as an application that needs the powerful graphic card, MPDN uses the intel one (defined in the osd) but I have no video at all, black screen. In the MPDN's options there is only intel hd3000 gpu.

I tried with madvr and mpv, the amd dedicated gpu is used correctly, so I think it's not a driver's issue.

Is it a known issue? Is there a fix?

[aufkrawall]

Should be fixed with latest github. Can you check and see if there's anything else I've missed?
It shows it correctly now. :)
I think there's just one little glitch when opening a PNG image: The arrows of the OSD implicate that there is no chroma upscaling performed (which is correct), but it still shows the chroma scaler selected in MPDN next to it:
http://abload.de/img/1rgslf.png (http://abload.de/image.php?img=1rgslf.png)
I think it would be cleaner to show no chroma scaler if no chroma scaling is actually performed.

[Shiandow]

It shows it correctly now. :)
I think there's just one little glitch when opening a PNG image: The arrows of the OSD implicate that there is no chroma upscaling performed (which is correct), but it still shows the chroma scaler selected in MPDN next to it:
http://abload.de/img/1rgslf.png (http://abload.de/image.php?img=1rgslf.png)
I think it would be cleaner to show no chroma scaler if no chroma scaling is actually performed.

That's not a glitch. It tells you that while it isn't scaling the chroma, it is in fact still shifting the chroma (to correct the offset) using Sinc-Blackman.

[aufkrawall]

Oo, that's quite techie.
Maybe too techie, since chroma offset is usually not of interest?
I could imagine that there often will be questions about it because the main purpose of render script OSD is probably to just check if the render script is doing what the user has intended.

[Dlget]

<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="EWA Chroma Scaler" Guid="af9fde70-2a83-460b-89c0-8b7f2e54f1ba">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Six</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincHann" />
</Settings>
</Script>
</Preset>
<Preset Name="Deband" Guid="bf52b2f3-9452-4c70-9bbd-fea981c65c92">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Deband.DebandUi">
<Settings>
<maxbitdepth>10</maxbitdepth>
<power>0.75</power>
<grain>False</grain>
</Settings>
</Script>
</Preset>
<Preset Name="OpenCL NNEDI3" Guid="9a5e5b59-c324-437b-8ede-90a30fd11bd6">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.OclNNedi3Scaler">
<Settings>
<Neurons1>Neurons64</Neurons1>
<Neurons2>Neurons32</Neurons2>
<ChromaScalers>
<ChromaScalerPreset Name="None" Guid="f204908b-1256-4e68-93d6-568cec967020">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="Bilateral" Guid="8356bb60-57ba-4f29-a57d-1690313b7f29">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Bilateral.BilateralUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="ChromaScaler" Guid="db16d7d7-6de8-43d3-858d-1cbe9e60ae38">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Chroma.ChromaScaler">
<Settings>
<B>0.3333333</B>
<C>0.3333333</C>
<Preset>MitchellNetravali</Preset>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="EWA Chroma Scaler" Guid="9ed782fc-4c8d-4845-91ae-e266e818fdd8">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Four</TapCount>
<AntiRingingEnabled>True</AntiRingingEnabled>
<AntiRingingStrength>1</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.CustomLinearScalers.SincHann" />
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="NNEDI3 Chroma Doubler" Guid="61f1486d-51f7-4b80-8b03-f6558e9711a6">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.Chroma.NNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
<CodePath>ScalarMad</CodePath>
<Structured>False</Structured>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="OpenCL NNEDI3 Chroma Doubler" Guid="c3648a86-aba5-4281-844d-8bc11dc8dbbf">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.Chroma.OclNNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="SuperChromaRes" Guid="bc96da87-ff1e-4186-a39b-42762db557fd">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</ChromaScalerPreset>
</ChromaScalers>
<ChromaScalerGuid>d93e8c6f-1a4c-40d2-913a-3773c00d1541</ChromaScalerGuid>
</Settings>
</Script>
</Preset>
<Preset Name="Image Processor" Guid="56def538-6693-4e59-b69e-9e3e517d7b36">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.ImageProcessor.ImageProcessorScript">
<Settings>
<ShaderFileNames>
<String>AdaptiveSharpen\adaptive sharpen - 2015-08-19 - Pass one.hlsl</String>
<String>AdaptiveSharpen\adaptive sharpen - 2015-08-19 - Pass two.hlsl</String>
</ShaderFileNames>
<CompatibilityMode>True</CompatibilityMode>
</Settings>
</Script>
</Preset>
</ListOfPreset>

This is setting for 720p you gave in pastebin link.(i just modified tap count & neuron count for my use)
what does it do???
EWA croma scaler-->deband-->nnedi3 image doubler(with ewa as croma scaler)-->image processor --> & then required luma or croma scaler in option menu(not in scriptchain one)

what i understand from this- 1st croma is upscaled for target display size-->then image is doubled & then again croma is upscaled to match wih luma part-->image processor-->any scaler needed for making the video fit to target display size

why croma is upscaled twice?isn't image doubler is enough for requirement(as video width 1280 & display width 1920,image doubler would make it 2560 pixel & ewa croma scaler would make croma part near to it).at end required luma or croma scaler is used(downscaler)


Below one is setting for <720p
<ListOfPreset xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<Preset Name="OpenCL NNEDI3 Chroma Doubler" Guid="8d7e0110-9910-4f9c-95a7-aaef0292e552">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.Chroma.OclNNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons16</Neurons2>
</Settings>
</Script>
</Preset>
<Preset Name="OpenCL NNEDI3" Guid="6fc6523a-1ad2-4a50-a750-41d750e63a09">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.OclNNedi3Scaler">
<Settings>
<Neurons1>Neurons64</Neurons1>
<Neurons2>Neurons32</Neurons2>
<ChromaScalers>
<ChromaScalerPreset Name="None" Guid="c6c5e8a1-d85d-402a-b86f-d2c3ec4d2824">
<Script yaxlib:realtype="Mpdn.Extensions.Framework.RenderChain.RenderChainUi+IdentityRenderChainUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="Bilateral" Guid="1886fda1-1e2d-47fc-9278-29394a96d373">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Bilateral.BilateralUi">
<Settings />
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="ChromaScaler" Guid="7a869184-7175-41bc-ae78-2a3642c02094">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Chroma.ChromaScaler">
<Settings>
<B>0.3333333</B>
<C>0.3333333</C>
<Preset>MitchellNetravali</Preset>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="EWA Chroma Scaler" Guid="ffdb9baf-ed88-4eff-95fa-8052ffd3f2c4">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScalerChromaScalerUi">
<Settings>
<TapCount>Four</TapCount>
<AntiRingingEnabled>False</AntiRingingEnabled>
<AntiRingingStrength>0.85</AntiRingingStrength>
<Scaler yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.EwaScaler.EwaScaler+JincScaler" />
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="NNEDI3 Chroma Doubler" Guid="277eff2a-151e-4f4c-b389-12b47c500a2f">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.Chroma.NNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
<CodePath>ScalarMad</CodePath>
<Structured>False</Structured>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="OpenCL NNEDI3 Chroma Doubler" Guid="007f58aa-7326-49fb-80f3-9993e89233a0">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.Chroma.OclNNedi3ChromaScaler">
<Settings>
<Neurons1>Neurons32</Neurons1>
<Neurons2>Neurons16</Neurons2>
</Settings>
</Script>
</ChromaScalerPreset>
<ChromaScalerPreset Name="SuperChromaRes" Guid="a63240c1-4c6a-4abc-82bb-ce28f40de23b">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>1</Passes>
<Strength>1</Strength>
<Softness>0</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</ChromaScalerPreset>
</ChromaScalers>
<ChromaScalerGuid>2b4780b6-7aa7-478b-be30-36eb5a92d296</ChromaScalerGuid>
</Settings>
</Script>
</Preset>
</ListOfPreset>

in this 1st croma is doubled.Then luma is doubled along with croma.
why croma is doubled twice? isn't luma more imp part?



I could not make sense of these setting.In madVR,it just use doubler ,then any scaling required.

[aufkrawall]

When NNEDI3 for luma is used, chroma has to be scaled again.
If you don't choose chroma doubling in madVR, it automatically falls back to the main chroma scaler. In MPDN, you have do define this scaler by yourself (but have freedeom of choice for it, apart from NNEDI3 and main chroma scaler).

[Shiandow]

Oo, that's quite techie.
Maybe too techie, since chroma offset is usually not of interest?
I could imagine that there often will be questions about it because the main purpose of render script OSD is probably to just check if the render script is doing what the user has intended.

Maybe it's a bit technical, but it's not like it isn't doing anything. Exactly which scaler is used is important for both the image quality and the performance. Besides I'd rather the OSD displays something unimportant than not display something important.

[Zachs]

in this 1st croma is doubled.Then luma is doubled along with croma.
why croma is doubled twice? isn't luma more imp part?

I could not make sense of these setting.In madVR,it just use doubler ,then any scaling required.

The choice is yours. By default, chroma in NNEDI3 image doubler is set to "None", which means only Luma is doubled (chroma is scaled with MPDN's internal scalers). If you want to use NNEDI3 for chroma all the way, then it has to be doubled twice since chroma is half (in each direction) the size of the luma for 4:2:0 sources.

[aufkrawall]

Maybe it's a bit technical, but it's not like it isn't doing anything. Exactly which scaler is used is important for both the image quality and the performance. Besides I'd rather the OSD displays something unimportant than not display something important.
What about adding information that it's regarding chroma offset, and not actually upscaling? Or do you think this is already covered by the two arrows?

[Shiandow]

What about adding information that it's regarding chroma offset, and not actually upscaling? Or do you think this is already covered by the two arrows?

Well I don't mind changing the symbol, but ideally the symbol should be enough to tell you that the image is interpolated in some way, but without upscaling or downscaling. The statuses aren't really meant to explain why it's doing something, they're just there to tell you that it is. We're probably going to need to put together a short explanation of all the symbols at some point (especially for scalers, which have become quite complex).

[ryrynz]

The statuses aren't really meant to explain why it's doing something, they're just there to tell you that it is. We're probably going to need to put together a short explanation of all the symbols at some point (especially for scalers, which have become quite complex).

Why not just list the chain vertically?
Offset =
First chroma resize =
Luma resize =
Post processing =

Something like that.

[Zachs]

It already does.

[ryrynz]

Oh yup, got the wrong idea from the offset screenshot, I see now.

I notice this OSD sits not far alongside the player OSD before entering FSE mode and then shifts to it's default far right position, can that be avoided?

[aufkrawall]

Another topic: With conditional render scripts there can be done almost everything even today.
But I think it would be much easier to use if we had variables like "scaling factor". When people share render scripts they would also be more applicable in a universal way.

[Zachs]

That's not a glitch. It tells you that while it isn't scaling the chroma, it is in fact still shifting the chroma (to correct the offset) using Sinc-Blackman.

Actually it is a bug. Even when ChromaOffset is Zero (e.g. PNG file with RGB input), the chroma convolver still gets shown, which is incorrect - I've committed a fix on GitHub.

[Zachs]

Another topic: With conditional render scripts there can be done almost everything even today.
But I think it would be much easier to use if we had variables like "scaling factor". When people share render scripts they would also be more applicable in a universal way.

Good idea. Can you add a request on GitHub please?

[Zachs]

Hi guys,

There's a new test build available: http://mpdn.zachsaw.com/Test%20Builds/3440/

With the latest MPDN Extensions source from GitHub, you should now be able to use youtube-dl (and similar utilities) with MPDN.

The major change to allow this to work required MPDN to use just a single DirectShow graph. This change unfortunately also means seamless track changes for video / subtitles can no longer function (hence removed). OTOH XySubFilter's tray icon menus now work as expected.

This change also allows the YouTube source filter to download just the one stream. Subtitles are now also supported (3DYD will be releasing the fixed version soon that no longer hangs when subtitles are enabled).

It's a major change so I'd appreciate it if you guys could give it a proper test!

Cheers,
Zachs

EDIT: Found a problem "pipe:0" (youtube-dl) isn't working so don't test that yet.

[Zachs]

Latest test build 3442 (http://mpdn.zachsaw.com/Test%20Builds/3442/) should now properly support streaming via stdin.

Usage example:

youtube-dl -o - https://www.youtube.com/watch?v=C0DPdy98e4c | MediaPlayerDotNet.exe --stdin


Note: stdin is non-seekable so if you try to seek the stream will terminate. You can't restart the stream either so you'll need to close MPDN and start it via the command line arguments again. I'll tidy up the GUI to make sure seeking has no effect and stopping the stream is equivalent to closing a media file.

[Shiandow]

Hi,

Just thought I should mention that the latest MPDN extensions also have a new system for displaying the render script status. In principle the OSD should be the same (or equivalent) to the old system, so you probably won't see any difference, but please report any unexpected behaviour or bugs. Especially things like doubly nested scripts (e.g. using SuperRes in a SuperRes prescaler), and the Scripted Render Chain might still need some testing.

In theory the new system should be more resistant to bugs than the old system, though, so feel free to really try to break it. Of course illegible, but technically correct, statuses don't count as "breaking" it.

[Anima123]

Could developers please offer other builds together with AnyCPU version for Test?

[Zachs]

I'll upload other builds later, probably after I'm done with polishing up the UI for streaming.

[trandoanhung1991]

Is it possible to allow configuring Navigation extension? I'd like to be able to change right/left to jump x second instead of 5 key frames.

Currently to do this I'd have to use the sources and modify them myself, correct?

[Zachs]

Yeah unfortunately a user friendly key mapper is still missing. If anyone feels like contributing we'd really appreciate it.

[Zachs]

Could developers please offer other builds together with AnyCPU version for Test?

You can give test build 3445 (http://mpdn.zachsaw.com/Test%20Builds/3445/) a go.

[elraffa]

Hi Zachs..I would like to ask if you could incorporate an option to stretch hardcoded videos in MPDN.. I have a small monitor and films with black bands are too small for me.. In potplayer there is an option to do this (frame height +), like the image
http://i.imgur.com/Rn5fd5h.jpg

[Zachs]

I'll add that to my todo list.

[Zachs]

Just a heads up. A new test build (3448 (http://mpdn.zachsaw.com/Test%20Builds/3448/)) is now available which fixes a number of bugs.

[japa]

I get one or more extensions failed to load with this latest version, is this expected ?

[Shiandow]

There have been some changes to the extension API, so you need to combine the right version of the extensions with the right version of MPDN.

The only supported combinations are:

-The last released versions of MPDN with the last released version of the extensions.

-The latest test version of MPDN with the latest extensions on github.

All other combinations are deprecated, or won't work.

[japa]

Thanks, all sorted now. :D

BTW I like the @elraffa's suggestion I used that in MPC-HC and I miss it here.

[aufkrawall]

Good idea. Can you add a request on GitHub please?
Shiandow did it.
Thanks, with intervals it should be fairly easy to use. However, I'm too lazy atm to change my current script building. :o
I think considering the total amount of pixels of a source is the much better way than having to define every single bit manually regarding certain source resolutions (e.g. when a video with vertical format shows up, you may have to add a rewritten version of every existing scripts if they were just applicable for 16:9).

[aufkrawall]

Double post time, isn't it? :D

Well, I rebuilt my scripts with ScalingFactor conditionals and it took just five minutes and it works already like a charm. :devil:
I spent hours on my previous scripts which are based on source and target resolutions. :o

Since they work in a universal way, maybe bundle some presets?
e.g. installer could ask wich preset to use (they'd differ in required GPU power) and also makes some adjustments to the internal MPDN scalers based on the choice (e.g. Spline for downscaling etc.). Of course this would be totally optional.

Advantage would be that also users who don't want to understand conditional render scripts could get "best" quality, so it would be more attractive to a wider audience. Also, users could learn by the render scripts templates and e.g. start with simple modifications of them.

[Shiandow]

Should be possible with a player extension, feel free to add it as a feature request on github.

[Zachs]

Yeah it's actually one of my todo items. Since you've done so much IQ comparisons, perhaps you might be the best person to help out in this regard?
I'd start with just uploading the different presets (from low quality to max quality) to pastebin and linking them on the OP.
I'd also take into account of the framerate since it dictates how much render time per frame you actually get.
I'm always tied up with higher priority stuff so would appreciate it if you could help out.

[Zachs]

Should be possible with a player extension, feel free to add it as a feature request on github.

Ah I was thinking we could just use existing the copy and paste feature.

[Shiandow]

Ah I was thinking we could just use existing the copy and paste feature.

That's also possible, but a dialog would also be nice. Of course that dialog could use the copy/paste feature internally.

[Zachs]

That's also possible, but a dialog would also be nice. Of course that dialog could use the copy/paste feature internally.

Agreed.

[ryrynz]

Zach, can you duplicate the seekbar issue I mentioned? Drag the seekbar to the end when 'play next file in folder' is set and quite often
it'll just sit there stuck on the last decoded frame but in a playing state. Dragging the seekbar once again to the right from it's current end of seekbar position will force it to progress to the next file.

[Zachs]

I did try to reproduce the issue but haven't had any luck. Does this still happen with the latest test build?

[ryrynz]

Updated to latest build & extensions and it's working fine now. Is it possible to prevent a previously displayed frame flashing up when switching from FSE to windowed mode?
This occurs most noticeably when seeking whilst paused and then right clicking.

[aufkrawall]

Dunno if it's the driver or MPDN update, but flickering for me when seeking in windowed or exclusive fullscreen is gone.
I btw. really like how MPDN feels when seeking, very responsive.
Would it be possible now to also add time information of the cursor's spot in windowed fullscreen when moving the cursor onto the seekbar? I'm still really missing it from MPC HC. :)

[cascardian]

Hi, I'm one of those normally silent users, because I'm usually not experiencing any problems. I thought I'd inform you that with the latest release (2.44.0/1.21.0) I'm suddenly getting continually desyncing audio as the videos continue to play. Seeking corrects it for an instant and then it begins drifting again seemingly at the rate set in the rate tuner.

I have tried all sorts of videos and all sorts of settings in both MPDN and LAVFilters. It's definitely fixable by disabling the rate tuner. It seems that the API upgrade this version messed with the rate tuner somehow so that the audio is not kept in sync with the video.

Can anyone else please confirm this? I've reproduced it on a clean install, too.

Thank you for your hard work, guys!

[diimaan]

The Audio Script is completely gone missing in the (2.44.0/1.21.0) Build

https://copy.com/6YoDZifQ3O7pxJcc

And XySubFilter is not getting initialized and running even though it is selected and installed. :(

https://copy.com/e0wnQw27jwGLSEyO

https://copy.com/RoJ3klsjUDEMUqPa

https://copy.com/aRe10ah6Q8DolryM

Reverted back to the previous release as of now...

[aufkrawall]

In the madVR thread, there was a debate about whether or whether not to choose BT.2020 for 10/12 bit 4k video.
Result was that it shouldn't be used unless the stream is explicitly flagged accordingly, since basically every encode in the wild isn't BT.2020 and isn't flagged either. So MPDN should use Rec.709 for such streams, unless explicitly told otherwise by splitter.

[aufkrawall]

Ok, I created three simple quality presets which I think make sense for different GPU performance classes:
http://pastebin.com/RNLyLk9a

Settings for main MPDN:
low end:
chroma and luma up: bicubic 100 (MPDN's bicubic is very soft)
chroma and luma down: softcubic 50 or bicubic 50 (I think it's really just a matter of taste, I suppose most people rather like a sharp image)
(enhanced chroma reconstruction off, although this probably is redundant due to render scripts)

mid range:
chroma and luma up: Sinc-Hann8 AR
chroma and luma down: Spline4 AR

high end:
chroma and luma up: Sinc-Hann8 AR
chroma and luma down: Spline4 AR
(maybe also reduction of XySubFilter aliasing turned on)

I'm happy for hearing your opinions. :)

Some comments for some settings:
I set deband strength to 0.3 as a compromise (same goes for its grain).
I think edge strength of 0.3 for super-xbr makes sense to reduce aliasing and I find the line boosting rather unwanted as well. Sharpness of 0.75 is a compromise regarding this and ringing.
There is not a huge performance difference between different tap counts for EWA scalers and low tap count clearly has aliasing issues. I didn't choose Jinc for chroma for mid range and high end GPUs because it can clearly look too soft in cartoons. I find Sinc-Hann8 AR (max) as a good compromise between ringing and sharpness for good sources.

[Zachs]

Hi, I'm one of those normally silent users, because I'm usually not experiencing any problems. I thought I'd inform you that with the latest release (2.44.0/1.21.0) I'm suddenly getting continually desyncing audio as the videos continue to play. Seeking corrects it for an instant and then it begins drifting again seemingly at the rate set in the rate tuner.

I have tried all sorts of videos and all sorts of settings in both MPDN and LAVFilters. It's definitely fixable by disabling the rate tuner. It seems that the API upgrade this version messed with the rate tuner somehow so that the audio is not kept in sync with the video.

Can anyone else please confirm this? I've reproduced it on a clean install, too.

Thank you for your hard work, guys!

Yeah I noticed this myself last night. I think it might be worth going back to the previous release for now until I get this sorted. (BTW the reclock audio script is also affected by the same issue)

The Audio Script is completely gone missing in the (2.44.0/1.21.0) Build

https://copy.com/6YoDZifQ3O7pxJcc

And XySubFilter is not getting initialized and running even though it is selected and installed. :(

https://copy.com/e0wnQw27jwGLSEyO

https://copy.com/RoJ3klsjUDEMUqPa

https://copy.com/aRe10ah6Q8DolryM

Reverted back to the previous release as of now...

I noticed that too. XySubFilter doesn't seem to get loaded when there's no embedded subs. Oh Audio Script isn't gone -- just moved to the "Audio Renderer" category (look at the very bottom of your screenshot).

In the madVR thread, there was a debate about whether or whether not to choose BT.2020 for 10/12 bit 4k video.
Result was that it shouldn't be used unless the stream is explicitly flagged accordingly, since basically every encode in the wild isn't BT.2020 and isn't flagged either. So MPDN should use Rec.709 for such streams, unless explicitly told otherwise by splitter.

From what I've read, every new 10/12 bit 4k encode in the wild is BT.2020.

[Zachs]

Dunno if it's the driver or MPDN update, but flickering for me when seeking in windowed or exclusive fullscreen is gone.
I btw. really like how MPDN feels when seeking, very responsive.
Would it be possible now to also add time information of the cursor's spot in windowed fullscreen when moving the cursor onto the seekbar? I'm still really missing it from MPC HC. :)

Is the request on GitHub? If not could you add it? We'll implement it when we get to it.

[Zachs]

Ok, I created three simple quality presets which I think make sense for different GPU performance classes:
http://pastebin.com/RNLyLk9a

Settings for main MPDN:
low end:
chroma and luma up: bicubic 100 (MPDN's bicubic is very soft)
chroma and luma down: softcubic 50 or bicubic 50 (I think it's really just a matter of taste, I suppose most people rather like a sharp image)
(enhanced chroma reconstruction off, although this probably is redundant due to render scripts)

mid range:
chroma and luma up: Sinc-Hann8 AR
chroma and luma down: Spline4 AR

high end:
chroma and luma up: Sinc-Hann8 AR
chroma and luma down: Spline4 AR
(maybe also reduction of XySubFilter aliasing turned on)

I'm happy for hearing your opinions. :)

Some comments for some settings:
I set deband strength to 0.3 as a compromise (same goes for its grain).
I think edge strength of 0.3 for super-xbr makes sense to reduce aliasing and I find the line boosting rather unwanted as well. Sharpness of 0.75 is a compromise regarding this and ringing.
There is not a huge performance difference between different tap counts for EWA scalers and low tap count clearly has aliasing issues. I didn't choose Jinc for chroma for mid range and high end GPUs because it can clearly look too soft in cartoons. I find Sinc-Hann8 AR (max) as a good compromise between ringing and sharpness for good sources.

It would be nice if you could make a script group that has 3 (or more) presets (i.e. low medium high) and assign hotkeys to it so users could toggle between the 3 different settings. Having had a look at your preset I think that probably belongs in the "low/mid" group?

[aufkrawall]

wth, something went wrong with copy&paste.

Edit: It was meant to look like this:
http://pastebin.com/4rK474Br

[ryrynz]

Zach, in windowed mode the seekbar is quite responsive and basically refreshes instantly, can the seekbar in FSE mode have the same responsiveness?

[aufkrawall]

Is the request on GitHub? If not could you add it? We'll implement it when we get to it.
Alright, will do. Nice to see more things moving into render scripts, as they are open source.

[Zachs]

Zach, in windowed mode the seekbar is quite responsive and basically refreshes instantly, can the seekbar in FSE mode have the same responsiveness?

Fixed in the next release.

EDIT: Fixed in v2.44.1.

[Zachs]

wth, something went wrong with copy&paste.

Edit: It was meant to look like this:
http://pastebin.com/4rK474Br

Very nice, mate!
I'll add your post as a link in the OP.
Perhaps a few words to describe to newbies how to use the presets with screenshots?
e.g.
Copy and paste into Script Group.
Set hotkeys.
etc.

I'd imagine those who don't know much about render scripts won't know how to use the presets either, so a few lines of straight forward instructions + screenshot would probably do the trick.

[diimaan]

Oh Audio Script isn't gone -- just moved to the "Audio Renderer" category (look at the very bottom of your screenshot).


Oops my bad! Sorry about that! :)

[THX-UltraII]

does this player support frame-packed 3D Blu-Ray .ISO movies?

[Zachs]

Seeking corrects it for an instant and then it begins drifting again seemingly at the rate set in the rate tuner.

And XySubFilter is not getting initialized and running even though it is selected and installed. :(

Both issues have been fixed in v2.44.1.

[Zachs]

does this player support frame-packed 3D Blu-Ray .ISO movies?

Not sure. Haven't looked into it.

[Dlget]

wth, something went wrong with copy&paste.

Edit: It was meant to look like this:
http://pastebin.com/4rK474Br

This can be more tuned if you use fps in condition.
As for 23fps video rendering time should be less than 41ms & for 29 fps video rendering time should be less than 33ms.


I'll post my scrip later.
I liked that u used scaling factor instead of resolution.That makes things a lot easier.

I would like to hear about your opinion on scaling for anime related material

[aufkrawall]

Yep, it could be much more sophisticated. It would also make sense to additionally use source resolution information as well, e.g. to always force NNEDI3 chroma scaler for very low-res content, despite of the target resolution. But I wanted to keep it really simple and just discuss if people are ok with the chosen scaling settings. :)
(Now I think it might be better to choose super-xbr for second quadrupling pass for high end preset, as there is just little of a difference when doing NNEDI3 doubling before.)

Btw: Zachs, please check 2.44.1. It gives me lots of dropped frames, even without luma scaling. Queues are full, really weird. I reverted back to 2.44.0 and everything's back to normal.

[diimaan]

I get this when I try to use conditional script.

https://copy.com/GcNuBtpoZDUJXjTy

Are there any requisites to be installed? I am in Windows 10 64 bit pro. I have the requited dot net version installed. Only conditional script fails.

I can use script chain.

[huhn]

Not sure. Haven't looked into it.

can lavfilter even decode/handle MVC?

[Zachs]

I get this when I try to use conditional script.

https://copy.com/GcNuBtpoZDUJXjTy

Are there any requisites to be installed? I am in Windows 10 64 bit pro. I have the requited dot net version installed. Only conditional script fails.

I can use script chain.

It's likely something you've installed has removed ProcessDebugManager from the registry which is required by ClearScript (the library that enabled Conditionals). It is part of Windows and shouldn't be removed. If you do a clean Win10 install you won't encounter this problem (you can try doing a clean install in a VM just to see if Conditionals work).

[Zachs]

Btw: Zachs, please check 2.44.1. It gives me lots of dropped frames, even without luma scaling. Queues are full, really weird. I reverted back to 2.44.0 and everything's back to normal.

That's weird. Must've been something I changed. I'll check.

[diimaan]

It's likely something you've installed has removed ProcessDebugManager from the registry which is required by ClearScript (the library that enabled Conditionals). It is part of Windows and shouldn't be removed. If you do a clean Win10 install you won't encounter this problem (you can try doing a clean install in a VM just to see if Conditionals work).

Is it possible to install that component alone? Or dll?

[Zachs]

Is it possible to install that component alone? Or dll?

'pdm.dll' should be in your Windows\System32\F12 directory. You could try registering it with regsvr32 to see if it works.

[Zachs]

Btw: Zachs, please check 2.44.1. It gives me lots of dropped frames, even without luma scaling. Queues are full, really weird. I reverted back to 2.44.0 and everything's back to normal.

I can't seem to reproduce the issue here so I need more info.
Does this happen with all sources regardless of FPS?
Does it continually drop frames or does it do it in bunches?

[diimaan]

'pdm.dll' should be in your Windows\System32\F12 directory. You could try registering it with regsvr32 to see if it works.

WoW! it worked :):thanks:

[Zachs]

WoW! it worked :):thanks:

That's good but you might want to find out what third party software corrupted your registry (and probably other parts of Windows) to begin with.

[aufkrawall]

I can't seem to reproduce the issue here so I need more info.
Does this happen with all sources regardless of FPS?
Does it continually drop frames or does it do it in bunches?
Fluid Motion is the culprit. When playing e.g. 1080p 29,7fps video@75Hz with it enabled, it's constantly dropping frames.
Without FM (or with 2.44.0.), everything's fine.

[Zachs]

Fluid Motion is the culprit. When playing e.g. 1080p 29,7fps video@75Hz with it enabled, it's constantly dropping frames.
Without FM (or with 2.44.0.), everything's fine.

Ah found the problem. One of the bug fixes in 2.44.1 exposed another bug.

[ryrynz]

Zach if I select FSE whilst in paused windowed mode the seekbar disappears.
Also when I right click whilst in FSE mode the resolution and the frame changes giving me this.

http://i.imgur.com/x6vsjH8t.jpg

[Zachs]

Already fixed in the next release. I noticed it myself too yesterday just as I uploaded 2.44.1.

[trandoanhung1991]

This is weird. On my system, whenever I try to use NNEDI3 Chroma doubling, it causes a render script error. Both normal and OpenCL versions. Any idea how I can start to debug? Using GTX 970 with 355.98

[Zachs]

Well it would help if we know what the error message is.

[trandoanhung1991]

Well it would help if we know what the error message is.

That's the thing, I don't know. I just see "<!> Error in render script" when I enable Statistics OSD. I don't know where to dig in further.

[Zachs]

Did you use the installer or the source code from github?

[Zachs]

Hi guys,

If you follow our development on GitHub, you'd have noticed there's been quite a few recent changes to the Reclock audio script.

The new Reclock should work very well now, at least on the machines that I have tested on. If you guys could give it a proper test I'd appreciate it. What you'll find is your video frame rate (ref clock) will creep towards your display refresh rate until it eventually converges. As usual, Reclock will only be active when your source is within 3% of your display refresh rate.

Cheers.

[Shiandow]

That's the thing, I don't know. I just see "<!> Error in render script" when I enable Statistics OSD. I don't know where to dig in further.

I think that means that DirectX11 isn't available.

That said, I'm not sure why NNEDI3 still uses "Renderer.FallbackOccurred" and doesn't display the actual error message, might be a good idea to change that.

[Zachs]

Hmm fallback doesn't actually show up as "<!> Error in render script". This particular error message only comes up when an exception has occurred in the render script system.
@trandoanhung1991 You can try running dbgview to get hold of the detailed errors.

[Zachs]

Fluid Motion is the culprit. When playing e.g. 1080p 29,7fps video@75Hz with it enabled, it's constantly dropping frames.
Without FM (or with 2.44.0.), everything's fine.

Should be fixed in v2.44.2.

[trandoanhung1991]

Did you use the installer or the source code from github?

I think that means that DirectX11 isn't available.

That said, I'm not sure why NNEDI3 still uses "Renderer.FallbackOccurred" and doesn't display the actual error message, might be a good idea to change that.

This is from using OpenCL NNEDI3 Chroma Doubler. What's interesting is that it also seems to lock my GPU's clock speed to a low power state. Must be a bug with OpenCL and NVIDIA drivers?

[1708] System.NullReferenceException: Object reference not set to an instance of an object.
[1708] at Mpdn.Extensions.Framework.RenderChain.RenderChainScript.Render()
[1708] at Mpdn.D3D9VideoRenderer.VideoRenderer.A(Int64 , C , C )
[1708] at Mpdn.D3D9VideoRenderer.VideoRenderer.A(Int64 , IList`1 , IList`1 )


Normal NNEDI3 Chroma Doubler works normally, weird. I remember it having the same issues. Must be my memory playing tricks on me.

[aufkrawall]

Should be fixed in v2.44.2.
No apparent problems with this build/render scripts so far, nice.

What's interesting is that it also seems to lock my GPU's clock speed to a low power state. Must be a bug with OpenCL and NVIDIA drivers?

To what clock speed does it lock your GPU?
P2 is normal for OpenCL/Cuda, as it can't run stable with maximum VRAM clock.

[trandoanhung1991]

No apparent problems with this build/render scripts so far, nice.


To what clock speed does it lock your GPU?
P2 is normal for OpenCL/Cuda, as it can't run stable with maximum VRAM clock.

405/405. Can't unlock it unless I reboot.

[Shiandow]

405/405. Can't unlock it unless I reboot.

That sounds suspicious. At first I just thought it was using a lower power state because it wasn't doing anything, but then it should be able to return to a higher power state.

[aufkrawall]

Very weird, I have the samer driver and almost the same GPU (980) and OpenCL NNEDI3 is working fine here on Windows 10.

Hm, it even works here without the new interop API, even though D3D9 sharing extension isn't available.

Edit: Ok, two little issues to report:
Updater again offers me 2.44.2 although it is already installed. Deleted MPDN's config folder, didn't help.

When a PNG is opened in MPDN, it doesn't stop showing it when opening other media with it like video file.

[Zachs]

Hm, it even works here without the new interop API, even though D3D9 sharing extension isn't available.

It isn't possible to disable the new interop API -- the option says "Always use new OpenCL interop API". Disabling isn't the same as "Never use" - it simply allows MPDN to use the old interop API if it exists.

When a PNG is opened in MPDN, it doesn't stop showing it when opening other media with it like video file.

You mean the video file you're opening next no longer plays?

FWIW It worked fine when I tried opening a PNG file and then a video file.

[Dlget]

Feature Request:

Hotkey for script chain
We can save screenshot using PrintScreen & paste them in wordpad or anyother.
Why not save them directly in a directory.(with a name generated from script chain setting/active filters)

[ryrynz]

Getting consistently dropped frames when right clicking whilst in FSE mode while a video is playing.
Happens the worst with d3d 9ex, d3d 10 is better but still drops and d3d 11 has no problem.
HD 4000 64 bit w7 latest MPDN.

[aufkrawall]

It isn't possible to disable the new interop API -- the option says "Always use new OpenCL interop API". Disabling isn't the same as "Never use" - it simply allows MPDN to use the old interop API if it exists.

Ok, thx for information.


You mean the video file you're opening next no longer plays?

FWIW It worked fine when I tried opening a PNG file and then a video file.
Strange, it's totally reproducible here. I first drag&drop a 24bpp PNG into MPDN, which is rendered fine, and then I do the same with a video file. Then the player window changes to the size of the video, but the portrayed picture is still the previous PNG.

[aufkrawall]

That Fluid Motion bug is still not fully fixed with 2.44.2, I'm still getting constantly dropped frames with 60fps@75Hz.
Queues are full and without fluid motion, there are no dropped frames.

[Zachs]

Does it happen in fse mode as well?

[aufkrawall]

Yes, no difference.

[Dreamland]

MPDN v2.44.2

http://t6.pixhost.org/thumbs/3922/29606803_error.jpg (http://www.pixhost.org/show/3922/29606803_error.jpg)

:thanks:

[eX``]

Hi there ;)

I bet you're already aware of that, but Lav .66 is out (https://github.com/Nevcairiel/LAVFilters/releases)

[Zachs]

Strange, it's totally reproducible here. I first drag&drop a 24bpp PNG into MPDN, which is rendered fine, and then I do the same with a video file. Then the player window changes to the size of the video, but the portrayed picture is still the previous PNG.

That Fluid Motion bug is still not fully fixed with 2.44.2, I'm still getting constantly dropped frames with 60fps@75Hz.
Queues are full and without fluid motion, there are no dropped frames.

MPDN v2.44.2

http://t6.pixhost.org/thumbs/3922/29606803_error.jpg (http://www.pixhost.org/show/3922/29606803_error.jpg)

:thanks:

Can you guys try test build 3466 and let me know if it fixes all these problems?

Getting consistently dropped frames when right clicking whilst in FSE mode while a video is playing.
Happens the worst with d3d 9ex, d3d 10 is better but still drops and d3d 11 has no problem.
HD 4000 64 bit w7 latest MPDN.

The dropped frames only rack up for a while after the mode switch right? If so I wouldn't worry too much about it -- GPU takes a fair bit of time to do mode switch so dropped frames are unavoidable.

[Zachs]

Hi there ;)

I bet you're already aware of that, but Lav .66 is out (https://github.com/Nevcairiel/LAVFilters/releases)

Ah thanks. I'll update the installer.

[aufkrawall]

Unfortunately, still getting drops with FM. PNG problem is fixed.

[Zachs]

Unfortunately, still getting drops with FM. PNG problem is fixed.
I can't replicate the problem on any of my systems actually. Which version did this problem appear?

[aufkrawall]

With 2.44.1.

[Zachs]

Did you try fse mode with the latest version? What sort of dropped frames are you getting? One every few seconds?

[aufkrawall]

Yes, one or two drops every few seconds, FSE the same.

[Zachs]

The only thing that could've caused it to drop frame when render queue is full is the driver is reporting it is not ready to receive another frame in its presentation queue for a longer duration than allowed. When you have fluid motion enabled, it shortens this duration since it needs to present multiple blended frames in quick succession.

Hmm btw does it happen with every 60fps sources?

[aufkrawall]

I suppose yes, it also happens with 720p 60fps.
1080p 30fps works fine, at least I played a video for a few minutes and there were no frame drops.

[Zachs]

You might want to try different drivers to see if it helps because I'm just not seeing it here.

[aufkrawall]

Well, I rather stick to 2.44.0 for some time, as it mostly works without any issues here.

[Dreamland]

test build 3466:
Video Patagonia 8k (downloaded from you tube 3824x2160)

http://t6.pixhost.org/thumbs/3946/29633609_error1.jpg (http://www.pixhost.org/show/3946/29633609_error1.jpg)
Other players (example MPC-HC) it's ok

no other issue in this moment

anyway .. with MPDN have a great quality video!:p

[petoulachi]

Hello there,

I'm interested with this "new" player (currently using MPC HC + MadVR), but I only have a limited HTPC with an i5 and HD5000. Does MPDN can use both CPU and GPU (instead of MadVR that only use GPU) ?


Regards,

[Braum]

It's a little off track but I want to know your opinion on a 21:9 monitor I'm planning to buy : https://www.asus.com/Commercial-Monitors/PB298Q/
Does anyone here got a 21:9 monitor ? I would love to see movie in fullscreen as today most of them are in anamorphic :D

[eX``]

@Off topic: http://www.prad.de/new/monitore/test/2013/test-asus-pb298q-teil8.html#Fazit
I have no clue about 21:9 monitors, but you can find a lot of useful informations on tftcentral, prad, pcmonitors.info or while searching for reviews through the internet. Still, TFT and prad do very deep and exact reviews (color, black depth, Delta, ...) and sometimes even have ICC profiles.

[Braum]

@Off topic: http://www.prad.de/new/monitore/test/2013/test-asus-pb298q-teil8.html#Fazit
I have no clue about 21:9 monitors, but you can find a lot of useful informations on tftcentral, prad, pcmonitors.info or while searching for reviews through the internet. Still, TFT and prad do very deep and exact reviews (color, black depth, Delta, ...) and sometimes even have ICC profiles.

Thank you a lot ! :)

[burfadel]

test build 3466:
Video Patagonia 8k (downloaded from you tube 3824x2160)

http://t6.pixhost.org/thumbs/3946/29633609_error1.jpg (http://www.pixhost.org/show/3946/29633609_error1.jpg)
Other players (example MPC-HC) it's ok

no other issue in this moment

anyway .. with MPDN have a great quality video!:p

I'm getting the same, but with numerous different files. I did have one file that did work... but only when fluid motion was disabled. Actually going back to windowed mode and full-screen again, it was extremely jerky (like a massive number of dropped frames) disabling fluid motion did not help. Also, the same error message eventually showed when going back to Windows and full screen mode.

Current version mostly works fine, although I did notice that on some videos fluid motion looks shuddery.

[Zachs]

Well, I rather stick to 2.44.0 for some time, as it mostly works without any issues here.

I think I've found the problem. Will upload a test build later.

test build 3466:
Video Patagonia 8k (downloaded from you tube 3824x2160)

http://t6.pixhost.org/thumbs/3946/29633609_error1.jpg (http://www.pixhost.org/show/3946/29633609_error1.jpg)
Other players (example MPC-HC) it's ok

no other issue in this moment

anyway .. with MPDN have a great quality video!:p

FWIW the error says you're out of video memory.

Hello there,

I'm interested with this "new" player (currently using MPC HC + MadVR), but I only have a limited HTPC with an i5 and HD5000. Does MPDN can use both CPU and GPU (instead of MadVR that only use GPU) ?


Regards,

You can apportion part of the processing to AviSynth if you want to - LAV Splitter Source (which MPDN uses) supports .avs files.

I'm getting the same, but with numerous different files. I did have one file that did work... but only when fluid motion was disabled. Actually going back to windowed mode and full-screen again, it was extremely jerky (like a massive number of dropped frames) disabling fluid motion did not help. Also, the same error message eventually showed when going back to Windows and full screen mode.

Current version mostly works fine, although I did notice that on some videos fluid motion looks shuddery.

Yeah I found a couple of issues with FluidMotion with the recent releases (post 2.44.0). Will upload a test build later.

[burfadel]

Some of my media playing issues were driver related, with Catalyst 15.9 beta. Went back to my previous version and all was good again :). It might be the driver, it might be a combination of the driver + Insider Preview 10547.
+

[petoulachi]

You can apportion part of the processing to AviSynth if you want to - LAV Splitter Source (which MPDN uses) supports .avs files.
Thanks for your answer. I though that AviSynth was a not as good in a quality approach, but maybe I'm wrong ?

[Zachs]

Thanks for your answer. I though that AviSynth was a not as good in a quality approach, but maybe I'm wrong ?

It can be better in quality since there are some algorithms that not very suited to the GPU. However, it is almost always much slower, so unless you have a really fast CPU, it is usually not suitable for anything real time.

[Zachs]

Test build 3470 (http://mpdn.zachsaw.com/Test%20Builds/3470/) is now available.

@aufkrawall Can you try and see if this one works?

[aufkrawall]

Now I'm getting D3D presentation glitches instead of frame drops.
With D3D9, they occur much oftener than with D3D11 (was also the case with older builds and framedrops).

[Anima123]

There render error with wmv file playback now, with the latest release (don't know which version after). Would you developers please take a look?

[Zachs]

There render error with wmv file playback now, with the latest release (don't know which version after). Would you developers please take a look?

What's the error message? I just tried opening the various wmv files I have and they all work fine.

[Anima123]

There's pop up error windows says: Failed to render file 'xxxxx.wmv'.

[Zachs]

There's pop up error windows says: Failed to render file 'xxxxx.wmv'.

Every single wmv file?

[Anima123]

Every single wmv file?

Just files with video encoded as WMV2 (Windows Media Video 8) as mediainfo reported.

Edit: Just double-checked the LAV Video Decoder configurations, the wmv12 ticked.

[Zachs]

Just files with video encoded as WMV2 (Windows Media Video 8) as mediainfo reported.

Perhaps you installed the new LAV filters?

[Anima123]

Perhaps you installed the new LAV filters?

Yes, the latest 0.66 one, with which other players, like mpc-hc work well.

[Zachs]

Yes, the latest 0.66 one, with which other players, like mpc-hc work well.

I tested the wmv file you sent me via PM and that one worked fine. Can't tell if it's WMV2 though.

EDIT: Looks like all the wmv files I have are WMV3.

EDIT2: Is MPC-HC using LAV Source Splitter of the same version and not the internal one? Also make sure it isn't using any other source filters.

[Anima123]

I tested the wmv file you sent me via PM and that one worked fine. Can't tell if it's WMV2 though.

Just sent you a sample via pm.

[Zachs]

Just sent you a sample via pm.

Works fine.

[Anima123]

EDIT2: Is MPC-HC using LAV Source Splitter of the same version and not the internal one? Also make sure it isn't using any other source filters.

You're right, mpc-hc is using 'WM SF Reader' as splitter. I don't know how to force it to use LAVSplitter though.

[Zachs]

You're right, mpc-hc is using 'WM SF Reader' as splitter. I don't know how to force it to use LAVSplitter though.

Using the latest LAV Filters, your sample clip works fine for me with MPDN. Something is causing LAV Source Splitter to refuse decoding the file on your system.

[Anima123]

It's strange, only this particular file has this problem. Other same encoded files do not.

[Zachs]

It's strange, only this particular file has this problem. Other same encoded files do not.

Might want to ask in the LAV filters thread instead.

[burfadel]

It's strange, only this particular file has this problem. Other same encoded files do not.

Try out this post-0.66 build and see if it still happens:
http://files.1f0.de/lavf/nightly/LAVFilters-0.66.0-12.exe

[Shiandow]

Might also be a good idea to make sure that "wmv" is selected in the input formats tab of the LAV splitter settings.

[aufkrawall]

Good news, with build 3475 playback is fine again with FM.
Did you have to roll anything back, or did you find a better way? :)

Edit: What I've noticed: Is it intended that MPDN doesn't show the frame you jump to when moving the seeking slider, unless playback is resumed?

[Zachs]

Partially reverted. I did find the problem but the newer method wasn't as smooth as the old one so done parts were reverted.

[aufkrawall]

That build doesn't render PNGs or this video file:
http://www.mediafire.com/download/frvugpkdfteg01f/files.7z
I guess the problem with the video file might be that it contains too few frames (less than fps).

[Zachs]

That build doesn't render PNGs or this video file:
http://www.mediafire.com/download/frvugpkdfteg01f/files.7z
I guess the problem with the video file might be that it contains too few frames (less than fps).

Ah yes that one will be fixed for proper release.

[edison]

Could you tell me where the settings of MPDN saved? I want to reset the config of MPDN?

[Zachs]

Could you tell me where the settings of MPDN saved? I want to reset the config of MPDN?

%localappdata%\MediaPlayerDotNet

[cca]

I had a try of this player, decent enough for 23p/24p content but cannot get smooth playback with 29.97p content, it just stutters no matter what option I select for rendering. Refresh rate is 60Hz, tried with and without fluid motion, tried DirectX9 and 11, no difference. Very strange. Not using render script, just normal scaling options at default.

Sent from my Nexus 9 using Tapatalk

[Zachs]

I have a try of this player, decent enough for 23p/24p content but cannot get smooth playback with 29.97p content, it just stutters no matter what option I select for rendering. Refresh rate is 60Hz, tried with and without fluid motion, tried DirectX9 and 11, no difference. Very strange.

Sent from my Nexus 9 using Tapatalk

Are you sure FluidMotion is actually enabled? By default it is deactivated for 29p on 60Hz.

[cca]

Are you sure FluidMotion is actually enabled? By default it is deactivated for 29p on 60Hz.

I know, so I changed the option myself to try with it. The result is the same, the video has stuttering, playback is not smooth and the stutters very noticeable.

On the same PC, MPC+madvr do not have any issues, so I do not think it's my drivers or something PC specific.

[cca]

Ok, since I updated the AMD drivers 10 mins ago I also reinstalled MPDN x86 from scratch (removed config folders), now it seems to work, will keep poking to see what happens as I change options.

EDIT: Argh, the second time I try to play the same video, stutter begins. I do not understand why, I get incosistent behavior with MPDN here.

[Zachs]

I know, so I changed the option myself to try with it. The result is the same, the video has stuttering, playback is not smooth and the stutters very noticeable.

On the same PC, MPC+madvr do not have any issues, so I do not think it's my drivers or something PC specific.

2.44.1 and beyond has known FluidMotion bugs. This has been fixed in v2.44.3. Which version did you use?

[Zachs]

Ok, since I updated the AMD drivers 10 mins ago I also reinstalled MPDN x86 from scratch (removed config folders), now it seems to work, will keep poking to see what happens as I change options.

Oh yes -- if you look just one or two pages back you'll find another user with a problem with AMD drivers causing FluidMotion stuttering issue.

[cca]

Oh yes -- if you look just one or two pages back you'll find another user with a problem with AMD drivers causing FluidMotion stuttering issue.

I am now trying without fluid motion, 29.97 video on a 60Hz display. Still microstutters.

[cca]

Ok, this is what I have noticed so far: after seeking, the stutters start to occur. If I pause and resume, it starts to smooth out. For windowed mode, checking the box "Rely on DWM to handle vertical sync in windowed mode" makes the stutter completely dissapear. The problem remains in FSE mode though.

[cca]

OK, seem to have nailed what was causing this; LAV being in DXVA Copy Back mode. I switched to pure software mode, stuttering completely vanished! Very strange, but that seems to be the problem.

[Zachs]

That might be purely driver related then.

[cca]

Following up, unfortunately the stutter came back after a while. I tried other options too, I cannot get consistent behavior from MPDN for 29.97 videos. On the other hand, no problems with 24p either when played in 24Hz or in 60Hz with fluid motion.

[Zachs]

FWIW I just tested this on my NVIDIA and its just as smooth as 23p.

[aufkrawall]

29,7fps is fine here as well with NV.
cca, did you try with the latest v2.44.3? If yes, does the issue also occur with 2.44.0?

[Zachs]

Another user mentioned a couple of pages ago that the AMD beta driver is the culprit. I don't have a recent AMD GPU to test with but my old 4350 has no problem with 29p either. Then again it's running legacy drivers.

[cca]

I doubt it's the drivers since my alternate setup with mpc + madvr doesn't have this issue. I'm using version 2.44.3.3480 32bit of MPDN for my tests. I even started exploring possible timing issues so I followed the guide for the Rate Tuner extension but it still stutters. By the way the AMD beta drivers were update last night and they fixed the memory leaking bug. I am using Windows 10, it could affect things, but I have not been able to pinpoint what really causes those stutters, the OSD reports no drops, no repeats and no glitches, all stay at zero!

[aufkrawall]

How perceivable is the stuttering? Severe, or rather mini-stutter like?

[cca]

How perceivable is the stuttering? Severe, or rather mini-stutter like?

Mm, I would describe it as mini-stutter but it is rather perceivable, like it repeates a frame then continues, but it does it every 2-3 seconds. That's why I tried the Rate Tuner too, but did not help.

[aufkrawall]

You can try some juddertest videos:
http://www.oc-burner.de/ftp/Videos/juddertest/
Do you see it there as well, without OSD and other MPDN GUI elements hidden too?

[burfadel]

Mediainfo DLL updated to version 0.7.78

[phoenix369]

Hi guys.
First I would like to give congratulations to everybody involved in the development of this project. I have seen it evolve in very little time to the kind of player that I would use as my main daily player, and done so against some fierce competition like MPC+madVR.
The only thing that is keeping me from a complete switch its this problem I have in MPDN with oversaturated colors. Skin tones are orange or red and my dinamic range is suffering from this saturated and contrasted rendering of images with great loss in shadow details.
Im no expert but im guessing it could be related to calibration support, since with MPC+madVR saturation and contrast its way more precise because it supports selecting my screen calibrated gamut (BT.2020). If I select BT.709 in madVR the output, color wise, is almost the same as with MPDN, maybe a touch less saturated.
I havent found anywhere the typical saturation, contrast and color controls that could help alleviate this problem either.
I hope this can be solved eventually so I can make desired the full switch.
Also wanted to ask if there is a guide of any kind to understand which are the best options, quality wise, to create an script chain?
There are so many already that its difficult to be sure if Iam combining them the right way. Right now I have superChromaRes+Super xBR image scaling+Deband+LumaSharpen.
ThanKs a lot !!

[madshi]

my screen calibrated gamut (BT.2020).
OT: FWIW, today's displays usually can't even come close to BT.2020. It might make more sense to try to reach DCI-P3. That's at least within reach of some displays.

[aufkrawall]

As usual, MPDN Updater is offering me the version I already have installed as new.
There should be some additional checks implemented to prevent this.

[LaurentG]

Hi everyone (and sorry for my bad english),

I am a HTPC-addict and I've just installed MPDN (after having used MPC-HC, ffdshow/avisynth and MadVR for years).

Maybe using this forum to ask the following question is not "right". If it is the case, I apologize.

I'd like to know where to find samples of scripts and settings for MPDN. I can't find any tutorial when using Google and it's quite demanding for the newbies (like me) to understand how to use MPDN at its full potential.

The "buzz" surrounding MPDN is growing among people who want to setup or update their HTPC (thanks to you developers !), and I think (or want to believe) that I'm not alone to look for that kind of help.

Of course, every setting depends on the display, the CPU, the GPU, the video source (...) and the needs of each of us. But anyway, maybe you know if there is (or will be) a "project" which will be a start point to share the knowledge about the MPDN setup (the basics of "MPDN for the dummies" and later, maybe a kind of "how-to for advanced settings" ?).

Thanks for your help.

[burfadel]

The AMD Catalyst 15.9.1 driver was released shortly after the 15.9. They quickly fixed the bug present in 15.9, there could be other minor code changes as it runs brilliantly here compared to 15.9.

[aufkrawall]

I'd like to know where to find samples of scripts and settings for MPDN.
I made an example here:
http://pastebin.com/4rK474Br
Copy&paste it into script group and feel free to ask any questions.

Documentation is yet not very friendly for new users yet, but Zachs and Shiandow don't have infinite time and currently rather spent it on new code, which is reasonable.

[LaurentG]

@aufkrawall : :thanks:

[Zachs]

Mm, I would describe it as mini-stutter but it is rather perceivable, like it repeates a frame then continues, but it does it every 2-3 seconds. That's why I tried the Rate Tuner too, but did not help.

Did MPDN show "Fluid Motion active" in its OSD screen? And what's the max render & presentation times when the mini-stutter occurs?

[cca]

Did MPDN show "Fluid Motion active" in its OSD screen? And what's the max render & presentation times when the mini-stutter occurs?

Fluid motion is not active, absent from the OSD too. Render times is 2.55ms, Present 0,16ms, MAX Render 16,56ms MAX Present 0,28ms. Doubt that this is lack of horsepower from the card.

[Zachs]

Ah and refclk deviation was nowhere to be found too?

[cca]

Ah, the refclk deviation appears after a few seconds, I can get you a reading in a few minutes, I'm doing only basic testing until this is sorted, no render scripts, nothing extra enabled really. The numbers above are from windowed rendering, in full screen there a little higher but there's no difference in the stutter issue.

Sent from my Nexus 9 using Tapatalk

[cca]

Here's a screenshot of the stats:
https://i.gyazo.com/9071d90114f254b936305d32c411d283.png

[Zachs]

Did you enable fluid motion? It doesn't look like you have.

[cca]

No I didn't, should I? It's a 29.97 video on a 59Hz refresh, it shouldn't be needed, and when I tried to force it on it still stuttered.

[Shiandow]

No I didn't, should I? It's a 29.97 video on a 59Hz refresh, it shouldn't be needed, and when I tried to force it on it still stuttered.

In your case it's apparently a 29.709 video on a 59.93039 screen, which means that your screen has about half a frame more every second than the video, which causes an extra frame every 2 seconds. It should work better when you turn fluid-motion on though.

[cca]

I'll do some testing after work then, the weird part is that I tried to correct it with Rate Tuner but stutter persists.

[Zachs]

I'll do some testing after work then, the weird part is that I tried to correct it with Rate Tuner but stutter persists.

Can you give test build 3483 a try to see if Fluid Motion works better on your AMD GPU? This build should make it more consistent in terms of when a frame gets presented on an AMD GPU so it should help even in the case where Fluid Motion isn't enabled. As I've mentioned previously though I only have an old AMD GPU to test (it's actually ATI so that's how old it is) so YMMV. But do let me know if it helps at all.

[cca]

Okies, I'll try the new build soon.

Sent from my Nexus 9 using Tapatalk

[Zachs]

@aufkrawall

Could you also give the new test build a try and let me know if you get any D3D presentation glitches when FluidMotion is enabled?

Cheers.

[aufkrawall]

Yes, glitches with this build and FM. 2.44.3 is smoother.

[Zachs]

Yes, glitches with this build and FM. 2.44.3 is smoother.

There's definitely a problem with your NVidia drivers then. This test build simply reports glitches whenever a VBlank has happened but at no time during that new VSync cycle was the driver able to accept another Present() call. This means MPDN has to drop frame (in the test build) or keep trying until it's capable of accepting that call (in 2.44.3). In the latter case, it may be enough to mask the frame delay but there's still going to be a slight stutter.

I've only managed to replicate the driver bug under DWM mode, FSE mode works fine. Another thing you might want to try -- change your display resolution to 1080p or lower to see if it still happens. If you no longer get the glitches, it may be specifically to do a bug in the driver with high res monitors only.

[aufkrawall]

The issue also occurs with D3D9, FSE and 1080p display resolution.
I could try some drivers but I have the vague feeling that it won't go away.

[Zachs]

Here's the test build for the next extensions API (level 8) - http://mpdn.zachsaw.com/Test%20Builds/3521/.

This is to be used with the APILevel8 branch: https://github.com/zachsaw/MPDN_Extensions/tree/APILevel8

Main changes:
Preload the next file in playlist (eliminates the wait when loading youtube sources)
AudioScript API overhaul - Audio Chain removed for now and will make a come back before proper release.

Cheers.

[ryrynz]

Main changes:
Preload the next file in playlist (eliminates the wait when loading youtube sources)


Nice! No jarring playback lag at the start of the next file! Can the mouse cursor flash be prevented when the next file is being loaded at all? I guess windowed mode playback might be the only mode with the possibility of that happening.

Does this work alongside 'play next file in folder'? Because with that option enabled it's still a smooth start to the next file which TBH I wasn't expecting.

The lag of the current stable build happens when I start the playback of the second file in the playlist (off SSD) but not when there's only one file in the playlist and 'play next file in folder' is active.

[Garteal]

Nice, one Zachs as always. Going to give it a try this weekend.

[Anima123]

Unlike previous versions, from maybe 2.44.1，switch from window mode to full-screen mode, the video will flicker for several times, the last flicker almost paused the playback, no matter if the script chain is used or not.

Is it because of the on-screen algorithm display feature, and can it be optimized further more to avoid such flickers?

[Zachs]

Nice! No jarring playback lag at the start of the next file! Can the mouse cursor flash be prevented when the next file is being loaded at all? I guess windowed mode playback might be the only mode with the possibility of that happening.

Does this work alongside 'play next file in folder'? Because with that option enabled it's still a smooth start to the next file which TBH I wasn't expecting.

The lag of the current stable build happens when I start the playback of the second file in the playlist (off SSD) but not when there's only one file in the playlist and 'play next file in folder' is active.

Yeah I'm still trying to hunt down the cause of the mouse cursor appearing when transitioning from one file to another.

Current stable build doesn't have the preload mechanism. If there's a lag it's due to something else. I think the new preload implementation doesn't do anything for play next file in folder but I'm not too sure. I'll take a look when I get some time. It's more of a proof of concept at this stage. It should only preload the next item in your playlist.

Nice, one Zachs as always. Going to give it a try this weekend.

Could you take a look and see if there's anywhere else that might benefit from the new preload mechanism? Obviously as mentioned above "play next file in folder" could definitely use the same improvement. I'd restrict it to preload just one file in the background though as loading too many may cause excessive memory usage.

Unlike previous versions, from maybe 2.44.1，switch from window mode to full-screen mode, the video will flicker for several times, the last flicker almost paused the playback, no matter if the script chain is used or not.

Is it because of the on-screen algorithm display feature, and can it be optimized further more to avoid such flickers?

I don't think it's due to that. I personally haven't seen the problem though so I'm not sure what it could be. I had to "disable" Optimus recently on Win10 too since the latest drivers suffer the same problem I encountered on Win8.1, except this time it crashes Chrome instead of .NET apps. So I haven't been able to test the newer versions on Optimus systems (technically speaking my dev laptop is an Optimus system but Optimus technology isn't enabled - i.e. the NV GPU only drives an external display)

[aufkrawall]

I tried how fluid motion looks on Intel. The result is, it doesn't work there at all (Haswell, newest driver on Windows 10), frames are repeated.
madVR works. :confused:
And also with madVR on Intel I'm seeing some minor judder with smooth motion. However, this is really hard to judge since many displays aren't totally judder free.

Edit: 2.44.0 is still definitely smoother than build 3521 for me with Nvidia. Tested it with 72fps juddertest video@75Hz display.

[Anima123]

I don't think it's due to that. I personally haven't seen the problem though so I'm not sure what it could be. I had to "disable" Optimus recently on Win10 too since the latest drivers suffer the same problem I encountered on Win8.1, except this time it crashes Chrome instead of .NET apps. So I haven't been able to test the newer versions on Optimus systems (technically speaking my dev laptop is an Optimus system but Optimus technology isn't enabled - i.e. the NV GPU only drives an external display)

Now I realized it probably is related to the driver upgrade. You might want to look at the way MPDN entered FSE mode. As I tested, full-screen windows mode works fine.

[Zachs]

Going into FSE mode is very driver dependent. There is really nothing much an app has to do to go into FSE mode so not much could really go wrong. NVIDIA driver is really bad in this regard unfortunately. For example the NVIDIA has a known bug in D3D10/11 that goes into the wrong refresh rate when both AMD and Intel work fine when driving the same monitor. In fact the reports about it taking a long time to get into FSE mode is due to that bug. Only happens if you have a display that reports both 59hz and 60hz in its EDID though.

[cascardian]

Going into FSE mode is very driver dependent. There is really nothing much an app has to do to go into FSE mode so not much could really go wrong. NVIDIA driver is really bad in this regard unfortunately. For example the NVIDIA has a known bug in D3D10/11 that goes into the wrong refresh rate when both AMD and Intel work fine when driving the same monitor. In fact the reports about it taking a long time to get into FSE mode is due to that bug. Only happens if you have a display that reports both 59hz and 60hz in its EDID though.

I'm on 358.50 on a Fermi card (so WDDM 1.3 on Win10 for now) and it seems I'm affected by this with my Dell U2414H. That combination gets me a native HDMI connection only (no DP on the card, no DVI-D on the monitor) and NVIDIA is already annoying me by having to change output to Full Range every driver update, because it can't figure out this is not a TV usage case... which AMD apparently still can, despite the EDID reporting the HDTV 1080p parameters.

Anyway, the U2414H's EDID reports both 59 and 60 Hz. While in DWM mode, this seemingly translates to 60 Hz meaning true 60.00 Hz, while it changes to 59.94 in FSE mode. That leads to the following, measured within MPDN via CTRL+J (DWM VSync and new rendering path were enabled):

With 60 Hz selected in driver settings:
D3D10.1/11 in DWM: 60.00 Hz display output, 60.00 Hz composition rate.
D3D10.1/11 in FSE: 59.94, 59.94.
D3D9Ex in DWM: 60.00, 60.00.
D3D9Ex in FSE: 60.00, 59.94.

With 59 Hz selected:
D3D10.1/11 in DWM: 59.94, 59.94.
D3D10.1/11 in FSE: 59.94, 60.00.
D3D9Ex in DWM: 59.94, 59.94
D3D9Ex in FSE: 59.94, 59.94.

D3D9Ex applications in FSE mode are annoying (older games primarily), of course, because it disables f.lux.

Anyway, changing refresh rates on entering FSE mode actually takes long in both D3D9Ex and D3D10.1/11, but it might just be that this particular monitor is slow at switching, I don't know. If NVIDIA already knows about this, I wonder how long it'll take to fix it.

[nevcairiel]

FSE shouldn't really have a "composition rate" as the DWM compositor is not used in that scenario. Not sure what the second number would really represent then.

[Zachs]

I tried how fluid motion looks on Intel. The result is, it doesn't work there at all (Haswell, newest driver on Windows 10), frames are repeated.
madVR works. :confused:
And also with madVR on Intel I'm seeing some minor judder with smooth motion. However, this is really hard to judge since many displays aren't totally judder free.

Edit: 2.44.0 is still definitely smoother than build 3521 for me with Nvidia. Tested it with 72fps juddertest video@75Hz display.

P4600 on an i7 Haswell (admittedly on Win7) is one of my main development platforms so it definitely works. I also frequently test it on Broadwell (Atom) and HD3000 (both on Win10). In fact Intel gets the most attention since I have so many systems with integrated GPUs. That said, do you get the same problem with all the different source frame rates?

Anyway, changing refresh rates on entering FSE mode actually takes long in both D3D9Ex and D3D10.1/11, but it might just be that this particular monitor is slow at switching, I don't know. If NVIDIA already knows about this, I wonder how long it'll take to fix it.

NVIDIA aren't too interested in fixing bugs in the HTPC arena unfortunately. They aren't even interested in fixing their Optimus bug which crashes Chrome, Visual Studio etc. I took the time and effort to report the bugs, only to get their frontline customer service reps asking me to do nonsense tests. And when I did, they went silent (been a few months now).

FSE shouldn't really have a "composition rate" as the DWM compositor is not used in that scenario. Not sure what the second number would really represent then.

The display refresh rate reported by Windows.

[Zachs]

Can the mouse cursor flash be prevented when the next file is being loaded at all?

Should be fixed in 3525.

Edit: 2.44.0 is still definitely smoother than build 3521 for me with Nvidia. Tested it with 72fps juddertest video@75Hz display.

Build 3525 is now smoother on all my test systems vs 2.44.0. Can you try that and see if it's the same for you?

[ryrynz]

Should be fixed in 3525.


Slick. Not a mouse cursor blink or a window flash in sight.

[Garteal]

^ can you check if preloading with "Play next file in directory" works fine now if you grab the latest from my fork & playlist branch? If not, feel free to open an issue about it.

[aufkrawall]

Build 3525 is now smoother on all my test systems vs 2.44.0. Can you try that and see if it's the same for you?
OSD is logging lots of delayed frames as long as it is turned off.
Yes, when OSD is activated, it doesn't count such rendering problems.
But playback is not smooth, whether OSD is enabled or not.

[Zachs]

OSD is logging lots of delayed frames as long as it is turned off.
Yes, when OSD is activated, it doesn't count such rendering problems.
But playback is not smooth, whether OSD is enabled or not.

How are you testing it? On all of my 5 different test machines with 9 different GPUs, the latest build gives much smoother playback with fluid motion on.

@ryrynz How's fluid motion for you in this latest build?

[Shiandow]

By the way, it seems that script group hotkeys no longer work with the latest build. They'll change the settings, but I can't get MPDN to reload the renderscripts, so the settings won't be applied until you resize / play a new file.

[Zachs]

By the way, it seems that script group hotkeys no longer work with the latest build. They'll change the settings, but I can't get MPDN to reload the renderscripts, so the settings won't be applied until you resize / play a new file.

I'll investigate.

EDIT: Fixed. https://github.com/zachsaw/MPDN_Extensions/commit/1e2a874cd4081456b95e526386c8b5625c520b14

[Shiandow]

I'll investigate.

EDIT: Fixed. https://github.com/zachsaw/MPDN_Extensions/commit/1e2a874cd4081456b95e526386c8b5625c520b14

Thanks!

[Zachs]

^ can you check if preloading with "Play next file in directory" works fine now if you grab the latest from my fork & playlist branch? If not, feel free to open an issue about it.

I've made further changes and I think most (if not all) cases that we could preload files are now covered.

[ryrynz]

@ryrynz How's fluid motion for you in this latest build?

Will give it a shot when I'm home. Everything looked fine when I was checking out the file preloading.


I've made further changes and I think most (if not all) cases that we could preload files are now covered.

Cool, just had a look at the commits and saw Garteal's fix in there too, so I expect things will go smoothly. Are script conditionals still working properly now that we can preload?

[Zachs]

Cool, just had a look at the commits and saw Garteal's fix in there too, so I expect things will go smoothly. Are script conditionals still working properly now that we can preload?

The Conditional script should still work properly.

Oh and you'll need to test the latest commits with the latest MPDN test build 3526.

[ryrynz]

Oh and you'll need to test the latest commits with the latest MPDN test build 3526.

Build 3526 is giving me constant delayed frames on the Intel HD 4000 when Fluid Motion is enabled only when using D3D9. 2.44.3.3480 works fine.

[Zachs]

Build 3526 is giving me constant delayed frames on the Intel HD 4000 when Fluid Motion is enabled only when using D3D9. 2.44.3.3480 works fine.

Can you guys try Build 3528? Found and fixed a glaring mistake that was introduced in 2.44.1.

[ryrynz]

Can you guys try Build 3528? Found and fixed a glaring mistake that was introduced in 2.44.1.

All good here.

[aufkrawall]

The counter for delayed frames is unfortunately still increasing when debug OSD is closed. When it is opened again, it e.g. looks like this:
http://abload.de/thumb/15iu6a.jpg (http://abload.de/image.php?img=15iu6a.jpg)

[Zachs]

And there's no delayed frames when OSD is displayed? Did you test it in FSE mode?

[aufkrawall]

Yes (they are not reported, but I can see judder which is worse than just the monitor's judder.). In FSE, there is the same problem.

[Zachs]

Is fluidity as good as 2.44.0 though? You still get D3D glitches when playing video that has the same rate as your display right?

[Zachs]

Yes (they are not reported, but I can see judder which is worse than just the monitor's judder.). In FSE, there is the same problem.

Can you try D3D9 to see if the problem is the same?

[Anima123]

Unstopped delayed frames here with the latest test version, with both d3d9 and d3d11 with FSE. The playback is fluent to me though.

Edit: Wow, that's with Fluid Motion on. The delayed frames keep constant without Fluid Motion, at lease with d3d9.

Edit2: No delayed frames with d3d11 once Fluid Motion turned off.

It seems the delayed frames increase constantly with Fluid Motion on, unless OSD is on.

[Zachs]

Can you guys give build 3533 a try?

[ryrynz]

It seems the delayed frames increase constantly with Fluid Motion on, unless OSD is on.


Can you guys give build 3533 a try?

Can confirm this issue is fixed.

[aufkrawall]

Yes. I think now we got it as smooth again as with 2.44.0.
But I must say that I find madVR smooth motion a bit more fluid. There seems to be less erratic judder.

[Zachs]

Can you explain a bit further on what you mean by erratic judder? How often do you get this?

[Zachs]

New test build 3535 (http://mpdn.zachsaw.com/Test%20Builds/3535/) available with FluidMotion improvements. This particular one is the result of intensive testing on ATI 4350, NVIDIA 8400GS / NVS4200M / 560GTX, Intel HD3000 and HD4600 across 3 different machines and 6 different displays. There should be no more erratic jerks. Could you guys give it a go and report back please?

Cheers.

p.s. FSE mode only please - Windowed mode (DWM) is too dependent on what Windows does to be a reliable base for testing.

[aufkrawall]

Very nice work, now it's definitely smooth.
There's btw. no difference for me between FSE or windowed, even DX11 works absolutely glitch-free here.

Some questions:
What exactly does "improve chroma reconstruction" do?
I don't notice any difference whether it's off or on with render script chroma upscaling.

When we do chroma and luma upscaling via render script, main MPDN's scalers are still used for offset correction. Should bilinear be enough for this?

Unfortunately, I have a severe problem with render script level 8 API and my old script chain:
I'm just getting lots of errors with it with any file:
http://abload.de/thumb/mpdntdu7l.png (http://abload.de/image.php?img=mpdntdu7l.png)
My script:
http://pastebin.com/cpQMnqLC

[Shiandow]

SuperRes has been changed a bit to make it work better with the new API, unfortunately this change made old settings invalid. I think most other scripts somehow managed to keep working. So if you just remove all copies of SuperRes and replace them with a new one it should work.

[aufkrawall]

There must be another problem, I deleted everything except of a conditional that defines deband for native resolution and it still shows this:
http://abload.de/thumb/2fgsqp.png (http://abload.de/image.php?img=2fgsqp.png)

[Shiandow]

There must be another problem, I deleted everything except of a conditional that defines deband for native resolution and it still shows this:
http://abload.de/thumb/2fgsqp.png (http://abload.de/image.php?img=2fgsqp.png)

Seems it was not just SuperRes that didn't work then. I'll have a look what goes wrong, but there might not be an easy way to solve it.

[aufkrawall]

I was hoping the times to completely recreate the render script config would be over. :o
Not a big drama, but it may be very inconvenient for some users. :)

[Shiandow]

Not over yet, just more infrequent :(

One way to prevent it that would be worth a try is making the setting system use the default values when the config file doesn't specify one. Currently it just makes stuff up whenever it encounters an unspecified value, which is not too helpful.

Not sure what's going wrong for you though since I'm getting a somewhat different error message, might be something recent that's messing things up.

[DeadlyEmbrace]

Wow, so after ages the Remote Control finally got an update :)
The remote control now supports video tracks, opening of playlist files and URLs and the interface has finally gotten a long overdue overhaul (no more fugly! :o)

Please note that the Remote will not work with extensions v1.22.0 or below.

For the full set of release notes see https://github.com/DeadlyEmbrace/MPDN_RemoteControl/releases/tag/v1.3.0.67

Or grab the release https://github.com/DeadlyEmbrace/MPDN_RemoteControl/releases/download/v1.3.0.67/MPDNRemoteControl.zip

[burfadel]

I mentioned this in the Guru3D forums as I originally thought it was the display driver. A member there suggested I post here about it in case it's a compatibility issue with MPDN and Windows 10 TH2 build 10565+.

I have an issue that seems to be pretty random when playing MPDN in Windows 10 build 10565. It seems to cause a crash in quartz.dll (Directshow runtime) which in turn crashes the desktop explorer, which reloads quickly. If it happens to anyone else, it just looks like the taskbar is refreshing along with the icons (but it does show in event viewer). It has only happened since build 10565, which is why I believe it is Windows related. It is an Insider Preview, hence beta, so I wouldn't expect it to be perfect! I've already reported it through the Insider Preview feedback tool.

[aufkrawall]

Not over yet, just more infrequent :(

Ok, found out it's not related to my old script.
I created the deband example from above completely from the scratch and it fails as well when the conditional script is active (when ScalingFactor == 1).

[Shiandow]

As far as I can tell it was still (more or less) working at the time of my last commit, so it'd have to be a very recent bug. For now you could try using an older version of the extensions. And it would also help if you paste a copy of the minimal configuration that triggers the bug, either here or on the git repository.

[aufkrawall]

How can I get an old render script version from Github?
Simple script with error (open video file unscaled to activate conditional):
http://pastebin.com/9duy1paJ

[Shiandow]

How can I get an old render script version from Github?
Simple script with error (open video file unscaled to activate conditional):
http://pastebin.com/9duy1paJ

If you browse through the commits (https://github.com/zachsaw/MPDN_Extensions/commits/APILevel8), you can click on the "<>" button to browse the repository as it was at the time of that commit (including the commit itself). I think it was all still working at the time of my latest commit (although there have been some fixes and improvements since then) but if you want to you could reasonably quickly check at which point it stops working.

[aufkrawall]

With this commit it has been broken:
https://github.com/zachsaw/MPDN_Extensions/tree/08de818de103e1a4a15cc8430f048ad07e939fa5

[Shiandow]

Ah, yeah I think I see what went wrong, I knew something was bothering me about that commit. I'll just revert the commit for now, then it should work again. I fear the bug is a symptom of a more fundamental problem, though.

[Zachs]

Ah, yeah I think I see what went wrong, I knew something was bothering me about that commit. I'll just revert the commit for now, then it should work again. I fear the bug is a symptom of a more fundamental problem, though.

Looks like changes to Options by introducing ActiveOptions failed. Any idea why it's failing yet? I'll have a look into it anyhow.

[Shiandow]

Looks like changes to Options by introducing ActiveOptions failed. Any idea why it's failing yet? I'll have a look into it anyhow.

Probably because "Initialized()" was never called. It technically should be called whenever a RenderChain is used, but so far it's largely been ignored. Which was fine since it didn't really do anything, but then "ActiveOptions" was introduced, which was only ever intialized when "Intialized()" was called.

Edit: seems it's also initialized in the constructor, but only if "Options" isn't null, which it won't be because we're in the constructor.

[Zachs]

Probably because "Initialized()" was never called. It technically should be called whenever a Preset is used, but so far it's largely been ignored. Which was fine since it didn't really do anything, but then "ActiveOptions" was introduced, which was only ever intialized when "Intialized()" was called.

Edit: seems it's also initialized in the constructor, but only if "Options" isn't null, which it won't be because we're in the constructor.

Looks like we need a bit more discussion so let's continue on GitHub.

[Zachs]

Very nice work, now it's definitely smooth.
There's btw. no difference for me between FSE or windowed, even DX11 works absolutely glitch-free here.

Some questions:
What exactly does "improve chroma reconstruction" do?
I don't notice any difference whether it's off or on with render script chroma upscaling.

When we do chroma and luma upscaling via render script, main MPDN's scalers are still used for offset correction. Should bilinear be enough for this?

Unfortunately, I have a severe problem with render script level 8 API and my old script chain:
I'm just getting lots of errors with it with any file:
http://abload.de/thumb/mpdntdu7l.png (http://abload.de/image.php?img=mpdntdu7l.png)
My script:
http://pastebin.com/cpQMnqLC

Latest commit on GitHub fixes this issue.

[Zachs]

Wow, so after ages the Remote Control finally got an update :)
The remote control now supports video tracks, opening of playlist files and URLs and the interface has finally gotten a long overdue overhaul (no more fugly! :o)

Please note that the Remote will not work with extensions v1.22.0 or below.

For the full set of release notes see https://github.com/DeadlyEmbrace/MPDN_RemoteControl/releases/tag/v1.3.0.67

Or grab the release https://github.com/DeadlyEmbrace/MPDN_RemoteControl/releases/download/v1.3.0.67/MPDNRemoteControl.zip

Thank you so much!

The next MPDN release is going to be a massive step forward!

[Zachs]

I mentioned this in the Guru3D forums as I originally thought it was the display driver. A member there suggested I post here about it in case it's a compatibility issue with MPDN and Windows 10 TH2 build 10565+.

I have an issue that seems to be pretty random when playing MPDN in Windows 10 build 10565. It seems to cause a crash in quartz.dll (Directshow runtime) which in turn crashes the desktop explorer, which reloads quickly. If it happens to anyone else, it just looks like the taskbar is refreshing along with the icons (but it does show in event viewer). It has only happened since build 10565, which is why I believe it is Windows related. It is an Insider Preview, hence beta, so I wouldn't expect it to be perfect! I've already reported it through the Insider Preview feedback tool.

Thanks for the heads up!

[Zachs]

Very nice work, now it's definitely smooth.
There's btw. no difference for me between FSE or windowed, even DX11 works absolutely glitch-free here.

Some questions:
What exactly does "improve chroma reconstruction" do?
I don't notice any difference whether it's off or on with render script chroma upscaling.

When we do chroma and luma upscaling via render script, main MPDN's scalers are still used for offset correction. Should bilinear be enough for this?

On some systems windowed mode will cause a lot of DWM glitches so it's not a good base to test FluidMotion, which was why I asked for it to be tested in FSE mode only - doesn't mean it won't work in windowed mode.

"Improve chroma reconstruction" only works with MPDN's internal chroma scalers, so if you have render script chroma scaling, it has no effect.

What linear interpolation to use for offset correction really depends on how sensitive you are to chroma. I haven't tested this myself, so would appreciate it if you could give it a proper test to see if you could see any difference at all going to bicubic or higher. Since you know what to look for in chroma, you'd be a good candidate to give us some feedback.

[Shiandow]

"Improve chroma reconstruction" only works with MPDN's internal chroma scalers, so if you have render script chroma scaling, it has no effect.

That's not quite true, sure chroma scalers can override it, but the default behaviour is to use the same setting as MPDN's internal scalers.

[Zachs]

That's not quite true, sure chroma scalers can override it, but the default behaviour is to use the same setting as MPDN's internal scalers.

Oh yes that's right - Renderer.ConvertToRgb() has supported that for some time now... Totally forgot. :o

[aufkrawall]

Latest commit on GitHub fixes this issue.
Yes.
However, SuperRes is still broken. As soon as I select it (already renewed config, can access SuperRes' settings), I'm getting this:
http://abload.de/thumb/superresu3qxc.png (http://abload.de/image.php?img=superresu3qxc.png)

[Zachs]

We know what the problem is but it'll take some time to figure out how to do it right.

[Anima123]

Latest extension doesn't compatible with test version 3535. Is there another test version coming out soon?

[Zachs]

No. We are in the middle of a rather big change so don't use the latest for now.

EDIT: It should now work but keep in mind we're still working on it so it'll break from time to time.

[Anima123]

Thanks for the heads ups. It still brings red errors with no images when using with Fluid Motion right now, basically the texture is not the right type.

Edit: It also doesn't work if set passes of SuperRes to 8.

[Zachs]

Thanks for the heads ups. It still brings red errors with no images when using with Fluid Motion right now, basically the texture is not the right type.

Edit: It also doesn't work if set passes of SuperRes to 8.

Works fine here. You may need to clear your settings.

[Zachs]

MPDN v2.45 will be released soon so we'd appreciate it if you guys could give the latest APILevel8 branch a test to make sure there isn't any outstanding bugs.

Cheers.

[Anima123]

There's Error messages as follow:

One or more extensions failed to load!

The extensions you have installed may be incompatible with the current version of MPDN.

You are strongly advised to correct this problem before using MPDN.

I'm using the latest test build 64-bit and latest APILevel8 branch, settings just cleared before start MPDN.

Edit: The script version with no latest [Playlist] changes works. I will test it with my own settings to see if there's any problems.

[Zachs]

Yes the latest playlist changes will require a new test build and it's a work in progress.

[Anima123]

With the last render script Level8 version that worked with 3535, I got the following error messages when increased the pass of SuperRes to 8, SuperRes with NNEDI3 32x16 as image doulber, both steps with EWA ChromaScaler.

https://www.dropbox.com/s/4qdy7u6fx0h5odq/ErrorRenderScript.png?dl=0

Edit: Once the error appears, there's no way to get it back normal with simply revert the settings.

[Zachs]

Can you post your script settings so we could try and replicate the problem?

[Zachs]

Just a heads up. Test build 3550 is up and is compatible with the latest level8 commits on GitHub.

[aufkrawall]

Yes, 3551 + latest render scripts look good, it accepts my old script without issues (just have to reconfigure SuperRes settings, but no errors).

Just two little things:
-main MPDN scaling settings aren't saved on close
-MPDN hangs when closing if a video file was opened and a render script like deband is active

[Olivier C.]

I have encountered a bug when tried to use ffdshow raw filter + Avisynth for video processing :

MPDN :
http://img15.hostingpics.net/thumbs/mini_137513mpdn.jpg (http://www.hostingpics.net/viewer.php?id=137513mpdn.jpg)

MPC-HC :
http://img15.hostingpics.net/thumbs/mini_613863mpchc.jpg (http://www.hostingpics.net/viewer.php?id=613863mpchc.jpg)

MPDN settings :
Render script : None

Without avisynth, i.e. if i enable sharpen filter within ffdshow raw filter, i can not see any difference. For example, i have enabled the post-processing only on a half part and set sharpness to Max value. It seems that MPDN can not manage properly this post-processing filter, but maybe i am wrong in the way i use it ?

[lanzorg]

I have encountered a bug when tried to use ffdshow raw filter + Avisynth for video processing :

MPDN :
http://img15.hostingpics.net/thumbs/mini_137513mpdn.jpg (http://www.hostingpics.net/viewer.php?id=137513mpdn.jpg)

MPC-HC :
http://img15.hostingpics.net/thumbs/mini_613863mpchc.jpg (http://www.hostingpics.net/viewer.php?id=613863mpchc.jpg)

MPDN settings :
Render script : None

Without avisynth, i.e. if i enable sharpen filter within ffdshow raw filter, i can not see any difference. For example, i have enabled the post-processing only on a half part and set sharpness to Max value. It seems that MPDN can not manage properly this post-processing filter, but maybe i am wrong in the way i use it ?

Why are you still using these old things?

[Olivier C.]

Why are you still using these old things?

Because :
- i am using temporal denoisers or other stuff available only in ffdshow / avisynth.
- on first page, i can see this :
"Each chain can be customized by adding postprocessing filters (e.g. AC3Filter, ffdshow, SVP)"

But maybe i am wrong ...

[Zachs]

The biggest difference from your screenshots is both players aren't getting the same input. Take a look at the video resolution.

[Zachs]

Yes, 3551 + latest render scripts look good, it accepts my old script without issues (just have to reconfigure SuperRes settings, but no errors).

Just two little things:
-main MPDN scaling settings aren't saved on close
-MPDN hangs when closing if a video file was opened and a render script like deband is active

Found the problem. Will be fixed in the next build.

[Zachs]

With the last render script Level8 version that worked with 3535, I got the following error messages when increased the pass of SuperRes to 8, SuperRes with NNEDI3 32x16 as image doulber, both steps with EWA ChromaScaler.

https://www.dropbox.com/s/4qdy7u6fx0h5odq/ErrorRenderScript.png?dl=0

Edit: Once the error appears, there's no way to get it back normal with simply revert the settings.

Can you try the latest source on GitHub? It should now be fixed.

[Anima123]

Can you try the latest source on GitHub? It should now be fixed.

It seems fixed. Thank you developers.

[Zachs]

It seems fixed. Thank you developers.

Fantastic! Thank you for testing.

[Olivier C.]

The biggest difference from your screenshots is both players aren't getting the same input. Take a look at the video resolution.
You are right. The reason is that the avisynth script is supposed to denoise and upscale the movie.

[ryrynz]

Without avisynth, i.e. if i enable sharpen filter within ffdshow raw filter, i can not see any difference. For example, i have enabled the post-processing only on a half part and set sharpness to Max value. It seems that MPDN can not manage properly this post-processing filter, but maybe i am wrong in the way i use it ?

Zach I'm wondering if this is related to the image corruption I experience with Avisynth and NNEDI3.

[Zachs]

Not sure but what can be certain from the two screenshots is ffdshow isn't feeding the same video to MPDN.

[Zachs]

Hi guys

Test build 3555 is now available. Would appreciate if the testers could give it a proper test. We should be quite close to an official release.

Cheers.

[aufkrawall]

Seems to be fine.

Btw: I really don't notice any difference when choosing any algorithm for offset correction.
I scaled a problematic image from 720p to WQHD with SuperRes (NNEDI3 for chroma and luma) and there is basically no difference whether nearest neighbor or Spline is used for offset correction.
So I think it just should be set to bilinear when doing upscaling fully via render scripts. It's cheap and you are someway on the safe side, compared to NN.

[Zachs]

BTW could you post your updated script settings with the different performance presets?
I'm planning to release the new MPDN soon and would like to link to it in the OP. Something like your previous post would do just fine.

[aufkrawall]

Here they are:
http://pastebin.com/3XDMsaei
I made some small changes, e.g. replacing second NNEDI3 pass for highend GPU quadrupling with SuperRes super-xbr.

[Olivier C.]

The biggest difference from your screenshots is both players aren't getting the same input. Take a look at the video resolution.

It's the same video, the only difference between MPDN and MPCHC is that the video is denoised and upscaled on one side and doesn't processed at all on the other side.

But i understand your comment so i have reproduced it again and this time with the same resolution on OSD and with Mediainfo properties in order to convince you this time :) :

MPDN :
http://img15.hostingpics.net/thumbs/mini_799622mpdn.jpg (http://www.hostingpics.net/viewer.php?id=799622mpdn.jpg)

MPCHC :
http://img15.hostingpics.net/thumbs/mini_945502mpchc.jpg (http://www.hostingpics.net/viewer.php?id=945502mpchc.jpg)


Thanks Zachs for your awesome player

[Zachs]

Is this with the latest test build or the latest stable release? If it's not the test build could you give that a try because there was a related bug that was fixed some time ago but hasn't yet made it into the stable build.

EDIT: The test build I mentioned has now been released so once you've updated your current version, please test it again to see if the problem still occurs.

[Zachs]

Here they are:
http://pastebin.com/3XDMsaei
I made some small changes, e.g. replacing second NNEDI3 pass for highend GPU quadrupling with SuperRes super-xbr.

Thanks! Oh and MPDN internal scaler settings for each of the presets would be nice too.

[Zachs]

The following is a collection of render script presets that you can use if you're not familiar with the render script system, or you just wanted something to start off with. Simply copy and paste the text in the link and paste them into the Script Group renderscript.

By aufkrawall:


3 presets (for low / mid / high end GPUs) of scaling and postprocessing
http://pastebin.com/3XDMsaei

Please feel free to contribute your own presets and I'll add them to the collection.

[Olivier C.]

Is this with the latest test build or the latest stable release? If it's not the test build could you give that a try because there was a related bug that was fixed some time ago but hasn't yet made it into the stable build.

EDIT: The test build I mentioned has now been released so once you've updated your current version, please test it again to see if the problem still occurs.

Yes, it was with the latest build.

I checked with the fresh latest build (2.45.0.3555) and now it works like a charm ! Great job Zachs !

Maybe if i may say one little thing, i noticed that the input resolution is correct until i update the upscale factor while the video is playing. In this case, the input resolution is not refreshed (OSD, Ctrl+J).
I have to admit that it's a minor thing.

Thanks a lot Zachs (and others) for your (very) great job !

[Zachs]

Maybe if i may say one little thing, i noticed that the input resolution is correct until i update the upscale factor while the video is playing. In this case, the input resolution is not refreshed (OSD, Ctrl+J).
I have to admit that it's a minor thing.

Bear in mind that MPDN doesn't support input resolution changes in the middle of playback - you'll need to reopen the file.

[burfadel]

Under Windows 10 Insider Preview build 10565, along with the recent update that pushed the build decriptor to 10565.41737.amd64fre.th2_release_flt.151020-1357 (notice the 20 Oct stamp there), MPDN crashes when changing media files. This did not occur before the update. I tried resetting MPDN and using default settings, changing the decoder settings, colour space etc, nothing stops it from crashing when changing media files.

I tried MPC-HC 64-bit version and it changes media files without hassle, only MPDN is affected.

I also tried the latest nightly for MPC-HC 64-bit, tweaked all the settings, using the latest madVR with settings tweaked, custom shaders, and so on, and no matter what I tried I can not get MPC-HC to crash like MPDN does when changing media files (occurs even with a complete fresh install of MPDN). There must be something in the way that MPDN changes media files that something in the latest Windows 10 TH2 Insider Preview build with the aforementioned update that it does not like!

I suspect that maybe there is a long standing bug in MPDN that has only become apparent with a change in the way Windows does stuff, or a bug in Windows that affects the way MPDN does this, which is different to how MPC-HC etc works? In any case, reopening MPDN after each file is a little annoying!

The bug I mentioned earlier with quartz.dll, or what was causing quartz.dll to crash, has been fixed :).

[Anima123]

I don't use Fluid Motion a lot, but I just did a test with the latest version and it still has increasing number of delayed frames when OSD is off.

[ryrynz]

I don't use Fluid Motion a lot, but I just did a test with the latest version and it still has increasing number of delayed frames when OSD is off.

I'm not seeing that here, using build 3555 with either D3D9, 10.1 or 11 in windowed and exclusive modes on Intel HD graphics.

[Anima123]

I'm not seeing that here, using build 3555 with either D3D9, 10.1 or 11 in windowed and exclusive modes on Intel HD graphics.

Here's on nVidia 880M Optimus, the latest release builds with render script chain applied.

[ryrynz]

Here's on nVidia 880M Optimus, the latest release builds with render script chain applied.

Not seeing it on my 750 Ti either. It is all D3D versions and both windowed and FSE modes? If you disable the renderscript does it still occur?

[Zachs]

It's a problem with Optimus. There's not much that could be done because of the way the technology works. The old version never reported delayed frames when fluid motion is active. The new one does.

[Zachs]

Under Windows 10 Insider Preview build 10565, along with the recent update that pushed the build decriptor to 10565.41737.amd64fre.th2_release_flt.151020-1357 (notice the 20 Oct stamp there), MPDN crashes when changing media files. This did not occur before the update. I tried resetting MPDN and using default settings, changing the decoder settings, colour space etc, nothing stops it from crashing when changing media files.

Does this happen with the older build as well (i.e. API level 7)?

[aufkrawall]

Thanks! Oh and MPDN internal scaler settings for each of the presets would be nice too.
Settings for main MPDN:
low end GPU:
chroma and luma up: bilinear
chroma and luma down: bicubic sharpness 50
(enhanced chroma reconstruction off)

mid range:
chroma and luma up: bilinear
chroma and luma down: spline4 AR

high end:
chroma and luma up: bilinear
chroma and luma down: spline4 AR
(reduction of XySubFilter aliasing turned on)

Since there is no proof that a better scaler for offset correction really makes a difference, imho we shouldn't recommend to burn performance until we might know better one day.

[aufkrawall]

There might be still something a bit weird with fluid motion: OSD reports me delayed frames when it is closed while I'm playing 59,94fps@75Hz, e.g. this video:
http://www.oc-burner.de/ftp/Videos/juddertest/juddertest_59,94.mp4
I'm not sure if this is just a OSD bug or if it's really less smooth, I'm not used to 60Hz.

[Zachs]

Well delayed frames occur when your GPU had failed to present the frame in the vsync MPDN wanted. This can be due to various reasons external to MPDN. But if it's just delayed by one frame on a 75hz display, chances are you won't even notice it.

[christopherw]

Interesting player with some useful potential. Small footprint is a bonus.

I'm having unexpected results with interlaced content though - with identical LAV filter settings, using either DXVA or CUVID, it's not properly deinterlacing. Even with 50/60p enabled and attempting to force either just hardware or just Yadif software, and aggressively deinterlacing, everything is being blend deinterlaced.

The same video files, with MPC-HC and LAV configured identically, play perfectly.

I'm happy to communicate privately and send some example video files so you can attempt to replicate the issue if you wish, PM me through the board.

[burfadel]

Does this happen with the older build as well (i.e. API level 7)?

Yes, I tried the last release with the old extension, and when the media file changes it comes up with the crash dialogue that says MPDN - Media Player .NET has stopped working.

I might try a really ancient version and associated extension and see if something has changed over the versions.

[burfadel]

I found where the issue is introduced, it's the extensions version v1.22.0. I tried numerous MPDN's and extensions, and they all worked fine (they would not crash). I then went back to MPDN v2.44.3.3480 and tried the numerous supported extensions. The v1.21.x series works fine (including v1.21.4), I went previous/next about 60 times trying to get it to crash, perfectly stable! I then tried extension version v1.22.0 (also supported under the same MPDN), and it crashed just like the latest versions, without fail. I then tried extension version v1.21.4 again, and no issues!

So, the issue was definitely introduced in v1.22.0. I hope that makes it easier to track down what could be causing it!

[ryrynz]

So, the issue was definitely introduced in v1.22.0. I hope that makes it easier to track down what could be causing it!

Possibly unrelated but I had mpdn crash on Windows 7 last night when changing files. Not sure what caused it but Icaros crashed also.

[burfadel]

Possibly unrelated but I had mpdn crash on Windows 7 last night when changing files. Not sure what caused it but Icaros crashed also.

It could be? I don't use Icaros so I can't say about that. All I know is that anything before extensions v1.22.0 works perfectly, v1.22.0 or later it crashes when changing files consistently.

[Anima123]

It could be? I don't use Icaros so I can't say about that. All I know is that anything before extensions v1.22.0 works perfectly, v1.22.0 or later it crashes when changing files consistently.

I've experienced similar crashes too. Now I had to quit MPDN and restarted it playing another file in order to avoid the crash.

[burfadel]

I've experienced similar crashes too. Now I had to quit MPDN and restarted it playing another file in order to avoid the crash.

It's good to know that's it's not just me then! I'm sure it's something simple.

[Zachs]

Do you have preload next file enabled in your playlist settings?

If you do, try disabling that to isolate the problem.

[edison]

The program will crash if the speaker/headphone was unpluged.

[burfadel]

Do you have preload next file enabled in your playlist settings?

If you do, try disabling that to isolate the problem.

No luck here. I tried that and it still crashes as before.

[Zachs]

Can you try disable xysubfilter and try again?

[Anima123]

Crash still occurs after disabling subtitle feature.

[burfadel]

Same here, and I also have subtitles disabled by default and double checked to make sure.

[ryrynz]

I sometimes get visual corruption when loading the next file (pink or green lines throughout the image) briefly when preloading is active.'
I haven't been able to reliably reproduce the next file load crashing so far.

[Zachs]

Well I can't seem to be able to replicate the crashing problem. Are you guys using the LAV Filters included with MPDN?

[Anima123]

In recent days I twiddled parameters around SuperRes, since I am not satisfied with the 'artificial' kind effect NNEDI3 produced.

Now I am quite satisfied with Super-xBR as image doubler with Edge Strength 1.00 and Sharpness 0.35, and without 'third pass', the SuperRes configured with the default parameters. Though not perfect, but it gives me natural and vivid images watching a movie from a fare distance.

Tried with other image doublers, some, e.g. the NNEDI3 may better closed look, yet the overall visual pleasure it brings is not as good as with the settings I mentioned.

[Zachs]

So, the issue was definitely introduced in v1.22.0. I hope that makes it easier to track down what could be causing it!

This is strange. v1.22.0 was mainly about the playlist supporting URLs. When it crashes, do you get an error message?

[burfadel]

I was using the latest LAVFilters 0.66.0-31 which works perfectly with all other media players, and even all versions of MPDN with the extensions prior to v1.22.0. However, I just tried a fresh install of MPDN (clean install option) and extensions (same, clean install), installing the version of LAV Filters that came with MPDN. Same result, crash on changing media files.

Something between the extensions in v1.21.4 and v1.22.0 is causing the crash. Was the compile environment exactly the same for v1.21.4 and v1.22.0? Since it is such a specific fault (only with extensions v1.22.0 and later) and for me doesn't affect anything else, I still suspect the issue lies with one of the commits. It could either be the commit itself, or something in the commit that causes something in MPDN to go awry?

Is it possible for you to compile a few different test MPDN extensions omitting some of the different commits between v1.21.4 and v1.22.0 to see which commit is causing the issue?

Thanks

[Zachs]

Interesting player with some useful potential. Small footprint is a bonus.

I'm having unexpected results with interlaced content though - with identical LAV filter settings, using either DXVA or CUVID, it's not properly deinterlacing. Even with 50/60p enabled and attempting to force either just hardware or just Yadif software, and aggressively deinterlacing, everything is being blend deinterlaced.

The same video files, with MPC-HC and LAV configured identically, play perfectly.

I'm happy to communicate privately and send some example video files so you can attempt to replicate the issue if you wish, PM me through the board.

Could you cut a small section and post it somewhere so others could have a look at it too?

[burfadel]

This is strange. v1.22.0 was mainly about the playlist supporting URLs. When it crashes, do you get an error message?

No, just the standard crash error message that I was quick enough to capture :)

http://oi59.tinypic.com/168by92.jpg

and then:

http://oi57.tinypic.com/2ldkjkl.jpg

[Zachs]

Is it possible for you to compile a few different test MPDN extensions omitting some of the different commits between v1.21.4 and v1.22.0 to see which commit is causing the issue?

You can try grabbing the source code version from GitHub. You can get the incremental updates from https://github.com/zachsaw/MPDN_Extensions/compare/1.21.4...1.22.0

Can you try v1.22.0 (source version) and see if that fails as well? If it does, replace playlist*.* with the ones from v1.21.4.

[Zachs]

No, just the standard crash error message that I was quick enough to capture :)

http://oi59.tinypic.com/168by92.jpg

and then:

http://oi57.tinypic.com/2ldkjkl.jpg

You should be able to get the error message from EventViewer.

[Anima123]

I got this from EvenViewer:

Faulting application name: MediaPlayerDotNet.exe, version: 2.45.0.3555, time stamp: 0x5627415f
Faulting module name: VideoFrameServicesNative.dll, version: 0.0.0.0, time stamp: 0x55c2b55f
Exception code: 0xc0000005
Fault offset: 0x0000000000004370
Faulting process id: 0x14f4
Faulting application start time: 0x01d10f7964c41c3c
Faulting application path: C:\Program Files\MPDN\MediaPlayerDotNet.exe
Faulting module path: C:\Program Files\MPDN\Native\x64\VideoFrameServicesNative.dll
Report Id: b063df52-7b6c-11e5-8272-f81654b0da87
Faulting package full name:
Faulting package-relative application ID:

EventData as following:
MediaPlayerDotNet.exe
2.45.0.3555
5627415f
VideoFrameServicesNative.dll
0.0.0.0
55c2b55f
c0000005
0000000000004370
14f4
01d10f7964c41c3c
C:\Program Files\MPDN\MediaPlayerDotNet.exe
C:\Program Files\MPDN\Native\x64\VideoFrameServicesNative.dll
b063df52-7b6c-11e5-8272-f81654b0da87


Edit: And the .Net Runtime Error as following:
Application: MediaPlayerDotNet.exe
Framework Version: v4.0.30319
Description: The process was terminated due to an unhandled exception.
Exception Info: System.AccessViolationException
Stack:
at B.c.a(IntPtr, UInt32, UInt32, IntPtr, UInt32, IntPtr, UInt32, IntPtr, UInt32)
at B.c.a(IntPtr, UInt32, UInt32, IntPtr, UInt32, IntPtr, UInt32, IntPtr, UInt32)
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(IntPtr)
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(Int64, IntPtr, System.Collections.Generic.IList`1<IntPtr>, System.Collections.Generic.IList`1<System.Drawing.Size>, System.Collections.Generic.IList`1<SharpDX.RectangleF>, Boolean)
at B.K.c(B.I)
at B.K.f(Boolean)
at B.K.A(System.Action`1<Boolean>)
at B.K.s()
at MediaPlayerDotNet.MainForm.A(System.Drawing.Size)
at MediaPlayerDotNet.MainForm.A(Boolean, Boolean, Boolean)
at MediaPlayerDotNet.MainForm.OpenMedia(System.String, Boolean, Boolean, Boolean)
at MediaPlayerDotNet.MainForm.C(System.Object, System.EventArgs)
at System.Windows.Forms.ToolStripItem.RaiseEvent(System.Object, System.EventArgs)
at System.Windows.Forms.ToolStripMenuItem.OnClick(System.EventArgs)
at System.Windows.Forms.ToolStripItem.HandleClick(System.EventArgs)
at System.Windows.Forms.ToolStripItem.HandleMouseUp(System.Windows.Forms.MouseEventArgs)
at System.Windows.Forms.ToolStrip.OnMouseUp(System.Windows.Forms.MouseEventArgs)
at System.Windows.Forms.ToolStripDropDown.OnMouseUp(System.Windows.Forms.MouseEventArgs)
at System.Windows.Forms.Control.WmMouseUp(System.Windows.Forms.Message ByRef, System.Windows.Forms.MouseButtons, Int32)
at System.Windows.Forms.Control.WndProc(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.ToolStrip.WndProc(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.ToolStripDropDown.WndProc(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.NativeWindow.Callback(IntPtr, Int32, IntPtr, IntPtr)
at System.Windows.Forms.UnsafeNativeMethods.DispatchMessageW(MSG ByRef)
at System.Windows.Forms.UnsafeNativeMethods.DispatchMessageW(MSG ByRef)
at System.Windows.Forms.Application+ComponentManager.System.Windows.Forms.UnsafeNativeMethods.IMsoComponentManager.FPushMessageLoop(IntPtr, Int32, Int32)
at System.Windows.Forms.Application+ThreadContext.RunMessageLoopInner(Int32, System.Windows.Forms.ApplicationContext)
at System.Windows.Forms.Application+ThreadContext.RunMessageLoop(Int32, System.Windows.Forms.ApplicationContext)
at A.k.A(System.String[])

[Zachs]

Thank you!

Does this occur when you have preload next file disabled?

EDIT: Reproduced the error. Only happens when you have reset window size when closing media enabled.

EDIT2: Can you guys confirm if disabling the option fixes the issue?

[burfadel]

Thank you!

Does this occur when you have preload next file disabled?

EDIT: Reproduced the error. Only happens when you have reset window size when closing media enabled.

EDIT2: Can you guys confirm if disabling the option fixes the issue?

Yes, works fine with the reset windows size when closing disabled! Strange that it only occurred with extensions v1.22.0 and not v1.21.4 (under the previous MPDN version)?

The event view log for me shows basically the same as anima123:
Fault bucket 120342797506, type 4
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: MediaPlayerDotNet.exe
P2: 2.45.0.3555
P3: 5627415f
P4: VideoFrameServicesNative.dll
P5: 0.0.0.0
P6: 55c2b55f
P7: c0000005
P8: 0000000000004370
P9:
P10:

[Zachs]

Excellent!

Fixed in next release.

[burfadel]

Excellent!

Fixed in next release.

I also just found that enabling 'Reclock' under 'Audio Script' causes the same crash, the details are the same in the event viewer as the other crash.

[Zachs]

I also just found that enabling 'Reclock' under 'Audio Script' causes the same crash, the details are the same in the event viewer as the other crash.

With or without reset window size?

[Anima123]

Without 'reset window size closing media', I got these error messages when trying to playback a video file:
TITLE: SharpDX Error
------------------------------

An unexpected error 'SharpDX.SharpDXException' has occurred.

------------------------------
ADDITIONAL INFORMATION:

HRESULT: [0x80004005], Module: [General], ApiCode: [E_FAIL/Unspecified error], Message: Unspecified error
(SharpDX)

------------------------------
BUTTONS:

&Abort
------------------------------


And yes, I am using 'Reclock' together with Sanear Audio Renderer.

Edit: disable 'Reclock' does not make the error go away.
Edit2: reboot windows 8.1 seems to fix this error, you might want to ignore this post.

[burfadel]

With or without reset window size?

Without reset window size. It causes the exact same error details in the same file as with the reset window size option, so I guess if you fixed that it might have also fixed this issue?

The other thing I noticed with disabling the reset window size option is that manually changing media files seems to be very snappy, smooth, and seamless now. It could just be psychological, but I am sure that changing between files seems to be improved.

EDIT:
It seems that selecting any option under Audio Script causes the same crash (same details), whether it is reclock, DynamicRangeCompressor, Script Chain, or Script Group. This crash even occurs when nothing is enabled under script chain or script group, simply selecting it will cause the crash. I suspect though that if the error is fixed with the reset window size that these other issues may also hopefully be fixed (seeing as the crash is in the same file with the same crash details).

[Zachs]

Without reset window size. It causes the exact same error details in the same file as with the reset window size option, so I guess if you fixed that it might have also fixed this issue?

The same stack trace as well?

[burfadel]

Here are the crash results for each of the four audio render options:
Fault bucket , type 0
Event Name: APPCRASH
Response: Not available
Cab Id: 0

Problem signature:
P1: MediaPlayerDotNet.exe
P2: 2.45.0.3555
P3: 5627415f
P4: VideoFrameServicesNative.dll
P5: 0.0.0.0
P6: 55c2b55f
P7: c0000005
P8: 0000000000004370
P9:
P10:

I only pasted it once because all four crash reports are absolutely identical (well apart from the report ID further down, which makes sense since it's a different crash event).

and the associated .NET error (same for all of the crashes, include the reset window size option):

Application: MediaPlayerDotNet.exe
Framework Version: v4.0.30319
Description: The process was terminated due to an unhandled exception.
Exception Info: System.AccessViolationException
Stack:
at B.c.a(IntPtr, UInt32, UInt32, IntPtr, UInt32, IntPtr, UInt32, IntPtr, UInt32)
at B.c.a(IntPtr, UInt32, UInt32, IntPtr, UInt32, IntPtr, UInt32, IntPtr, UInt32)
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(IntPtr)
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(Int64, IntPtr, System.Collections.Generic.IList`1<IntPtr>, System.Collections.Generic.IList`1<System.Drawing.Size>, System.Collections.Generic.IList`1<SharpDX.RectangleF>, Boolean)
at B.K.c(B.I)
at B.K.f(Boolean)
at B.K.A(System.Action`1<Boolean>)
at B.K.s()
at MediaPlayerDotNet.MainForm.A(System.Drawing.Size)
at MediaPlayerDotNet.MainForm.A(Double)
at MediaPlayerDotNet.MainForm.A(Mpdn.IMedia, Boolean, Boolean, Boolean)
at MediaPlayerDotNet.MainForm.OpenMedia(Mpdn.IMedia, Boolean, Boolean, Boolean)
at Mpdn.Extensions.PlayerExtensions.Playlist.PlaylistForm.OpenMedia(Boolean)
at Mpdn.Extensions.Framework.PlayerExtension`2[[System.__Canon, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089],[System.__Canon, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089]].PlayerKeyDown(System.Object, Mpdn.PlayerControlEventArgs`1<System.Windows.Forms.KeyEventArgs>)
at System.EventHandler`1[[System.__Canon, mscorlib, Version=4.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089]].Invoke(System.Object, System.__Canon)
at Mpdn.PlayerControl.OnKeyDown(System.Windows.Forms.KeyEventArgs ByRef)
at MediaPlayerDotNet.MainForm.A(System.Object, System.Windows.Forms.KeyEventArgs)
at System.Windows.Forms.KeyEventHandler.Invoke(System.Object, System.Windows.Forms.KeyEventArgs)
at System.Windows.Forms.Control.OnKeyDown(System.Windows.Forms.KeyEventArgs)
at System.Windows.Forms.Control.ProcessKeyEventArgs(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.Form.ProcessKeyPreview(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.Control.ProcessKeyMessage(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.Control.WmKeyChar(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.Control.WndProc(System.Windows.Forms.Message ByRef)
at System.Windows.Forms.NativeWindow.Callback(IntPtr, Int32, IntPtr, IntPtr)
at System.Windows.Forms.UnsafeNativeMethods.DispatchMessageW(MSG ByRef)
at System.Windows.Forms.UnsafeNativeMethods.DispatchMessageW(MSG ByRef)
at System.Windows.Forms.Application+ComponentManager.System.Windows.Forms.UnsafeNativeMethods.IMsoComponentManager.FPushMessageLoop(IntPtr, Int32, Int32)
at System.Windows.Forms.Application+ThreadContext.RunMessageLoopInner(Int32, System.Windows.Forms.ApplicationContext)
at System.Windows.Forms.Application+ThreadContext.RunMessageLoop(Int32, System.Windows.Forms.ApplicationContext)
at A.k.A(System.String[])


I notice the line Mpdn.D3D9VideoRenderer.VideoRenderer.A(IntPtr), even though I have the presentation API set as Direct3D 11. When I press Ctrl+J for the stats when the video's are playing it does say Direct3D 11. Is that right?

[Zachs]

Ah disable "Resize window to fit when opening media" too and see if it still crashes.
If it doesn't then it's the same issue that I've fixed.

And yes D3D9VideoRenderer is really just because MPDN started out as just a D3D9 renderer and when I added D3D 10 and 11, I just left it as it is.

[burfadel]

The 'reset window size when closing media' causing the crash regardless of any other setting, and 'resize window to fit when opening media' causes the crash when any of the options is selected under the audio renderer, but is fine if 'none' is selected.

So I guess the issue is fixed with the next build, so all good!

[Zachs]

Should be fixed now in v2.45.1.

[burfadel]

Should be fixed now in v2.45.1.

Yes it seems to be! That's including the audio render options etc as well.

Thanks :)

[burfadel]

There is a new issue, however it seems to be related to the previous issue. It only occurs when 'Reset window size when closing media' is selected as well as an Audio renderer option. When you change from windowed mode to exclusive mode, more often than not when going back to exclusive mode the picture will stop, but the audio will continue. You just need to move the progress slider and it seems to correct itself. Well, apart from the one time when I was testing it when the whole program froze, but the audio continued! I had to use the task manager to close it. Sorry, no event viewer information this time as the program didn't actually crash. Just a video playback freeze.

Only an issue with that option and any audio renderer option selected (I use the reclock option). If either are selected on their own there doesn't seem to be an issue.

[djsolidsnake86]

image quality is better thant using mpc hd with madvr?

[Dreamland]

I think MPDN is the best player for quality video ! great!

[djsolidsnake86]

I think MPDN is the best player for quality video ! great!

doesn't support madvr
what is the output?

[Dreamland]

Madvr + MPC-HC is very good..but has not the same quality video of MPDN
my opinion watching in my Eizo monitor
(excuse me for my bad english ) I.m italian

[djsolidsnake86]

Madvr + MPC-HC is a very good..but has not the same quality video of MPDN
my opinion
(excuse me for my bad english ) I.m italian

come fai a dirlo?
how do you know that

[Dreamland]

mettendo a confronto un video con Madvr + MPC-HC e lo stesso video con MPDN
piu contrasto, colori piu vivi,l'immagine sembra anche piu pulita e fluida..
tanti settaggi con Madvr ma MPDN è uscita sempre vincente
(sempre secondo me)
se vuoi sapere i miei settaggi , te li dico senza problemi

[burfadel]

Madvr + MPC-HC is very good..but has not the same quality video of MPDN
my opinion watching in my Eizo monitor
(excuse me for my bad english ) I.m italian

how do you know that

I'm sure he is well aware oh his own opinion of things :). Dreamland, what you said makes perfect sense, and I agree!

@Zachs
That new issue I mentioned a couple of posts back, I can't seem to replicate it now! It was doing it without fail last night, but now I just can't seem to get it to do it with the same settings. I haven't restarted my computer or done anything else in the meantime, so I'm not sure what's going on there.

If anyone at all has the issue of the video pausing but audio continuing when going from windowed mode to exclusive mode (only happened when certain settings were selected) please report it. Issues will only be fixed if they are known!

[burfadel]

Okay, seems to be a bit temperamental. If I go to windows mode, do something on the computer like write this out in Chrome), then go back to the player and put it in exclusive mode, it may have paused video playback with the sound continuing, but it's not consistent. However, it doesn't seem to do it at all without the two settings I mentioned earlier selected. It's a difficult one to reproduce! I'm sure the first time it happened that I didn't have Chrome or anything open...

[Anima123]

burfadel, would you please try rebooting OS and see if this problem is still there? Sometimes rebooting can solve some kind of odd ones.

[Zachs]

Does it happen if you enter FSE mode with alt enter?

[ryrynz]

Encountered a few situations where switching between windowed and FSE modes results in the picture freezing and audio continuing. I've had it when going either way.
When performing a double click again to fix it the video then stutters quite badly for a few seconds before playing back as normal.
New windowed mode, D3D9Ex, HD Graphics 4000. Thought I'd mention it here before doing any further investigation into D3D modes, different systems etc.

I've also experienced a second or so pause of the video occasionally after seeking whilst using the arrow keys, the audio has no delay.

Anyone else experienced either of these?

[burfadel]

Encountered a few situations where switching between windowed and FSE modes results in the picture freezing and audio continuing. I've had it when going either way.
When performing a double click again to fix it the video then stutters quite badly for a few seconds before playing back as normal.
New windowed mode, D3D9Ex, HD Graphics 4000. Thought I'd mention it here before doing any further investigation into D3D modes, different systems etc.

I've also experienced a second or so pause of the video occasionally after seeking whilst using the arrow keys, the audio has no delay.

Anyone else experienced either of these?

Yes, me! The first one at least. I'm using Direct3D 11, R9-280X, exclusive mode etc, but the symptoms sound identical. It seems to be a bit hard to reproduce though, it doesn't happen all the time.

Under options --> general, do you have either of the resize options selected, if so, which?

Also, do you have any option at all selected under 'Audio Renderer' --> 'Audio Script'?

If so to any of the above, set audio script to 'none' and see if the behaviour continues. If so, try deselecting the resize options and see if the problem is resolved. The reason for this is a crash bug was recently fixed that affected these options, but I believe there is still some residual related 'buggyness'...

Of course, if your trials regarding these settings prove fruitful, report back which one was the offending options for you. Unfortunately for me it only occurs sometimes, which makes testing difficult. It is an annoying bug though!

[ryrynz]

Under options --> general, do you have either of the resize options selected, if so, which?


They are both selected right now, because I deleted MPDN's config for testing, so they're at defaults. I don't usually keep those enabled so I think it's unlikely they're related as I've experienced this before.. but we'll see.


Also, do you have any option at all selected under 'Audio Renderer' --> 'Audio Script'?


Nope.


The reason for this is a crash bug was recently fixed that affected these options, but I believe there is still some residual related 'buggyness'...


I'm aware of it.


Of course, if your trials regarding these settings prove fruitful, report back which one was the offending options for you.

Of course.

I'm replicating the seek freeze consistently but it does require a number of attempts for it to happen. It occurs in both Windowed mode and Exclusive.
What happens is when seeking the image freezes at a particular point but the audio has seeked further back. When the audio catches up to the point where the image is at, then MPDN continues playing like normal.
I have disabled fluidmotion, it's not related to that or the resize options.

*Both issues occur in 3555 as well*

Also had the FSE freeze as well, both resize options are unticked at this point, no scripts are active. Seems to happen quite often after the video is paused in windowed mode and then playback is resumed in FSE.
I'm getting both these issues consistently, they shouldn't be too hard for Zach to find.

Hey Zach, when Lock to video aspect ratio is enabled and I stretch the right side of the frame the window keeps it's left side position steady. If I stretch the left side the window jumps out to wherever I stretched the left side out to.

[burfadel]

Ah ok. Yeah I just experienced it again a second ago. Video does not play, but audio continues. Moving the progress bar the picture changed of course, but was a still picture still :), but audio played normally. I guess it would be easier if it threw a crash to track it through event manager, or that it was easily reproducible. Your setup is completely different to mine, I do believe though it is related to the recent fix as it wasn't happening before for me.

[burfadel]

Well, was experiencing it enough to try disabled 'Resize windows to fit when opening media', no go. Because it doesn't happen all the time for me it was seemingly hard to test. However, I've tested all combinations of settings of what was causing the now fixed crash in v2.45.0, and disabling them of course (which stopped it crashing previously) and it still happens. This confirms what ryrynz said above. I suspect though it is still related to the crash fix.

[burfadel]

Sorry for the third post in a row, since it's a different 'issue' I've made a separate post about it.

I notice that sometimes after playing files MPDN seems to stay in memory even if you close it off. If you look in task manager, under 'Details' tab, it shows programs in memory. If you sort it by name (by clicking on name) if not already in alphabetical order, you may see 'MediaPlayerDotNet.exe' even if you have already closed it off. If it is still there it uses much less memory than normal.

Can anyone else confirm that it still shows after closing, after playing media files? Issues can only be fixed if people report them, and more so if others confirm that such issues exist :). Not sure what causes it, whether it happens with only certain media files, or whether you have to watch a certain amount first, changed files etc. It's more so whether other people see the issue after closing with normal use. Thanks!

[ryrynz]

Closing down fine here. I think I recall a number of times where I had to wait a bit before relaunching MPDN as it wouldn't load again while it was still in memory, but I haven't experienced anything recently.
Might want to rename your config folder and give it another shot and see if it still happens and then if so, see if you can't narrow it down. Any particular types of files, play times? etc.

[burfadel]

It's not all the time, I've noticed it only a couple of times. Might be after playing a MPEG-2 stream (from recorded TV shows) to playing h.264 media etc... then going between full screen and windows etc. Not entirely sure, like the other issue of the picture freezing it only occurs some of the time it's probably hard to track down. I believe it only has happened recently. Could be related to the picture freezing issue? That is, might only happen after that has occurred?

[Zachs]

Hi guys just a heads up. If you haven't noticed I've been away but will be back end of next week. I won't be able to do much in the mean time but please do continue to provide more info as it would make it easier for be to reproduce the problems.

I've noticed MPDN not terminating before and found it was due to xysubfilter, so try disabling that to begin with. It is directshow filter related.

[aufkrawall]

Funny thing, I've noticed this for the second time: After installing the latest Windows 10 updates, MPDN needs longer to start.
Reinstalling it shortens the start times to the normal numbers.

[elraffa]

Hi Zachs..I would like to ask if you could incorporate an option to stretch hardcoded videos in MPDN.. I have a small monitor and films with black bands are too small for me.. In potplayer there is an option to do this (frame height +)
Hi zachs, please dont forget my request

[Zachs]

Yes, me! The first one at least. I'm using Direct3D 11, R9-280X, exclusive mode etc, but the symptoms sound identical. It seems to be a bit hard to reproduce though, it doesn't happen all the time.

Under options --> general, do you have either of the resize options selected, if so, which?

Also, do you have any option at all selected under 'Audio Renderer' --> 'Audio Script'?

If so to any of the above, set audio script to 'none' and see if the behaviour continues. If so, try deselecting the resize options and see if the problem is resolved. The reason for this is a crash bug was recently fixed that affected these options, but I believe there is still some residual related 'buggyness'...

Of course, if your trials regarding these settings prove fruitful, report back which one was the offending options for you. Unfortunately for me it only occurs sometimes, which makes testing difficult. It is an annoying bug though!

Hi guys,

Test build 3560 is now available. I haven't been able to replicate the issue but I did find a couple of problems with the code that might've contributed to the issue. Let me know if it fixes the problem for you.

[Zachs]

Hi zachs, please dont forget my request

I haven't forgotten but it has a very low priority as it's not a feature I use personally.

[ryrynz]

Looks good, no issues changing from Windowed to FSE. I still get the odd pause for a couple of seconds while seeking (generally backwards seeking using the arrow keys) I might have this happen 1/40 times or so.
The decoder and render queue just sit on 1 & 0 for a bit when this happens then playback continues as normal. This might be a bit of a non issue, and not be related to the player..

Okay just managed it again, this time is stalled for a good five seconds. Generally it's very responsive, guess I'll start testing options now.

I've changed about everything I can with no effect. I can generally cause this delay to happen by seeking about 5-7 steps forward and then a few backward. Audio continues while the picture and OSD is frozen.

[CruNcher]

I think MPDN is the best player for quality video ! great!

It's really interesting to watch this development

as interesting as to watch Sonys Vegas or the Paint.net development :)

He does a great job seperating the GUI from the Render Core and that is the most important :)

[Zachs]

It's really interesting to watch this development

as interesting as to watch Sonys Sony Vegas or the Paint.net development :)
If you fully understand how the GC works under .net desktop you'll find it to be a non issue. I've written my own production quality GC that is used for time sensitive applications on embedded platforms and even in those situations, you can always find ways to get around the GC pause. MPDN does a lot of its time sensitive stuff outside of .net. Memory overhead can be higher but GC tends to be faster in a highly multi threaded app, especially vs something like a smart pointer in C++.

[Zachs]

Looks good, no issues changing from Windowed to FSE. I still get the odd pause for a couple of seconds while seeking (generally backwards seeking using the arrow keys) I might have this happen 1/40 times or so.
The decoder and render queue just sit on 1 & 0 for a bit when this happens then playback continues as normal. This might be a bit of a non issue, and not be related to the player..

Okay just managed it again, this time is stalled for a good five seconds. Generally it's very responsive, guess I'll start testing options now.

I've changed about everything I can with no effect. I can generally cause this delay to happen by seeking about 5-7 steps forward and then a few backward. Audio continues while the picture and OSD is frozen.
It's actually not frozen but waiting for the next frame to arrive so it can be rendered and presented. I've tried seeking back and forth but haven't been able to replicate the issue though.

[ryrynz]

It's actually not frozen but waiting for the next frame to arrive so it can be rendered and presented. I've tried seeking back and forth but haven't been able to replicate the issue though.

I've managed to reproduce this on my HTPC system as well.
I don't have this issue with MPC-BE/madVR, I guess the next step is to try earlier MPDN versions as well, I do know that 3550 has this issue for me as well.

While I'm on the topic, I know I've asked before but what the hell..can MPDN hope to seek any faster than it currently does?
MPC seeks about twice the speed (using same LAV filters, Directsound, Direct3D etc)
I've changed all MPDN defaults to their fastest values but I can't reach anywhere near as fast as MPC.

[eX``]

Hi. I have a little problem with MPDN, but I do not think this is exclusively related to it, for I managed to reproduce the error on MPC-HC. Hope you guys still can help.

Faulting application name: MediaPlayerDotNet.exe, version: 2.45.1.3557, time stamp: 0x562e0218
Faulting module name: nvcuvid.dll, version: 7.17.13.5887, time stamp: 0x563756f0
Exception code: 0xc0000005
Fault offset: 0x0000000000068548
Faulting process id: 0x161c
Faulting application start time: 0x01d117b89f092a87
Faulting application path: C:\Program Files\MPDN\MediaPlayerDotNet.exe
Faulting module path: C:\Windows\system32\nvcuvid.dll
Report Id: dd74865c-83ab-11e5-b707-448a5b219869

Unfortunately I can't give you any MPDN stamp for it crashes before I can fetch it.

[burfadel]

Hi. I have a little problem with MPDN, but I do not think this is exclusively related to it, for I managed to reproduce the error on MPC-HC. Hope you guys still can help.



Unfortunately I can't give you any MPDN stamp for it crashes before I can fetch it.

That is an issue with your Nvidia driver. Try updating your driver to the latest and see if the problem persists :).

[CruNcher]

If you fully understand how the GC works under .net desktop you'll find it to be a non issue. I've written my own production quality GC that is used for time sensitive applications on embedded platforms and even in those situations, you can always find ways to get around the GC pause. MPDN does a lot of its time sensitive stuff outside of .net. Memory overhead can be higher but GC tends to be faster in a highly multi threaded app, especially vs something like a smart pointer in C++.

Yeah it's interesting to compare the approach vs MPC-BE which i find currently the best win32 optimized Player in time sensitivity in a Multithreaded Scenario (apart from Microsofts own MFT focused work) :)

And yeah sure GC can get better over time if you optimize it and have full controll over it best example Firefox, but is it enough to compete in the end on resources and would this not endup in performance loses on the time sensitive optimized parts im still skeptical ;)

I mean we not going for nothing into the Low Level API age and leaving DX 11s generic IHVs based GCs for DX 12 and more uniquely approaches tailored specific to the applications requirements ;)

And currently either Firefox or Google Chrome look really old vs Microsofts Edge, especialy in a low power scenario as you seem to care a lot about :)

[eX``]

That is an issue with your Nvidia driver. Try updating your driver to the latest and see if the problem persists :).

Yup I know, but I don't know where it comes from. Just updated this morning, clean install, before posting here, nothing changed. Anyway, I'll look around=). Let me know if rings a bell though.

[Zachs]

Yeah it's interesting to compare the approach vs MPC-BE which i find currently the best win32 optimized Player in time sensitivity in a Multithreaded Scenario (apart from Microsofts own MFT focused work) :)

And yeah sure GC can get better over time if you optimize it and have full controll over it best example Firefox, but is it enough to compete in the end on resources and would this not endup in performance loses on the time sensitive optimized parts im still skeptical ;)

I mean we not going for nothing into the Low Level API age and leaving DX 11s generic IHVs based GCs for DX 12 and more uniquely approaches tailored specific to the applications requirements ;)

And currently either Firefox or Google Chrome look really old vs Microsofts Edge, especialy in a low power scenario as you seem to care a lot about :)
Like I said you really need to fully understand GC as it is implemented in .net but this is not the forum to discuss it.

[Anima123]

There's controversial discuss around the topic: should SuperRes using high-quality downscaler or not, in madVR thread? However when I was playing around the options kept within the MPDN version of SuperRes, I found the high-quality downscaler altered straight lines in the video, which is obvious when playing video with hard-encoded subtitles. Only enabled the high-quality downscaler will cause that kind of artefacts, no matter which image-doubler used. BTW, NNEDI3 itself does not have that kind of 'effect'.

I will upload some images for comparison later today.

Shiandow, it would be nice if you can comment of this, maybe after I uploaded the images for camparison?

The pic with HQ downscaling off:
https://www.dropbox.com/s/jqm4ssqe1nznlik/Subtitle_simple_downscaler.png?dl=0

Pic with HQ downscaling on:
https://www.dropbox.com/s/fgbeq7gn6jcvyth/Subtitle_HQ_downscaling.png?dl=0

[Zachs]

I've managed to reproduce this on my HTPC system as well.
I don't have this issue with MPC-BE/madVR, I guess the next step is to try earlier MPDN versions as well, I do know that 3550 has this issue for me as well.

While I'm on the topic, I know I've asked before but what the hell..can MPDN hope to seek any faster than it currently does?
MPC seeks about twice the speed (using same LAV filters, Directsound, Direct3D etc)
I've changed all MPDN defaults to their fastest values but I can't reach anywhere near as fast as MPC.

I just tried seeking on my old i5 laptop via mechanical USB HDD and frame appear as soon as I click on the seekbar. Seek is instantaneous (if there's any delay it's imperceptible). This is with free-to-air footage that is 5GB in file size. With other files that have sparse keyframes it does take longer but in those cases MPDN's just waiting for the decoder to come back with a frame.

I still haven't been able to replicate any problems with audio playing while video waits regardless of how much I tried seeking back and forth and I've tested it on several machines already. Perhaps it's related to certain video files?

[aufkrawall]

ryrynz, are you using the same LAV settings/version for MPDN as with other players?
MPDN seeking is rather "instant" here with H.264 DXVA2 CB & GTX 980.

[Shiandow]

There's controversial discuss around the topic: should SuperRes using high-quality downscaler or not, in madVR thread? However when I was playing around the options kept within the MPDN version of SuperRes, I found the high-quality downscaler altered straight lines in the video, which is obvious when playing video with hard-encoded subtitles. Only enabled the high-quality downscaler will cause that kind of artefacts, no matter which image-doubler used. BTW, NNEDI3 itself does not have that kind of 'effect'.

I will upload some images for comparison later today.

Shiandow, it would be nice if you can comment of this, maybe after I uploaded the images for camparison?

The pic with HQ downscaling off:
https://www.dropbox.com/s/jqm4ssqe1nznlik/Subtitle_simple_downscaler.png?dl=0

Pic with HQ downscaling on:
https://www.dropbox.com/s/fgbeq7gn6jcvyth/Subtitle_HQ_downscaling.png?dl=0

Those look identical to me. In fact there's a decent chance they are identical, I just discovered a bug that essentially forces HQ downscaling on.

[Milardo]

Any chance that MPDN will have support for tv tuner/capture cards-like to apply all these rendering features of MPDN to devices i have like those?

[Anima123]

Those look identical to me. In fact there's a decent chance they are identical, I just discovered a bug that essentially forces HQ downscaling on.

Was this bug introduced recently? I do see difference with some version before.

[Shiandow]

Was this bug introduced recently? I do see difference with some version before.

Looks like it's been there from the start. Odd, I thought it had worked at some point. Anyway, I just tried the low quality downscaling, and I found it introduced a lot of aliasing and ringing, and it's not really much faster, so generally I don't think it's really worth using. Might need to add HQ downscaling to SuperChromaRes as well.

[eX``]

(Little off) Ok, just wanted to give you the follow-up :p. So the nvidia "clean install" was screwed, tons of updates above one another had something to do with this problem, for I formated, reinstalled everything and everything is working.
Still one little problem though: i STILL have this FU**NG white veil over my screen. My screen is calibrated, working perfectly, but everytime I install my drivers, it screws with the gamma or anything and I feel like watching at my screen through a white veil ... (like when I open MPDN, the deep black screen is now dark grey....)
(Little off)
édit : like this Before (http://www.prad.de/images/monitore/iiyama_xb2783hsu-b1/schwarz1.jpg) / After (http://www.prad.de/images/monitore/iiyama_xb2783hsu-b1/schwarz2.jpg)

SOLVED: The god of computers received my calls :p something in the nvidia panel was screwed. It touched everything 10 times, and thought "what's the point" and then 5min later I thought "ok one more time" and just clicked on the "Digital Vibrance thing (without changing the actual value) and...EUREKA ^^

Not solved in fact. Everytime I put MPDN in fullscreen mode, it screws back, and I have to go back to tweak this "Vibrance". This is not doing it with any game or application or anything.

[eX``]

Forget about it, it is another "driver fail", it does it too sometimes when I launch applications that have to go fullscreen.

[Braum]

Did someone try MPDN with Wine on linux ? Did it worked ? :thanks:

[BluesFanUK]

I'm having major difficulty installing this.

I downloaded the certificate and the installer, and I ran the certificate install bat file first (as Admin), all seemed fine. When I click the installer it just freezes up my explorer. The install circle just turns and the PC needs to be reset so I can delete the file. I've tried this about five times now using both 64 and 86 versions and both do the same thing.

I'm on Windows 10, any help appreciated.

EDIT - Disabled Windows Smartscreen too as despite installing the certificate the installer was still telling me the program is unrecognised. Same problem. Soon as I click the installer or run as Admin it just shows the loading circle and freezes that particular window. :(

EDIT 2 - Downloaded the portable version and despite giving me the warning about it being an unknown program, it opens. I don't get why the installer isn't working though... Back to MPC-HC for the time being.

[Zachs]

I downloaded the certificate and the installer, and I ran the certificate install bat file first (as Admin), all seemed fine. When I click the installer it just freezes up my explorer. The install circle just turns and the PC needs to be reset so I can delete the file. I've tried this about five times now using both 64 and 86 versions and both do the same thing.

I'm on Windows 10, any help appreciated.

Sounds like a bug in your virus scanner locking up.

[louiebh]

is this program able to replace madvr+mpc? id like to switch over if i can get everything in 64 bit atm reclock and svp are holding me back would this program be the one to succeed?

[BluesFanUK]

Sounds like a bug in your virus scanner locking up.

You're a star, disabled Avast, installed the player.

I'm sure I disabled it when I did a clean boot, anyway, many thanks! :-)

[Zachs]

The latest MPDN Extensions (v1.24.0) adds basic support for RAR archives and DVD.

You can now play from RAR archives directly without having to unpack them. Note that it only supports RAR archives created using the 'store' method (i.e. no compression). The filter was compiled straight from http://www.v12pwr.com/RARFileSource/.

As for playing DVD, you should now be able to open VTS_xx_0.IFO files in DVD\VIDEO_TS folder. As I hardly ever play DVDs, this is probably as far as I would support it in MPDN.

Cheers,
Zach

[Zachs]

is this program able to replace madvr+mpc? id like to switch over if i can get everything in 64 bit atm reclock and svp are holding me back would this program be the one to succeed?

MPDN has full replacement for reclock (in fact it has 2 different methods) but you'll have to find out if SVP even runs under 64-bit.

[louiebh]

MPDN has full replacement for reclock (in fact it has 2 different methods) but you'll have to find out if SVP even runs under 64-bit.

oh wow thats actually great in mpc i just need to set external filter to ffdshow raw and svp works so i will try it soon

my last query im wondering is in MPC there is a feature for capture which i hope to use in future my idea is to capture with a capture card then let my pc process it with my media player this feature currently has me sticking to mpc but 64 bit would be a bit less taxing on my poor laptop

the feature im speaking of is:

http://www.dropbox.com/s/gcz8y6mlignrn5c/Screenshot%202015-11-11%2011.09.29.png

in this screenshot i have linked does mpdn have a similar function?

[Zachs]

the feature im speaking of is:

http://www.dropbox.com/s/gcz8y6mlignrn5c/Screenshot%202015-11-11%2011.09.29.png

in this screenshot i have linked does mpdn have a similar function?

No. That feature could be added via player extension, but no one has contributed such an extension yet.

[ryrynz]

ryrynz, are you using the same LAV settings/version for MPDN as with other players?


Yes, I install LAV separately (running latest nightly)


I still haven't been able to replicate any problems with audio playing while video waits regardless of how much I tried seeking back and forth and I've tested it on several machines already. Perhaps it's related to certain video files?

I can duplicate it on just about every file I test. The delay can be up to six seconds long, all previous seeking up until that point is as fast as I'd expect.

[Zachs]

I've tried it in 4 different PCs now to no avail. What exactly do you have to do to reproduce the problem?

[ryrynz]

I've tried it in 4 different PCs now to no avail. What exactly do you have to do to reproduce the problem?

Hmm, I seek forwards say five or ten odd times then seek backwards a few times, then forwards again then backwards more etc.
I do it fairly quickly too basically trying to force it to freeze, it's not something anyone would do in normal operation of the player.

[Zachs]

Hmm, I seek forwards say five or ten odd times then seek backwards a few times, then forwards again then backwards more etc.
I do it fairly quickly too basically trying to force it to freeze, it's not something anyone would do in normal operation of the player.

I managed to replicate some sort of bug which may not be the same one but certainly caused the player to freeze using your method. When it freezes though, it doesn't continue unlike what you're seeing.

[ryrynz]

I managed to replicate some sort of bug

Cool, will try the next test build.

[Zachs]

Cool, will try the next test build.

It's available now.

[ryrynz]

It's available now.

Oh sneaky.. you did a proper release.

[Zachs]

Oh sneaky.. you did a proper release.

Yeah it fixes a bug anyway even if it's not the one you encountered.

[ryrynz]

Yeah it fixes a bug anyway even if it's not the one you encountered.


I thought you might've fixed it there.. but nah. Just had it pause for a few seconds, took me a little while longer than usual for it to happen. I pretty sure it always happens on a backwards seek.

[Zachs]

I thought you might've fixed it there.. but nah. Just had it pause for a few seconds, took me a little while longer than usual for it to happen. I pretty sure it always happens on a backwards seek.

I've tested this latest version on all the machines I have access to. I tried seeking forwards over 10 times and then go backwards in rapid succession. I tried repeating this for at least half an hour each on the different machines but I've not managed to replicate the issue. There must be something else that's causing the bug to manifest itself on your setup.

[ryrynz]

There must be something else that's causing the bug to manifest itself on your setup.

I've managed to replicate it on both my machines W7-Intel & a W10-Nvidia with different content. I'll install a fresh copy on one or two of the laptops at work and see if I can duplicate it on one of those as well.

[Anima123]

Shiandow, as I remembered, you've mentioned that you were trying to find a good softer for SuperRes. Is there any progress on that? Even on experimental stage, some of us would like to offer testing as your request, just in case you need it.

[foxyshadis]

Odd bug I just encountered: I have two external subtitles in the same folder as this mkv: same name, so they're automatically picked up. One is .srt, one is .smi. They're both labeled x264 in VSFilter and MPDN because the file names end in .x264.ext, must be a VSFilter thing, but perhaps because they're both given the same name, in MPDN's subtitles menu I have only x264 and No Subtitles, where No Subtitles actually shows the .smi. I can't turn subtitles off without moving or removing the file.

[Shiandow]

Shiandow, as I remembered, you've mentioned that you were trying to find a good softer for SuperRes. Is there any progress on that? Even on experimental stage, some of us would like to offer testing as your request, just in case you need it.

Well, there's been some progress. I'll see if I can get something ready for testing some time soon.

[Zachs]

Odd bug I just encountered: I have two external subtitles in the same folder as this mkv: same name, so they're automatically picked up. One is .srt, one is .smi. They're both labeled x264 in VSFilter and MPDN because the file names end in .x264.ext, must be a VSFilter thing, but perhaps because they're both given the same name, in MPDN's subtitles menu I have only x264 and No Subtitles, where No Subtitles actually shows the .smi. I can't turn subtitles off without moving or removing the file.

I'll have a look into it. I certainly wasn't expecting the subtitle track name to be the same though.

EDIT: I tested this and can't seem to be able to reproduce the bug. It came up with 2 entries of x264 and a "Hide Subtitle" entry.

EDIT: Nevermind, found the problem.

[Zachs]

Test build 3572 is available. Should fix the subtitle problem.

@ryrynz can you let me know if it fixes the problem with dropped frames after seeking too? You're right about the refresh rate. The bug does have something to do with it but not directly caused by the (lack of) initial precision.

[ryrynz]

Sorry to say I'm still getting dropped frames when seeking with this build. BTW the installer references LAV 0.65 on mouse over, 0.67 is now released.

[Zachs]

Ah thanks. I've updated LAV in the installer to 0.67.

Test build 3573 should have the frame drop problem fixed. It's the same problem I identified earlier, the fix was incorrectly implemented so it made it worse. Anyway should be properly fixed this time.

[aufkrawall]

Is it normal that Fluid Motion still generates reported delayed frames when the OSD is closed?
The really odd thing is that there are much more reported delayed frames when I have a higher GPU usage via render scripts, compared to not using render scripts at all.
The queues are always full and it happens both in windowed mode and FSE with DX9 and 11.
I think to best reproduce you should play a 60fps video at 75Hz with Fluid Motion, generate a high GPU load (but not too high to prevent dropped frames) and open up the OSD from time to time to see if the reported delayed frames count is increasing.
There are no reported delayed frames with FM off or if the OSD is always active.

[fagoatse]

Greets, I've come across an issue when video playback runs at ~30fps even though mpdn reports 23.976fps. Audio seems fine but the entire thing is desynced as a result and sometimes playback just freezes. It only applies to a certain encode(h.264 8bit dvdrip) and it runs fine in the default windows player. Any ideas what could be the culprit?
I use the latest mpdn, mostly default settings for the player and lavfilters except for qs decoding(already checked).

[Zachs]

Is it normal that Fluid Motion still generates reported delayed frames when the OSD is closed?
The really odd thing is that there are much more reported delayed frames when I have a higher GPU usage via render scripts, compared to not using render scripts at all.
The queues are always full and it happens both in windowed mode and FSE with DX9 and 11.
I think to best reproduce you should play a 60fps video at 75Hz with Fluid Motion, generate a high GPU load (but not too high to prevent dropped frames) and open up the OSD from time to time to see if the reported delayed frames count is increasing.
There are no reported delayed frames with FM off or if the OSD is always active.

How long do you have to run it without OSD to have delayed frames? Can't seem to be able to reproduce the problem here. Does it occur in FSE mode or just windowed?

Greets, I've come across an issue when video playback runs at ~30fps even though mpdn reports 23.976fps. Audio seems fine but the entire thing is desynced as a result and sometimes playback just freezes. It only applies to a certain encode(h.264 8bit dvdrip) and it runs fine in the default windows player. Any ideas what could be the culprit?
I use the latest mpdn, mostly default settings for the player and lavfilters except for qs decoding(already checked).

You'll need to give us a sample. There's no way to tell what the actual problem is otherwise.

[Anima123]

How long do you have to run it without OSD to have delayed frames? Can't seem to be able to reproduce the problem here. Does it occur in FSE mode or just windowed?


The delayed frames that I have been experienced seems been fixed with the latest version, at least on my systems.

[Zachs]

The delayed frames that I have been experienced seems been fixed with the latest version, at least on my systems.

Yeah it doesn't happen on any of my systems either.

[edison]

Does the audio script opencl version still not work with NVIDIA Kelper w/359.00 driver?

[Zachs]

You need to install Intel or AMD OpenCL drivers for audio scripts to work. The reason your discrete GPU isn't used is because it's not worth transporting across the PCIe link and back for something the CPU or iGPU can do perfectly fine.

[luigizaninoni]

Am I missing something, or in order to customize shortcut keys you need to install the source of extensions, customize cs files etc. ? Something more user friendly would be really appreciated

[Zachs]

Am I missing something, or in order to customize shortcut keys you need to install the source of extensions, customize cs files etc. ? Something more user friendly would be really appreciated

Yes this has been said before.

[Anima123]

The latest MPDN Extensions branch 'filter' works with MPDN 2.45.5 now after several commits.

Happy Thanks Giving to all developers! And to all members here.

[foozoor]

Should I install the certificate even if I do not use the installer?
Do you include lavfilters and xysubfilter in the portable/standalone version?

[Magik Mark]

I do not know if this has been asked. It seems I can't find it in the search option.

1. Is it possible to have error diffusion option? In madvr it gives better contrast ratio

2. Is it possible to have subtitles moved on the black bars? This gives maximum viewing experience

Thanks for the great work zach. We really appreciate it

[Zachs]

Should I install the certificate even if I do not use the installer?
Do you include lavfilters and xysubfilter in the portable/standalone version?

Certificate: Yes (as this allows you to verify MPDN has not been tempered with).
LAV/XySub aren't included in the portable version. I have plans (high priority) to change this so users without admin privileges can run MPDN without installing anything.

I do not know if this has been asked. It seems I can't find it in the search option.

1. Is it possible to have error diffusion option? In madvr it gives better contrast ratio

2. Is it possible to have subtitles moved on the black bars? This gives maximum viewing experience

Thanks for the great work zach. We really appreciate it

1. No one contributed such a thing yet. But I'm not sure why it would give you better contrast ratio. Should it even do that?
2. No but it's on my todo list.

[Magik Mark]

1. No one contributed such a thing yet. But I'm not sure why it would give you better contrast ratio. Should it even do that?
2. No but it's on my todo list.

I'm also baffled. It turns out it dims the video a little bit. The color pops more. If you have a display that has been isf calibrated, you would see the difference. It maybe subtle for some. But when you start watching videos without error diffusion. You would missed it

[huhn]

than it is buggy if it changes the gamma. and the contrast doesn't change what so ever no matter what type of dither is used.

[Magik Mark]

Zachs,

Is there a way to port MPDN to Kodi? Something similar to Dsplayer? The author decided to give up on its development. I'm sure there will be pipol here who are willing to donate for this cause.

We could see that you are a very dedicated and passionate developer. You are the man for this. Hope you consider

Should you be needing donation to help you defary the cost of this advocacy, just let us know

[Plutotype]

Hi folks,
Would be InMemory playback possible to code? I mean having 64/128GB RAM, put the whole FHD/UHD movie into the RAM and let the source HDD idle. How hard would be this possible to achieve?
Thanks

[Zachs]

Zachs,

Is there a way to port MPDN to Kodi? Something similar to Dsplayer? The author decided to give up on its development. I'm sure there will be pipol here who are willing to donate for this cause.

Media Browser Theatre already includes MPDN as its player - there's no reason why Kodi can't do the same. MPDN's extensions API is powerful enough to support that. If there's anything it lacks, I'd be more than happy to improve the API.

Hi folks,
Would be InMemory playback possible to code? I mean having 64/128GB RAM, put the whole FHD/UHD movie into the RAM and let the source HDD idle. How hard would be this possible to achieve?
Thanks

Uh... Copy the file into a RAMDisk?

[huhn]

waist of time just think about how long it would take to copy a BD from a harddisk to the memory.

[ryrynz]

Hi folks,
Would be InMemory playback possible to code? I mean having 64/128GB RAM, put the whole FHD/UHD movie into the RAM and let the source HDD idle. How hard would be this possible to achieve?
Thanks

Maybe set your LAV queues to maximum as an, alternative to a RAM drive? It's not exactly the solution you're after but it would perform a similar sort of job.

[nevcairiel]

Setting the packet queue size and queue memory limit in LAV very high would load the entire file into memory, however its important to note that it doesn't seek from that queue or anything, so if you seek it would read from the file again - so its not really optimized for such use-cases.

[ryrynz]

Is it possible we might see media player seeking being able to access these buffers in future?

[nevcairiel]

Is it possible we might see media player seeking being able to access these buffers in future?

No. I don't believe in such hacks.

[petoulachi]

Hello,

I'm starting using MPDN and wondering one simple question : do the 3DLUT files generated with DispCalGUI + MadTPG will work fine with MPDN ? Maybe it's a dumb question, but as the file is generated using MadVR, and as MPDN does not use MadVR, I Wonder if everything will still be ok ?

Also, is there any tutorials, or at least some adviced script to improve image quality with a GTX750 and a i5 6600 ?

Thanks !

[Zachs]

I'm starting using MPDN and wondering one simple question : do the 3DLUT files generated with DispCalGUI + MadTPG will work fine with MPDN ? Maybe it's a dumb question, but as the file is generated using MadVR, and as MPDN does not use MadVR, I Wonder if everything will still be ok ?

Yes it should be fine.


Also, is there any tutorials, or at least some adviced script to improve image quality with a GTX750 and a i5 6600 ?

You can start by using the presets in the OP.

[aufkrawall]

2.46 still gives me delayed frames with OSD closed when Fluid Motion is active with 60fps@75Hz (1080p -> WQHD SuperRes Sinc-Hann + deband).
:(

[Magik Mark]

Zach

Is it possible to put "stop pause & play" buttons while in FSE?

[Nullack]

Im trialling MPDN x64 on Windows 10 x64. I already had 64 bit lav filters and Xysubfilter installed. So I didnt install those. MPDN wont run, it tries to execute but it just ends with no system logs or any windows. How do I enable logging? I read Optimus problems with MPDN startup, but I dont have that just an Intel CPU and a GTX 960.

Ive got the Zac certificate and the OpenCL runtimes for Intel CPUs since I have an Nvidia GTX 960. I tried re-installing this time with the MPDN lav filters, xysubfilter and MPC audio but it still wont start.

[foxyshadis]

Delete everything in %localappdata%\MediaPlayerDotNet\, just in case.

Options:
1. Download Process Monitor (https://technet.microsoft.com/en-us/sysinternals/processmonitor.aspx). When you run it (will prompt for admin) you should be asked for a filter, otherwise click the funnel icon up top. Set the filter to "Path" "contains" "MediaPlayerDotNet.exe", click Add & OK, and then run MPDN again. With any luck something near the end will point to a missing file or something like that. Posting the log might help.
2. Right click the home icon (the windows Flag) and open Event Viewer, expand Windows Logs, and look in Application for any instances of Application Error. If the program is silently crashing it'll show up there.

[Nullack]

Thanks

Here is the log file for process monitor in XML:

https://www.sendspace.com/file/o8o2to

Nothing relevant MPDN is appearing anywhere in the event viewer be it app, system security etcetc

Also there is no %localappdata%\MediaPlayerDotNet\ directory, just of course the usual ProgramData master dir tree

I installed using the x64 package of MPDN

[foxyshadis]

Dammit, sorry about that, it was meant to be Process as the filter, not Path. This log wouldn't show the termination. I was wondering why it was so short. It'd be more handy to export as a pml file, too.

[BluesFanUK]

Anyone have recommended settings for 1080p 60fps videos? I have a 5820K, Palit SJS 980ti and a 4K monitor, some 1080 content looks a bit naff on the new monitor, so i'm trying to get the best quality with this player.

Seems like any settings I change cause a lot of dropped frames on 1080 60fps (4K 60fps is magic though!). Full Screen Exclusive mode is also horrendous. Slow to open, slow to close, and causes loads of dropped frames switching between FSE and normal full screen.

[aufkrawall]

Dropped frames when entering/leaving FSE are expected, as long as the playback afterwards is fine.

In the first post, Zachs linked to profiles which I think are suitable for different GPU classes:
http://forum.doom9.org/showpost.php?p=1743874&postcount=3689
Maybe you want to give the preset for highend GPUs a try?

[BluesFanUK]

Stupid question - how do I paste that into MPDN? The Script Group under RenderScript allows you to pick and choose different scripts with the click of a button, but it won't allow you to paste text into the screen.

Ta.

[Shiandow]

Stupid question - how do I paste that into MPDN? The Script Group under RenderScript allows you to pick and choose different scripts with the click of a button, but it won't allow you to paste text into the screen.

Ta.

Just select the place on the left pane where you want the scripts, and paste (either using the right click menu, or by pressing Ctrl+V).

[BluesFanUK]

Cheers, weirdly that didn't paste for me when I was at work. Probably an admin setting.

EDIT: I'm getting random frame drops and i'm not sure why. I've cleared the settings, added the high GPU one then played several files. Results:

Sub 720p - 0 dropped frames

720p 30fps - 5~ dropped frames per second.
720p 60fps - Even worse than above

1080p 30fps - 0 dropped frames
1080p 60fps - 8~ dropped frames per second.

4K 30fps/60fps - both flawless.

My only assumption for those results is because the setting was created on a 1080 monitor rather than 4K (In full screen mode)? The 720 one just seems odd to me. I thought the lower the res the more taxing it is on the GPU, yet anything below 720 is absolutely fine. I'm miffed. When I move out of full screen it's fine, but kind of defeats the purpose of using a video renderer. FSE turned off too as it's painful to use.

Edit 2 - The Mid range one is fine with all settings, though i'm not sure whether it's better than the default settings.

[Nullack]

Dammit, sorry about that, it was meant to be Process as the filter, not Path. This log wouldn't show the termination. I was wondering why it was so short. It'd be more handy to export as a pml file, too.

THanks, here it is filtered by process in PML:

https://www.sendspace.com/file/g4t31v

Also tried a complete driver purge of Nvidia with DDU, and a fresh install of the latest drivers. No difference, MPDN when executed displays nothing and just exits. Nothing in any logs that I can find.

[foxyshadis]

THanks, here it is filtered by process in PML:

https://www.sendspace.com/file/g4t31v

Also tried a complete driver purge of Nvidia with DDU, and a fresh install of the latest drivers. No difference, MPDN when executed displays nothing and just exits. Nothing in any logs that I can find.

According to that, the last thing that happens before it up and quits is the first attempt to access DirectShow. I suspect that that something to do with it, but I'm not exactly sure what; maybe you have a bad decoder?

[Nullack]

Hmmm strange. Its default windows 10 x64 with just lav filters as far as I know; it was rebuilt this month under a clean install. I cant point to an obvious bad filter. Is there a further diagnostic step on this?

[Zachs]

Hmm can you run other .net app at all?

[Nullack]

Yes, and I tried just for a random thing the MS .net fixer which didnt find any problems.

[Zachs]

What does event viewer say when you tried to run MPDN?

[Nullack]

No entries in app, sys or security

[Zachs]

So it just doesn't run? And no error messages whatsoever? Did you try both the installer and standalone versions?

[Nullack]

No it does'nt run. Just tried standalone x64 and that doesnt run either.

[Zachs]

You must have something that's preventing it from working properly. Have you tried disabling third party anti virus software?

[Nullack]

Mate there is none in the way of third party AV, its all default Windows build with the default Windows embedded AV. Thinking about it, its all default, with LAV filters, also I have geforce experience installed which I think installs some sorts of video filters too. Foxyshadis seemed to think it was filter related.

[Zachs]

Well I'm out of idea.

[Nullack]

Surely there is a way to do a debug build that will log it? This is one and zeros mate, it all makes logical sense at a low level :) MPC-HC and MPC-BE work fine. It would be a shame for you to give up, I genuinely believe I could provide helpful assistance with testing

[Zachs]

It depends. Anyway I will see what I can do.

[cascardian]

So, I've installed the Intel OpenCL 15.1 redistributable (old Nehalem here, no iGPU), but MPDN x64 doesn't seem to recognise its presence, so I of course get the newly implemented error message for the absence of the files when I turn on an audio script. Which files or file is MPDN specifically looking for and where? The intelopencl64.dll is present in System32.

The rest of the files are located in Program Files (x86)\Common Files\Intel\OpenCL\bin\x64, which is curiously not the folder specified by Intel's installer in the environment variables, though. The strange thing is that that folder does not exist at all, which leads to me to believe that Intel forgot to specify a new correct environment path in the 15.1 installer. Sadly, fixing the environment path to the folder above does not make MPDN recognise the files either.

Am I doing something wrong or is not having an iGPU in the first place responsible here?

[Zachs]

Am I doing something wrong or is not having an iGPU in the first place responsible here?

My HTPC CPU (Wolfdale) is older than your Nehalem and doesn't have an iGPU either. I'm running the same Intel OpenCL 15.1 drivers (https://software.intel.com/en-us/articles/opencl-drivers#win64).

Audio scripts don't load the DLLs directly - they're loaded through the OpenCL ICD.

You can also try installing the AMD CPU OpenCL drivers.

[foozoor]

Does MPDN work with hybrid laptops?

I use Leshcat's drivers that provide updated Intel and AMD drivers for hybrid laptops.

Unfortunately MPDN doesn't recognize my AMD gpu, only the Intel HD3000 is available in MPDN options.

[ryrynz]

Unfortunately MPDN doesn't recognize my AMD gpu, only the Intel HD3000 is available in MPDN options.

Are there any options to set a default GPU for particular programs in the driver's or perhaps in a tweak tool? Does setting it to a high power mode allow for the AMD GPU to be detected?

[foozoor]

Are there any options to set a default GPU for particular programs in the driver's or perhaps in a tweak tool? Does setting it to a high power mode allow for the AMD GPU to be detected?

When I set MPDN in high power mode, the AMD GPU is still not detected in MPDN and I've got black screen. If I set as default, based on alimentation, no more black screen but only the Intel HD3000 is detected. In GPU-Z, only the Intel HD3000 GPU charge increases.

I tried with mpc-hc and madVR and the AMD GPU works.

[Nullack]

It depends. Anyway I will see what I can do.

I had an idea. I dropped my 4K desktop down to 1080P and re-tested MPDN both on the 32 bit and 64 bit standalone builds. Sadly it didnt fix the problem but atleast I know its not 4K specific anymore.

[cascardian]

My HTPC CPU (Wolfdale) is older than your Nehalem and doesn't have an iGPU either. I'm running the same Intel OpenCL 15.1 drivers (https://software.intel.com/en-us/articles/opencl-drivers#win64).

Audio scripts don't load the DLLs directly - they're loaded through the OpenCL ICD.

You can also try installing the AMD CPU OpenCL drivers.

Already did try AMD's, still nothing, sadly. This appears to be a problem confined to my config, then, I take it?

As far as I understand, the ICD extension under Windows (https://www.khronos.org/registry/cl/extensions/khr/cl_khr_icd.txt) calls the ICD loader OpenCL.dll--which, in my case, gets replaced upon graphics driver installation with NVIDIA's 110 KB file--that then looks up the possible vendor files enumerated under HKEY_LOCAL_MACHINE\SOFTWARE\Khronos\OpenCL\Vendors, which, besides nvopencl.dll, also correctly lists Intelopencl64.dll for me. AMD's also gets listed when I try it.

I can't figure out what's wrong here. GPU Caps Viewer also confirms that Intel's OpenCL is present and usable. Starting from a fresh MPDN x64 install with default settings also doesn't help. If anybody has an idea, I'd really appreciate the help.

[Zachs]

Already did try AMD's, still nothing, sadly. This appears to be a problem confined to my config, then, I take it?

As far as I understand, the ICD extension under Windows (https://www.khronos.org/registry/cl/extensions/khr/cl_khr_icd.txt) calls the ICD loader OpenCL.dll--which, in my case, gets replaced upon graphics driver installation with NVIDIA's 110 KB file--that then looks up the possible vendor files enumerated under HKEY_LOCAL_MACHINE\SOFTWARE\Khronos\OpenCL\Vendors, which, besides nvopencl.dll, also correctly lists Intelopencl64.dll for me. AMD's also gets listed when I try it.

I can't figure out what's wrong here. GPU Caps Viewer also confirms that Intel's OpenCL is present and usable. Starting from a fresh MPDN x64 install with default settings also doesn't help. If anybody has an idea, I'd really appreciate the help.

Can you try running DebugView and see what errors the Audio Script framework spits out?

[Zachs]

Hi guys,

Belphemur has kindly setup a bug report / feature request tracking system here - http://mpdn.zachsaw.com/bug/view_all_bug_page.php.
We encourage everyone to use it to report bugs and request features so things can be tracked.

Cheers.

[Belphemur]

To add to what Zach said.

You can browse the issues without an account but you need to create an account to report them.

This is to ensure the traceability of the issue on both side (since you'll get notified if any comment/change happen in the issue) and reduce the risk for us to get spam by bots.

[GREG1292]

Zachs will your player be able to output 12 bit in the near future. I have a CRT projector and a Moomee input card that I set to 12 Bit but in your
player I am only able to set 10 bit. And I can clearly see the improvement nice work.

Thanks Greg

[madshi]

Direct3D/DXGI doesn't support 12bit rendering, only 10bit and 16bit.

[Shiandow]

Would a 16bit output get rounded down to 12bits by the GPU or can you only output 16bit for 16bit screens?

[madshi]

The GPU drivers do all sorts of funky things. Direct3D allows to render to 10bit or 16bit backbuffers. Some GPUs output 12bit even if you render to 10bit. Some output 10bit even if you render to 16bit. How they convert the bitdepths (rounding, shifting, dithering, whatever) is up to our best guess. I wish there were a way to force the GPU to do untouched passthrough of the rendered pixels to the display, but currently there isn't. The GPU driver does what it wants.

[cascardian]

Can you try running DebugView and see what errors the Audio Script framework spits out?

Hello again,

I apologise for not reporting back during the holidays. This is the DebugView output when starting, letting the OpenCL error display, and then stopping MPDN (which has debug messages enabled):
00000042 2.71643806 [7740] [150884'833,869ms] [INFO] Spent 501ms building/loading extensions
00000043 3.24853396 [7740] Information: Create ComputeContext(786013072) in Thread(1).
00000044 3.24962926 [7740] Information: Create ComputeCommandQueue(785183648) in Thread(1).
00000071 3.89513636 [7740] Mono.Cecil.AssemblyResolutionException: Failed to resolve assembly: 'DirectShowLib-2005, Version=2.1.0.0, Culture=neutral, PublicKeyToken=67e7b740cdfc2d3f'
00000072 3.89513636 [7740] at Mono.Cecil.BaseAssemblyResolver.Resolve(AssemblyNameReference name, ReaderParameters parameters)
00000073 3.89513636 [7740] at Mono.Cecil.DefaultAssemblyResolver.Resolve(AssemblyNameReference name)
00000074 3.89513636 [7740] at ICSharpCode.Decompiler.Ast.Transforms.IntroduceUsingDeclarations.Run(AstNode compilationUnit)
00000075 3.89513636 [7740] at ICSharpCode.Decompiler.Ast.Transforms.TransformationPipeline.RunTransformationsUntil(AstNode node, Predicate`1 abortCondition, DecompilerContext context)
00000076 3.89513636 [7740] at Cudafy.Translator.CUDALanguage.DecompileType(TypeDefinition type, ITextOutput output, DecompilationOptions options)
00000077 3.89513636 [7740] at Cudafy.Translator.CudafyTranslator.DoCudafy(CudafyModule cm, Type[] types)
00000078 3.89513636 [7740] at Cudafy.Translator.CudafyTranslator.Cudafy(IEnumerable`1 props, Type[] types)
00000079 3.89513636 [7740] at Cudafy.Translator.CudafyTranslator.Cudafy(eArchitecture arch, Type[] types)
00000080 3.89513636 [7740] at Mpdn.Extensions.Framework.AudioHelpers.GetCudafyModule(Type[] types)
00000081 3.89513636 [7740] at Mpdn.Extensions.Framework.AudioHelpers.LoadAudioKernel(GPGPU gpu, Type type, Type[] types)
00000082 3.89513636 [7740] at Mpdn.Extensions.Framework.AudioChain.AudioProc.Initialize()
00000149 6.85856342 [7740] Warning: ComputeCommandQueue(785183648) leaked!
00000150 6.85858440 [7740] Warning: ComputeContext(786013072) leaked!
00000151 6.86609030 [7740] Information: Dispose ComputeCommandQueue(785183648) in Thread(2).

[Zachs]

I see what the problem is. A bit strange it only happens on your machine though.
I'll put a patch in when I have the time. Been a bit busy over the last month or so.

[petoulachi]

Hello,

First of all, happy new year, wish you the best for 2016 :)

I'm havind sound/image delay issue with MPDN, check my bug report here : https://mpdn.zachsaw.com/bug/view.php?id=5

Don't have any issue using MPC/MadVR, or anything else in fact.

Regards,

[Shiandow]

Hi all,

Unfortunately there hasn't been too much news lately, development is still steadily carrying on, but with an emphasis on steadily.

Anyway, here (http://screenshotcomparison.com/comparison/158879)'s a teaser for a project I've been working on. It's an implementation of (and improvement on) the downscaling method described here (https://graphics.ethz.ch/~cengizo/imageDownscaling.htm), which was mentioned on doom9 a while back.

[Anima123]

Wow, this downscaling algorithm looks very impressive.

Will it be used as part of the SuperRes upscaling algorithm?

[Shiandow]

I might give it a try, I'm not too sure if it's a good idea to have multiple non-linear scaling algorithms on top of each other though. And SuperRes is sharper if you use a softer downscaling algorithm is, so using this method might backfire.

[ryrynz]

Anyway, here (http://screenshotcomparison.com/comparison/158879)'s a teaser for a project I've been working on.

Looks real good, that'll make us Full HD set owners happy with sticking with our sets for a while longer.
I don't downscale ATM, so what would be the best downscaler for this to really be compared to? Any reason for choosing Bicubic?

[Shiandow]

I chose Bicubic because, as far as I know, that's the one most frequently used as a general purpose downscaler. If anyone knows a better algorithm I could also compare it to that, but it's really a lot sharper than any other algorithm I know (maybe a bit too much in some cases, but that's relatively easy to fix).

[Anima123]

I might give it a try, I'm not too sure if it's a good idea to have multiple non-linear scaling algorithms on top of each other though. And SuperRes is sharper if you use a softer downscaling algorithm is, so using this method might backfire.

You mentioned earlier that you might borrow the similarity of this downscaling algorithm into SuperRes and discard the downscaling step in SuperRes for good.

[Shiandow]

You mentioned earlier that you might borrow the similarity of this downscaling algorithm into SuperRes and discard the downscaling step in SuperRes for good.

You're paraphrasing quite a bit there. Whether it can be done entirely without the downscaling step remains to be seen. It would definitely be interesting to create a Structural Similariy Super Resolution algorithm (if you'll pardon the pun). But that would really be an entirely different algorithm, it's not as easy as simply plugging in a different downscaling algorithm.

[madshi]

Hi all,

Unfortunately there hasn't been too much news lately, development is still steadily carrying on, but with an emphasis on steadily.

Anyway, here (http://screenshotcomparison.com/comparison/158879)'s a teaser for a project I've been working on. It's an implementation of (and improvement on) the downscaling method described here (https://graphics.ethz.ch/~cengizo/imageDownscaling.htm), which was mentioned on doom9 a while back.
Looking good. As a comparison Bicubic100 (Catmull-Rom = Bicubic50, in case the name isn't clear) might make sense, which is also a very sharp downscaling algorithm, but probably still softer than yours.

[Anima123]

I just cannot resist seeing how far a new idea can achieve, ;).

[ryrynz]

Shiandow when you do a comparison vs Bicubic 100 can you also give the performance difference as well?

[PetitDragon]

Hi all,

Unfortunately there hasn't been too much news lately, development is still steadily carrying on, but with an emphasis on steadily.

Anyway, here (http://screenshotcomparison.com/comparison/158879)'s a teaser for a project I've been working on. It's an implementation of (and improvement on) the downscaling method described here (https://graphics.ethz.ch/~cengizo/imageDownscaling.htm), which was mentioned on doom9 a while back.

Glad to hear this before they get the patent. :D

[ryrynz]

It's an implementation of (and improvement on) the downscaling method described here (https://graphics.ethz.ch/~cengizo/imageDownscaling.htm), which was mentioned on doom9 a while back.

Not sure if you still have the original but I thought it would be nice to see how you've improved on this, if it's possible. Might be best to open a new thread on this if discussion continues.

[Shiandow]

Looking good. As a comparison Bicubic100 (Catmull-Rom = Bicubic50, in case the name isn't clear) might make sense, which is also a very sharp downscaling algorithm, but probably still softer than yours.

Did a comparison with bicubic100 here (http://screenshotcomparison.com/comparison/158996), it's closer, but Structural Similarity based Downscaling still wins. And when you scale even more the difference becomes larger.

At the moment using structural similarity downscaling takes an extra 5% GPU on a GTX 560Ti, to downscale 4k to 1080p (at 24hz). Which is well within reach for most GPUs.

Actually, the main difficulty is that (right now) you need to scale the chroma to 4k first, which actually takes longer than the downscaling. But it should be possible to scale the chroma directly to the correct size. To compare the performance I scaled the chroma for both Bicubic and the Structural Similarity Downscaler.

[Shiandow]

Not sure if you still have the original but I thought it would be nice to see how you've improved on this, if it's possible. Might be best to open a new thread on this if discussion continues.

To stay somewhat on topic I won't go into the details (if you want me to we should start a new thread), but basically their method is not numerically stable, so you need to use very high precision calculations to get a decent result, which is slow. If you did their calculations with enough precision then you'd get a similar result, maybe even without any visible difference, although I wouldn't be surprised if there was.

[ryrynz]

Did a comparison with bicubic100 here (http://screenshotcomparison.com/comparison/158996), it's closer, but Structural Similarity based Downscaling still wins. And when you scale even more the difference becomes larger.

At the moment using structural similarity downscaling takes an extra 5% GPU on a GTX 560Ti, to downscale 4k to 1080p (at 24hz). Which is well within reach for most GPUs.


Those stones on the road really come to life.. This would have to be given the performance lack of any significant performance impact the go to or perhaps even the default downscaler for MPDN & madVR

If you did their calculations with enough precision then you'd get a similar result, maybe even without any visible difference, although I wouldn't be surprised if there was.

Thanks, what's left to fix or improve on?

[Shiandow]

Thanks, what's left to fix or improve on?

Surprisingly little. The algorithm is mathematically pretty air tight, and numerically now as well. What's left is essentially just testing it. Well that and I might need to look at this 'patent pending' issue, although as far as I can tell there's no way they would be granted a patent broad enough to cover the changes I made.

[Zachs]

Does anyone have a link to the actual patent itself? Even if it's just pending a grant, we should have access to the patent text itself or at least the patent number.

[ryrynz]

Does anyone have a link to the actual patent itself? Even if it's just pending a grant, we should have access to the patent text itself or at least the patent number.

I did some searching around, looked through most of his published stuff on it, videos etc but there's no link to it, probably because it's still pending. Someone could always email and ask.

This is as close as I got, Markus Gross's patents. (http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.htm&r=0&f=S&l=50&d=PTXT&RS=IN%2Fgross-markus&Refine=Refine+Search&Refine=Refine+Search&Query=IN%2FGross-Markus)

Shiandow would you be so kind as to take a ~720P animated film screenshot, give it NNEDI3 doubling (maybe quadrupling too ;) @64+ neurons) and then a downscale with this to 1080 and compare it to the same with Catmull Rom? I believe many are using Catmull Rom as a downscaler these days when performing doubling/quadrupling, would be nice to see how much perceived detail this adds after some high quality upscaling. Thanks!

[louiebh]

is this working with livestreamer yet everytime i try i get a stdin error i dont know how to get it working :(

[Zachs]

is this working with livestreamer yet everytime i try i get a stdin error i dont know how to get it working :(

What error did you get?

[louiebh]

What error did you get?

http://i.imgur.com/oqXWI5C.png

===================================

An unexpected error 'System.IndexOutOfRangeException' has occurred.

------------------------------
Error Type = System.IndexOutOfRangeException
Error Message = Index was outside the bounds of the array.
Error Source = CommandLine
Error Site = Boolean MoveNext()
Error occurred = at CommandLine.Core.Tokenizer.<TokenizeShortName>d__16.MoveNext()
at System.Linq.Enumerable.<SelectManyIterator>d__22`3.MoveNext()
at System.Linq.Enumerable.WhereEnumerableIterator`1.MoveNext()
at System.Linq.Enumerable.<ExceptIterator>d__72`1.MoveNext()
at System.Linq.Enumerable.WhereSelectEnumerableIterator`2.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.SingleOrDefault[TSource](IEnumerable`1 source, Func`2 predicate)
at CommandLine.Core.OptionMapper.<>c__DisplayClassa.<MapValues>b__0(SpecificationProperty pt)
at System.Linq.Enumerable.<>c__DisplayClass7_0`3.<CombineSelectors>b__0(TSource x)
at System.Linq.Enumerable.WhereSelectEnumerableIterator`2.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.WhereEnumerableIterator`1.MoveNext()
at System.Linq.Enumerable.Aggregate[TSource,TAccumulate](IEnumerable`1 source, TAccumulate seed, Func`3 func)
at CommandLine.Core.ReflectionExtensions.SetProperties[T](T instance, IEnumerable`1 specProps, Func`2 predicate, Func`2 selector)
at CommandLine.Core.InstanceBuilder.Build[T](Func`1 factory, Func`3 tokenizer, IEnumerable`1 arguments, StringComparer nameComparer, CultureInfo parsingCulture)
at CommandLine.Parser.<>c__DisplayClass4`1.<ParseArguments>b__2()
at CommandLine.Parser.MakeParserResult[T](Func`1 parseFunc, ParserSettings settings)
at CommandLine.Parser.ParseArguments[T](Func`1 factory, String[] args)
at CommandLine.Parser.ParseArguments[T](String[] args)
at A.k.A(String[] , MainForm )
at A.k.A(String[] )
OS Version = Microsoft Windows NT 6.1.7601 Service Pack 1
Total Physical Memory = 15.9 GB
Available Physical Memory = 6 GB
Date Time = 27/01/2016 20:22:45
User Name = 3DGoD-PC\3DGoD
Application Startup Path = C:\Program Files (x86)\MPDN
Application Executable Path = C:\Program Files (x86)\MPDN\MediaPlayerDotNet.exe
CurrentDirectory = C:\Users\3DGoD\AppData\Local\Temp\nw9588_19391
SystemDirectory = C:\Windows\system32
Ran As Admin = False
HasShutdownStarted = False
MachineName = 3DGOD-PC
ProcessorCount = 4
LogicalDrives = C:\, R:\
EnvironmentVariables = PROCESSOR_ARCHITEW6432 = AMD64; COMPUTERNAME = 3DGOD-PC; CHROME_BREAKPAD_PIPE_NAME = \\.\pipe\GoogleCrashServices\S-1-5-21-516795912-4234960728-3803919610-1000; HOMEPATH = \Users\3DGoD; SHIM_MCCOMPAT = 0x810000001; NVIDIAWHITELISTED = 0x01; PROCESSOR_REVISION = 2a07; PATHEXT = .COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH;.MSC; TMP = C:\Users\3DGoD\AppData\Local\Temp; TEMP = C:\Users\3DGoD\AppData\Local\Temp; LOCALAPPDATA = C:\Users\3DGoD\AppData\Local; PUBLIC = C:\Users\Public; C_EM64T_REDIST11 = C:\Program Files (x86)\Common Files\Intel\Shared Files\cpp\; USERDOMAIN = 3DGoD-PC; ProgramFiles(x86) = C:\Program Files (x86); CG_BIN_PATH = C:\Program Files (x86)\NVIDIA Corporation\Cg\bin; Path = C:\Program Files (x86)\Razer Chroma SDK\bin;C:\Program Files\Razer Chroma SDK\bin;C:\Program Files (x86)\Common Files\Intel\Shared Libraries\redist\ia32_win\compiler;C:\Program Files (x86)\Common Files\Intel\Shared Libraries\redist\ia32\compiler;C:\Program Files (x86)\Common Files\Intel\Shared Files\cpp\bin\Intel64;C:\ProgramData\Oracle\Java\javapath;C:\Program Files (x86)\Common Files\Microsoft Shared\Windows Live;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0\;C:\Program Files\WIDCOMM\Bluetooth Software\;C:\Program Files\WIDCOMM\Bluetooth Software\syswow64;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files\Prio;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files (x86)\NVIDIA Corporation\Cg\bin;C:\Program Files (x86)\NVIDIA Corporation\Cg\bin.x64;C:\Program Files (x86)\NVIDIA Corporation\PhysX\Common;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files\Common Files\Microsoft Shared\Windows Live;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x86;C:\Program Files (x86)\Intel\OpenCL SDK\2.0\bin\x64;C:\Program Files (x86)\Skype\Phone\;C:\Program Files (x86)\Livestreamer;C:\Program Files\Common Files\Microsoft Shared\Windows Live; PROCESSOR_LEVEL = 6; PROCESSOR_IDENTIFIER = Intel64 Family 6 Model 42 Stepping 7, GenuineIntel; windows_tracing_flags = 3; PSModulePath = C:\Windows\system32\WindowsPowerShell\v1.0\Modules\; NUMBER_OF_PROCESSORS = 4; CommonProgramFiles = C:\Program Files (x86)\Common Files; QT_DEVICE_PIXEL_RATIO = auto; CG_LIB64_PATH = C:\Program Files (x86)\NVIDIA Corporation\Cg\lib.x64; asl.log = Destination=file; ProgramData = C:\ProgramData; ProgramW6432 = C:\Program Files; ProgramFiles = C:\Program Files (x86); CG_BIN64_PATH = C:\Program Files (x86)\NVIDIA Corporation\Cg\bin.x64; FP_NO_HOST_CHECK = NO; SystemRoot = C:\Windows; INTEL_DEV_REDIST = C:\Program Files (x86)\Common Files\Intel\Shared Libraries\; SESSIONNAME = Console; CommonProgramW6432 = C:\Program Files\Common Files; LOGONSERVER = \\3DGOD-PC; USERPROFILE = C:\Users\3DGoD; CommonProgramFiles(x86) = C:\Program Files (x86)\Common Files; APPDATA = C:\Users\3DGoD\AppData\Roaming; HOMEDRIVE = C:; windows_tracing_logfile = C:\BVTBin\Tests\installpackage\csilogfile.log; USERNAME = 3DGoD; PROCESSOR_ARCHITECTURE = x86; OS = Windows_NT; ComSpec = C:\Windows\system32\cmd.exe; SystemDrive = C:; windir = C:\Windows; ALLUSERSPROFILE = C:\ProgramData
RuntimeVersion = 4.0.30319.42000
Thread Count = 8
Handle Count = 258
VM Size = 246.6 MB
Peak VM Size = 255.1 MB
Working Set Size = 29.3 MB
Max Working Set Size = 1.3 MB
Min Working Set Size = 200 KB
Modules = MediaPlayerDotNet.exe, ntdll.dll, MSCOREE.DLL, KERNEL32.dll, KERNELBASE.dll, ADVAPI32.dll, msvcrt.dll, sechost.dll, RPCRT4.dll, SspiCli.dll, CRYPTBASE.dll, mscoreei.dll, SHLWAPI.dll, GDI32.dll, USER32.dll, LPK.dll, USP10.dll, IMM32.DLL, MSCTF.dll, nvinit.dll, VERSION.dll, detoured.dll, nvd3d9wrap.dll, SETUPAPI.dll, CFGMGR32.dll, OLEAUT32.dll, ole32.dll, DEVOBJ.dll, nvdxgiwrap.dll, prio32.dll, clr.dll, MSVCR120_CLR0400.dll, mscorlib.ni.dll, ShellEh552.dll, K_FPS.dll, dbghelp.dll, d3dx9_43.dll, d3dx10_43.dll, d3dx11_43.dll, D3DX8Wrapper.dll, D3DCOMPILER_43.dll, WINMM.dll, gdiplus.dll, PSAPI.DLL, OPENGL32.dll, GLU32.dll, DDRAW.dll, DCIMAN32.dll, dwmapi.dll, COMDLG32.dll, COMCTL32.dll, SHELL32.dll, EasyHook32.dll, DSOUND.dll, POWRPROF.dll, System.ni.dll, CRYPTSP.dll, rsaenh.dll, System.Core.ni.dll, MediaPlayerDotNet.ni.exe, System.Drawing.ni.dll, System.Windows.Forms.ni.dll, Mpdn.Core.ni.dll, uxtheme.dll, System.Configuration.ni.dll, System.Xml.ni.dll, nlssorting.dll, WindowsCodecs.dll, BtMmHook.dll, RpcRtRemote.dll, CommandLine.ni.dll, clrjit.dll, Microsoft.ExceptionMessageBox.ni.dll, Microsoft.VisualBasic.ni.dll, diasymreader.dll


===================================

Index was outside the bounds of the array. (CommandLine)

------------------------------
Program Location:

at CommandLine.Core.Tokenizer.<TokenizeShortName>d__16.MoveNext()
at System.Linq.Enumerable.<SelectManyIterator>d__22`3.MoveNext()
at System.Linq.Enumerable.WhereEnumerableIterator`1.MoveNext()
at System.Linq.Enumerable.<ExceptIterator>d__72`1.MoveNext()
at System.Linq.Enumerable.WhereSelectEnumerableIterator`2.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.SingleOrDefault[TSource](IEnumerable`1 source, Func`2 predicate)
at CommandLine.Core.OptionMapper.<>c__DisplayClassa.<MapValues>b__0(SpecificationProperty pt)
at System.Linq.Enumerable.<>c__DisplayClass7_0`3.<CombineSelectors>b__0(TSource x)
at System.Linq.Enumerable.WhereSelectEnumerableIterator`2.MoveNext()
at System.Linq.Enumerable.<ConcatIterator>d__58`1.MoveNext()
at System.Linq.Enumerable.WhereEnumerableIterator`1.MoveNext()
at System.Linq.Enumerable.Aggregate[TSource,TAccumulate](IEnumerable`1 source, TAccumulate seed, Func`3 func)
at CommandLine.Core.ReflectionExtensions.SetProperties[T](T instance, IEnumerable`1 specProps, Func`2 predicate, Func`2 selector)
at CommandLine.Core.InstanceBuilder.Build[T](Func`1 factory, Func`3 tokenizer, IEnumerable`1 arguments, StringComparer nameComparer, CultureInfo parsingCulture)
at CommandLine.Parser.<>c__DisplayClass4`1.<ParseArguments>b__2()
at CommandLine.Parser.MakeParserResult[T](Func`1 parseFunc, ParserSettings settings)
at CommandLine.Parser.ParseArguments[T](Func`1 factory, String[] args)
at CommandLine.Parser.ParseArguments[T](String[] args)
at A.k.A(String[] , MainForm )
at A.k.A(String[] )



SORRY ABOUT THE SIZE SPOILER TAGS DIDNT WORK :s

[Zachs]

And what was the actual command line parameters you used?

[louiebh]

And what was the actual command line parameters you used?

https://github.com/bastimeyer/livestreamer-twitch-gui

i used the custom video player parameter here

and when i tried the direct cmd i input:

livestreamer twitch.tv/doublelift source

while mpdn was set to default player in the text file

[Shiandow]

The new downscaler should be available in the next release of the MPDN extensions.

[burfadel]

Any plans on using SharpDX 3.0 now that it has been officially released? Direct3D 12 support would be nice :). I guess the only disadvantage is the drop of DirectX 10 support, but for cards not supporting DirectX 11 there's still DirectX 9 support so all good!

[huhn]

and more important what would be won?

[The Dark Lord]

I'm having problem with playing some MKVs mostly those downloaded from Youtube.

MPDN is rendering the video at average of 15 frames, which of course translates to massive stuttering.

https://i.imgur.com/3aJzfke.jpg

as you can see there are no dropped or delayed frames.
MPC-HC and other players like VLC have no problem rendering these files at 23 frames.

My specs:

win 10 64bit
i7- 4770K
Geforce 970 (Driver 361.43)
Latest MPDN 32-bit

[Anima123]

The new downscaler should be available in the next release of the MPDN extensions.

Right on this weekend?

[Shiandow]

Right on this weekend?

With any luck, yes. At any rate it'll be soon.

The code's already on Github, you could also copy that to the Extensions folder and use it that way. At least, I think that should work.

[Zachs]

I'm away on holiday at the moment though. Buy I have my laptop with be so hopefully I'll find some time to do a release later today.

[louiebh]

I'm away on holiday at the moment though. Buy I have my laptop with be so hopefully I'll find some time to do a release later today.

did the livestreamer issue i had work for you?

[Zachs]

Haven't had time to try it yet. But have your tried YouTube-dl? Does that work?

[louiebh]

Haven't had time to try it yet. But have your tried YouTube-dl? Does that work?

i saw it mentioned earlier in the thread when i good livestreamer and mpdn but I had literally no idea how to install it or get it working and it seemed from what i could see, be a way to download videos then watch them later rather than stream them directly through the browser?

but no im somwhat savvy but this was beyond me could you point me the github/mainpage didnt make any sense on how to install/work.

[huhn]

this is how it works: http://forum.doom9.org/showthread.php?p=1739563&highlight=youtube#post1739563

livestreamer a easier way to start livestreams but check youtube-dl first.

[ryrynz]

I was kinda expecting something amazing from the new SSIM downscaler but there wasn't that much difference when compared to lanczos 12AR IMO. It looks nice and all and I'd probably use it but it's no night and day difference.

[Shiandow]

I was kinda expecting something amazing from the new SSIM downscaler but there wasn't that much difference when compared to lanczos 12AR IMO. It looks nice and all and I'd probably use it but it's no night and day difference.

Thanks for the feedback, shame it wasn't as good as you thought it was going to be. Is there anything in particular that you think could be better?

FWIW something like 12 taps Lanczos is already pretty sharp to begin with, so in most case you shouldn't really need much additional sharpness (except if you downscale a lot). That said, in theory SSIM should be somewhat more intelligent, it adds exactly* the right amount of sharpness instead of just a lot of sharpness all the time. It could be sharper when downscaling very fine details, or for higher scaling factors. The SSIM downscaler will also be a bit smoother on less detailed regions, but that might not be particularly useful (although it does suppress source artefacts)*.

*Your mileage may vary.

[Anima123]

SSIM has it's own character for sure, you need to check several times to find the optimized settings working together with e.g., image doublers.

That is why I am more curious than ever, what it can produce with SuperRes or whatever you might call.

Edit: Right now I am using the following render chain, which seems to benefit from SSIM quite a lot, at least from 576p -> 1080p:
<ListOfPresetOfIFilterIRenderScript xmlns:yaxlib="http://www.sinairv.com/yaxlib/">
<PresetOfIFilterIRenderScript Name="SuperChromaRes" Guid="17c2c194-4a30-4355-bbcb-1dec7093d383">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperChromaResUi">
<Settings>
<Passes>2</Passes>
<Strength>1</Strength>
<Softness>0.25</Softness>
<Prescaler>True</Prescaler>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="SuperRes" Guid="1a020bec-87ad-4f83-b0dd-545e0ffdc9b6">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SuperRes.SuperResUi">
<Settings>
<PrescalerGroup>
<SelectedIndex>0</SelectedIndex>
<Hotkey />
<Options>
<PresetOfIFilterIRenderScript Name="Super-xBR" Guid="418b3ba0-af29-4724-9267-16461ed01a39">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Hylian.SuperXbr.SuperXbrUi">
<Settings>
<EdgeStrength>1</EdgeStrength>
<Sharpness>1</Sharpness>
<FastMethod>False</FastMethod>
<ThirdPass>True</ThirdPass>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="NEDI" Guid="91aa2996-b2fe-42d1-be0d-1a3479851567">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.Nedi.NediScaler">
<Settings>
<AlwaysDoubleImage>False</AlwaysDoubleImage>
<ForceCentered>False</ForceCentered>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="NNEDI3" Guid="316bb7d2-3886-4db3-b0b7-512615fdd76c">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.NNedi3.NNedi3Scaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
<CodePath>ScalarMad</CodePath>
<Structured>False</Structured>
<ChromaScalers />
<ChromaScalerGuid>00000000-0000-0000-0000-000000000000</ChromaScalerGuid>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="OpenCL NNEDI3" Guid="c92fad62-253a-4bfd-830f-a230111af648">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.OclNNedi3.OclNNedi3Scaler">
<Settings>
<Neurons1>Neurons16</Neurons1>
<Neurons2>Neurons16</Neurons2>
<ChromaScalers />
<ChromaScalerGuid>00000000-0000-0000-0000-000000000000</ChromaScalerGuid>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
</Options>
</PrescalerGroup>
<Passes>2</Passes>
<Strength>1</Strength>
<Softness>0.05</Softness>
<HQdownscaling>True</HQdownscaling>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="Resizer" Guid="06a38b00-8f92-4b6a-961e-12061ca127f0">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Mpdn.Resizer.ResizerScript">
<Settings>
<ResizerOption>TargetSize150Percent</ResizerOption>
</Settings>
</Script>
</PresetOfIFilterIRenderScript>
<PresetOfIFilterIRenderScript Name="SSIM downscaler" Guid="d447143f-8eaa-41fb-bacd-5cd022bb132a">
<Script yaxlib:realtype="Mpdn.Extensions.RenderScripts.Shiandow.SSimDownscaling.SSimDownscalerUi">
<Settings />
</Script>
</PresetOfIFilterIRenderScript>
</ListOfPresetOfIFilterIRenderScript>

[ryrynz]

Thanks for the feedback, shame it wasn't as good as you thought it was going to be. Is there anything in particular that you think could be better?


It just the content I was using. SSIM's strength is in the details, patterns and white areas. If you don't have this sort of content playing someone might instead prefer to run something else for a sharper picture. The content I was playing was computer generated so what I saw looked similar overall in sharpness to lanczos.


FWIW something like 12 taps Lanczos is already pretty sharp to begin with, so in most case you shouldn't really need much additional sharpness (except if you downscale a lot). That said, in theory SSIM should be somewhat more intelligent, it adds exactly* the right amount of sharpness instead of just a lot of sharpness all the time. It could be sharper when downscaling very fine details, or for higher scaling factors. The SSIM downscaler will also be a bit smoother on less detailed regions, but that might not be particularly useful (although it does suppress source artefacts)*.


Exactly right. Some might prefer other downscalers with less detailed content. I don't feel it's probably the one downscaler to rule them all but I have no doubt it'll be popular with general content, it's magic is in the finer details.

SSIM (https://mega.nz/#!jZhGnYRS!4UB4DneBZmje7i-f5C87C_Zs78p-F0UAkKGP_cVo2js)

Lanczos 16AR (https://mega.nz/#!SJIFSYyY!36OZd1GSndK2W1bT8FDhbaiFOIqCIzhzSQ14e4uk04g)

The broccoli, the quilt patterns and the shreds inside the crayfish are nice areas of improvement over Lanczos 16AR.

[madshi]

SSIM (https://mega.nz/#!jZhGnYRS!4UB4DneBZmje7i-f5C87C_Zs78p-F0UAkKGP_cVo2js)

Lanczos 16AR (https://mega.nz/#!SJIFSYyY!36OZd1GSndK2W1bT8FDhbaiFOIqCIzhzSQ14e4uk04g)
It's always helpful to include the original size image.

[Thunderbolt8]

It would also preferable to use an imagehost for the two pics instead of having to go through a download process.

[ryrynz]

It would also preferable to use an imagehost for the two pics instead of having to go through a download process.

I tried for over an hr last night.. Every image provider either resized them wouldn't accept them or changed the brightness.. Any recommendations?

It's always helpful to include the original size image.

I'll link the source later. I don't have a 4k screen.


http://4ksamples.com/

Screenshot was taken from the Chimei inn sample.

[Thunderbolt8]

did you try imagebam or someimage?

[ryrynz]

did you try imagebam or someimage?

For some reason I just cannot explain the Lanzcos image gets brighter when uploaded to various image uploading sites.

If someone could tell me why this happens I'd really appreciate it.

Imagebam link (http://113.imagebam.com/download/TwgwX69VRZAoY9ArAchQtA/46291/462902090/lanc16ar60.png)

Mega (https://mega.nz/#!SJIFSYyY!36OZd1GSndK2W1bT8FDhbaiFOIqCIzhzSQ14e4uk04g)

[eX``]

Hi,

I wanted to have a look / test SSIM but I got trouble using it for a simple reason: it's not showing in MPDN. When I go to Program Files / MPDN, the folder is in RenderScripts. Though it's not present in App data.

Thank you very much.

[Shiandow]

So you copied the "SSIMdownscaler" folder? Did you also copy the "SSimDownscaler.cs" file? That one should be in the RenderScripts folder as well.

[Zachs]

I've released the binary version yesterday. If you updated the extensions you should automatically have SSIM.

[Thunderbolt8]

For some reason I just cannot explain the Lanzcos image gets brighter when uploaded to various image uploading sites.did you take these images yourself? perhaps they get darker during the process for some reason.

edit: well guess that does not make much sense if they still get brighter afterwards when uploading. still, better check to rule anything out.

[eX``]

Nah I just updated. And for this was not working, I downloaded the .exe and reinstalled it, but same issue. Should I wipe it all and reinstall?
edit: I reread your message and checked, I don't have "SSimDownscaler.cs".
edit2: Dled both standalone and couldn't find the "SSimDownscaler.cs" I think I'm missing something here =)

[Shiandow]

Probably best to remove everything and reinstall both MPDN and the MPDN extensions.

[Thunderbolt8]

For some reason I just cannot explain the Lanzcos image gets brighter when uploaded to various image uploading sites.Ive heard that there could be brightness changes with some programs when resizing. try to use another program.

[ryrynz]

did you take these images yourself? perhaps they get darker during the process for some reason.

edit: well guess that does not make much sense if they still get brighter afterwards when uploading. still, better check to rule anything out.

Ive heard that there could be brightness changes with some programs when resizing. try to use another program.

No resizing was done, this is a straight Printscreen to paint and a save in PNG, both images are the same size.

And I've found it's Chrome that's causing the image to brighten.. Edge doesn't do it. Guess I got a bug to file.

*Edit* Bug confirmed, got the Chromium guys on to it.

[Shiandow]

Has anyone else tried EWA Sinc-Jinc (6 taps, anti-ringing set to 1.0) as a prescaler for SuperRes? I have yet to notice any obvious downsides, so I'm thinking of making it the new default prescaler.

[Anima123]

Has anyone else tried EWA Sinc-Jinc (6taps, nti-ringing set to 1.0) as a prescaler for SuperRes? I have yet to notice any obvious downsides, so I'm thinking of making it the new default prescaler.

I have tried it quite some time ago, it's good. Will do some more tests on it if needed.

[Shiandow]

Well, I was just wondering if there were any obvious downside to Sinc-Jinc. You might want to wait a bit before you do any rigorous testing, I've got some other small improvements for SuperRes that might change things.

[Anima123]

Well, I was just wondering if there were any obvious downside to Sinc-Jinc. You might want to wait a bit before you do any rigorous testing, I've got some other small improvements for SuperRes that might change things.

Just let me know once the improved SuperRes is ready for testing.

[aufkrawall]

Has anyone else tried EWA Sinc-Jinc (6 taps, anti-ringing set to 1.0) as a prescaler for SuperRes? I have yet to notice any obvious downsides, so I'm thinking of making it the new default prescaler.
I was already using Sinc-Hann as a prescaler for SuperRes for some time and didn't notice any drawback with low scaling factors towards super-xbr.
With madVR, I use its Jinc algorithm before SuperRes it looks fine to me as well.

[Anima123]

I was already using Sinc-Hann as a prescaler for SuperRes for some time and didn't notice any drawback with low scaling factors towards super-xbr.
With madVR, I use its Jinc algorithm before SuperRes it looks fine to me as well.

Sinc-Hann is the most sharpest upscaling algorithm in EWA scalers, sometimes it's too sharp for my preference.

[Shiandow]

Anyone else having trouble getting links to work with drag/drop? With Firefox 44 they don't seem to play automatically anymore. Weirdly enough they still work on chrome, and the problem seems to have started just after the last firefox update.

So, I'd like to know if it's just me, or if anyone else is seeing the same thing, and if I'm correct in assuming that the problem started with the update from Firefox v43 to v44.

[scott1256ca]

Just installed this as a possible replacement for mpc-hc. Is there an equivalent to 4x3 in mpc-hc? searching for 4x3 in this thread had no hits. stretch had a few but nothing relevent to me. I have an anamorphic lens so I need to stretch the image vertically to remove the black bars. I tried 4x3 under options, but that doesn't remove the top and bottom black bars.
Thx

[l0rdraiden]

Is there any guide explaning how to configure MPDN with SVP
and on how to configure MPDN itself? in particular the render script and upscaler sections.

Are the extensions / renders used in MPDN the same than in mpv? https://mpv.io/

What is the advantage of MPDN over Madvr?

Is MPDN close source? why?

[scott1256ca]

ok, I think I figured out the 4x3 thing. Too bad you have to stop and restart for it to take effect. Another thing, in full screen, there is a slider at the bottom, but the mouse pointer doesn't show up, which makes it hard to know where to click. Is there anything I can do to change that?

[Thunderbolt8]

Shiandow, will it be possible to use SSIM algorithm as standalone filter e.g. for downscaling/rezising encodes when doing x264 encoding? (avisynth usage?)

[Anima123]

Any progress on the SuperRes lately, Shiandow?

[Shiandow]

Shiandow, will it be possible to use SSIM algorithm as standalone filter e.g. for downscaling/rezising encodes when doing x264 encoding? (avisynth usage?)

Not unless someone ports the algorithm to avisynth, or if someone manages to write an avisynth wrapper for MPDN extensions.

Any progress on the SuperRes lately, Shiandow?

I managed to improve performance a bit while also improving the image quality slightly (I think). Combined with switching to Sinc-Jinc for the default prescaler this should give a decent increase in performance.

[Zachs]

Not unless someone ports the algorithm to avisynth, or if someone manages to write an avisynth wrapper for MPDN extensions.

The latter has been on my radar for some time. It would be quite easy to make an Avisynth wrapper - I just need some free time to work on it.

[Thunderbolt8]

The latter has been on my radar for some time. It would be quite easy to make an Avisynth wrapper - I just need some free time to work on it.would be great, thanks!

[XRyche]

I have a question that has probably been asked quite a few times before but..................................................................................has there been a script made for user configurable keyboard shortcuts? So MCE remotes can be used to control MPDN.

[trandoanhung1991]

Any idea how I can debug decoder stalls?

Just as I said that I got a Runtime Error while running MPDN.

[cascardian]

I have another bug to report, though it is extremely minor. Watching a video against the background of a running Twitch stream (Flash video with HTML5 controls) in Chrome will lead to occasional frame hitching/stuttering in MPDN. This is on DX11, using the new windowed mode rendering path. Switching that option off fixes the problem. Also happens in fullscreen, though not in exclusive.

I just checked: The problem does not occur with Youtube's HTML5 video in the background. In case it helps, here is Chrome's GPU log:
Graphics Feature Status
Canvas: Hardware accelerated
Flash: Hardware accelerated
Flash Stage3D: Hardware accelerated
Flash Stage3D Baseline profile: Hardware accelerated
Compositing: Hardware accelerated
Multiple Raster Threads: Enabled
Native GpuMemoryBuffers: Software only. Hardware acceleration disabled
Rasterization: Software only. Hardware acceleration disabled
Video Decode: Hardware accelerated
Video Encode: Hardware accelerated
WebGL: Hardware accelerated

Driver Bug Workarounds
clear_uniforms_before_first_program_use
disable_discard_framebuffer
exit_on_context_lost
force_cube_complete
scalarize_vec_and_mat_constructor_args
texsubimage_faster_than_teximage
use_current_program_after_successful_link

Problems Detected
Always call glUseProgram after a successful link to avoid a driver bug: 349137
Applied Workarounds: use_current_program_after_successful_link
Some drivers are unable to reset the D3D device in the GPU process sandbox
Applied Workarounds: exit_on_context_lost
TexSubImage is faster for full uploads on ANGLE
Applied Workarounds: texsubimage_faster_than_teximage
Clear uniforms before first program use on all platforms: 124764, 349137
Applied Workarounds: clear_uniforms_before_first_program_use
Always rewrite vec/mat constructors to be consistent: 398694
Applied Workarounds: scalarize_vec_and_mat_constructor_args
ANGLE crash on glReadPixels from incomplete cube map texture: 518889
Applied Workarounds: force_cube_complete
Framebuffer discarding can hurt performance on non-tilers: 570897
Applied Workarounds: disable_discard_framebuffer
Accelerated rasterization has been disabled, either via about:flags or command line.
Disabled Features: rasterization
Native GpuMemoryBuffers have been disabled, either via about:flags or command line.
Disabled Features: native_gpu_memory_buffers

[l0rdraiden]

Is there any guide explaning how to configure MPDN with SVP
and on how to configure MPDN itself? in particular the render script and upscaler sections.

Are the extensions / renders used in MPDN the same than in mpv? https://mpv.io/

What is the advantage of MPDN over Madvr?

Is MPDN close source? why?

Some help please?

[XRyche]

I have a question that has probably been asked quite a few times before but..................................................................................has there been a script made for user configurable keyboard shortcuts? So MCE remotes can be used to control MPDN.

shrugs

[Zachs]

I have a question that has probably been asked quite a few times before but..................................................................................has there been a script made for user configurable keyboard shortcuts? So MCE remotes can be used to control MPDN.

No not yet.

Any idea how I can debug decoder stalls?

Just as I said that I got a Runtime Error while running MPDN.

This usually means your CPU isn't fast enough (usually in the case of high bitrake 4K HEVC) or the data isn't being supplied fast enough to the decoder - e.g. If you're streaming high bitrate footage over wifi or via the internet, you could get stalls.

Runtime Error tends to happen when the unmanaged part of MPDN (i.e. DirectShow filters) have corrupted the heap - there's really not much that could be done.

I have another bug to report, though it is extremely minor. Watching a video against the background of a running Twitch stream (Flash video with HTML5 controls) in Chrome will lead to occasional frame hitching/stuttering in MPDN. This is on DX11, using the new windowed mode rendering path. Switching that option off fixes the problem. Also happens in fullscreen, though not in exclusive.

In windowed mode, the GPU has to be shared with every other app that requires it - and if one demands max priority and steals too much time from MPDN, you'll get a problem as it won't be able to present when it needs to.

Is there any guide explaning how to configure MPDN with SVP
and on how to configure MPDN itself? in particular the render script and upscaler sections.

Are the extensions / renders used in MPDN the same than in mpv? https://mpv.io/

What is the advantage of MPDN over Madvr?

Is MPDN close source? why?

Guide to use SVP with MPDN - :search:
Render script and upscaler configuration - :search:

I don't know much about MPV other than they took code from our GitHub repository, removed the LGPL license and then calling it their own. The issue was sorted after reminding them about the licensing issue.

AFAIK, madVR and mpv use much of the code from MPDN Extensions.

MPDN is closed source but the extensions where all the high-end scalers and filters are implemented, are LGPL v3. Why is MPDN closed source - as answered before multiple times - it's just is because I have to.

[foozoor]

Could you if possible fix this really annoying bug?
https://mpdn.zachsaw.com/bug/view.php?id=6

[Braum]

Hi ! I need some advice concerning the render script configuration.

I only play Bluray in 1920x1080
My screen resolution is 1680x1050
My gpu is an AMD 7980

Here is my downscaling configuration:

SuperChromaRes - 3 passes, strength 1, softness 0
SSIM downscaler
Deband - power 0,3
Image Processor - FineSharp tuned to my liking.

Am I doing it right ?

[Zachs]

Could you if possible fix this really annoying bug?
https://mpdn.zachsaw.com/bug/view.php?id=6

If you are using the playlist and have it set to repeat playback it will show two sets of filters in the system tray. Why would you assume it is " cause double performance consumption"? The other one, while loaded, will remain idle.

Disable "Preload next file" in the playlist if you don't want this feature.

[foozoor]

If you are using the playlist and have it set to repeat playback it will show two sets of filters in the system tray. Why would you assume it is " cause double performance consumption"? The other one, while loaded, will remain idle.

Disable "Preload next file" in the playlist if you don't want this feature.

I am not the one who reported the "bug" but I am using lavfilters megamix setup from the imouto.my blog.

However I had performance issues but it was maybe because of xysubfilter, I need to test it again.

[mecedo]

Can you help me with the following issue? I can't enable Hardware Acceleration in LAV Video in MPDN Player. If I set "DXVA Native" in LAV Video then "Available" is displayed and Active Decoder is "avcodec". In Potplayer with LAV Video which is configured in the same way as in MPDN displayed Active Decoder is "dxva2n" and hardware acceleration works.
Is it possible to debug or log MPDN Player to find out why hardware acceleration isn't applied?
Checked configurations: Intel i3 4130T (HD4400 video) with Win7 x32 and Win 7 x64.
Checked Intel video drivers: 10.18.10.3960 and 15.36.28.4332.
Any help will be appreciated.

[Zachs]

Can you help me with the following issue? I can't enable Hardware Acceleration in LAV Video in MPDN Player. If I set "DXVA Native" in LAV Video then "Available" is displayed and Active Decoder is "avcodec". In Potplayer with LAV Video which is configured in the same way as in MPDN displayed Active Decoder is "dxva2n" and hardware acceleration works.
Is it possible to debug or log MPDN Player to find out why hardware acceleration isn't applied?
Checked configurations: Intel i3 4130T (HD4400 video) with Win7 x32 and Win 7 x64.
Checked Intel video drivers: 10.18.10.3960 and 15.36.28.4332.
Any help will be appreciated.

MPDN doesn't support DXVA native. Other modes are fully supported.

[mecedo]

MPDN doesn't support DXVA native. Other modes are fully supported.

Thanks. Maybe you should mention it in description?
Here is useful explanation: LAV Video Configuration (http://wiki.team-mediaportal.com/1_MEDIAPORTAL_1/0_What's_New/1.3.x/1_UI_Related_Changes/Config_Related_Changes/Codecs_and_Renderer/LAV_Configuration/LAV_Video_Configuration)

And another question. Can MPDN use WASAPI as Audio renderer?

[ryrynz]

And another question. Can MPDN use WASAPI as Audio renderer?

Sanear uses WASAPI, I'd recommend using that, it's included in the installer.

Hopefully a new version of that isn't that far off, Alex has been committing some code recently.

[Zachs]

Quick update.

A new MPDN version will be released soon (later today or tomorrow) along with new extensions with plethora of changes (SuperRes/SuperChromaRes, SSIM etc).
Shiandow will do a quick write up on the changes later.

Someone requested a display gamma correction feature some time ago. It, in the form of a shader file to be used with Image Processor, has also made it into this next release.

Cheers.

[Shiandow]

Hi,

A few comments about the upcoming release.


Super-XBR has been updated by Hylian with an improved anti-ringing algorithm and tweaked parameters.
Image Processing no longer requires shaders to be in the 'ImageProcessingShaders' folder.
SSIM performance has been improved, and a limiter has been added to prevent it from amplifying details too much.
SuperRes / SuperChromaRes performance has improved with numerous tweaks to improve picture quality as well. Most of the tweaks are small but together they might make a significant difference. Also added 'Sinc-Jinc' as the default prescaler, since it is faster and sharper for low(ish) scaling factors (below 2~3), which makes it better for most practical purposes. Also added a fast preset for Super-Xbr that gives a good quality / performance ratio for higher scaling factors.
Bilateral now uses a more reconstructive method for retrieving chroma information from the luma, this method is still somewhat experimental and is prone to artefacts, but in some scenarios it blows away all other algorithms.
Lots of other tweaks and fixes (see the release notes / git repository for more details).


Anyone interested in developing may also be happy to hear that the framework has been improved quite a bit for this new release, also improving the readability of the RenderScripts (which should be convenient for people interested in porting them to other applications). Anyone not (yet) interested in developing will hopefully also see some of the benefits soon, in the form of less bugs and more features.

[foozoor]

Quick update.

A new MPDN version will be released soon (later today or tomorrow) along with new extensions with plethora of changes (SuperRes/SuperChromaRes, SSIM etc).
Shiandow will do a quick write up on the changes later.

Someone requested a display gamma correction feature some time ago. It, in the form of a shader file to be used with Image Processor, has also made it into this next release.

Cheers.

Hi,

A few comments about the upcoming release.


Super-XBR has been updated by Hylian with an improved anti-ringing algorithm and tweaked parameters.
Image Processing no longer requires shaders to be in the 'ImageProcessingShaders' folder.
SSIM performance has been improved, and a limiter has been added to prevent it from amplifying details too much.
SuperRes / SuperChromaRes performance has improved with numerous tweaks to improve picture quality as well. Most of the tweaks are small but together they might make a significant difference. Also added 'Sinc-Jinc' as the default prescaler, since it is faster and sharper for low(ish) scaling factors (below 2~3), which makes it better for most practical purposes. Also added a fast preset for Super-Xbr that gives a good quality / performance ratio for higher scaling factors.
Bilateral now uses a more reconstructive method for retrieving chroma information from the luma, this method is still somewhat experimental and is prone to artefacts, but in some scenarios it blows away all other algorithms.
Lots of other tweaks and fixes (see the release notes / git repository for more details).


Anyone interested in developing may also be happy to hear that the framework has been improved quite a bit for this new release, also improving the readability of the RenderScripts (which should be convenient for people interested in porting them to other applications). Anyone not (yet) interested in developing will hopefully also see some of the benefits soon, in the form of less bugs and more features.

You are amazing guys. ;)

[Zachs]

v2.46.0/v1.26.0 Released.

Please let us know if there's any issues.

Note: Your old audio script group config may throw up an error when you try to open the "Script Group" config dialog. To fix this, delete the %localappdata%\MediaPlayerDotNet\AudioScripts.?? (?? == 32/64/AnyCPU) folder and reconfigure your script chain.

This will be fixed in the next minor release. EDIT: Fixed in v1.26.1 (clears the config automatically instead of failing to bring up the config dialog altogether)

Cheers!

[cascardian]

Thank you for the newest version!

This is what I now get in DebugView on a fresh default config with any AudioScript enabled (I apologise if this is redundant for you):
00000002 5.56374979 [2344] [109346'021,584ms] [INFO] Spent 516ms building/loading extensions
00000003 6.08320141 [2344] Information: Create ComputeContext(806126304) in Thread(1).
00000004 6.08431005 [2344] Information: Create ComputeCommandQueue(634353200) in Thread(1).
00000005 6.69643497 [2344] Mono.Cecil.AssemblyResolutionException: Failed to resolve assembly: 'Mpdn.Core, Version=1.0.0.0, Culture=neutral, PublicKeyToken=null'
00000006 6.69643497 [2344] at Mono.Cecil.BaseAssemblyResolver.Resolve(AssemblyNameReference name, ReaderParameters parameters)
00000007 6.69643497 [2344] at Mono.Cecil.DefaultAssemblyResolver.Resolve(AssemblyNameReference name)
00000008 6.69643497 [2344] at ICSharpCode.Decompiler.Ast.Transforms.IntroduceUsingDeclarations.Run(AstNode compilationUnit)
00000009 6.69643497 [2344] at ICSharpCode.Decompiler.Ast.Transforms.TransformationPipeline.RunTransformationsUntil(AstNode node, Predicate`1 abortCondition, DecompilerContext context)
00000010 6.69643497 [2344] at Cudafy.Translator.CUDALanguage.DecompileType(TypeDefinition type, ITextOutput output, DecompilationOptions options)
00000011 6.69643497 [2344] at Cudafy.Translator.CudafyTranslator.DoCudafy(CudafyModule cm, Type[] types)
00000012 6.69643497 [2344] at Cudafy.Translator.CudafyTranslator.Cudafy(IEnumerable`1 props, Type[] types)
00000013 6.69643497 [2344] at Cudafy.Translator.CudafyTranslator.Cudafy(eArchitecture arch, Type[] types)
00000014 6.69643497 [2344] at Mpdn.Extensions.Framework.AudioHelpers.GetCudafyModule(Type[] types)
00000015 6.69643497 [2344] at Mpdn.Extensions.Framework.AudioHelpers.LoadAudioKernel(GPGPU gpu, Type type, Type[] types)
00000016 6.69643497 [2344] at Mpdn.Extensions.Framework.AudioChain.AudioProc.Initialize()
00000017 6.69762754 [2344] Error in AudioChainScript:
00000018 6.69762754 [2344]
00000019 6.69762754 [2344] Object reference not set to an instance of an object.
00000020 6.69762754 [2344]
00000021 6.69762754 [2344] ~
00000022 6.69762754 [2344] Stack Trace:
00000023 6.69762754 [2344] at Mpdn.Extensions.Framework.Chain.FilterChainScript`2.Execute()
00000024 6.70564222 [2344] Error in AudioChainScript:
00000025 6.70564222 [2344]
00000026 6.70564222 [2344] Object reference not set to an instance of an object.
00000027 6.70564222 [2344]
00000028 6.70564222 [2344] ~
00000029 6.70564222 [2344] Stack Trace:
00000030 6.70564222 [2344] at Mpdn.Extensions.Framework.Chain.FilterChainScript`2.Execute()
00000031 6.70576859 [2344] Error in AudioChainScript:
00000032 6.70576859 [2344]
00000033 6.70576859 [2344] Object reference not set to an instance of an object.
. . . and so on. It spams this.
00000311 7.86265707 [2344] Warning: ComputeContext(806126304) leaked!
00000312 7.86511326 [2344] Information: Dispose ComputeCommandQueue(634353200) in Thread(2).

[Zachs]

Thank you for the newest version!

This is what I now get in DebugView on a fresh default config with any AudioScript enabled (I apologise if this is redundant for you):

Ah I forgot to fix that.

Can you try it with the latest source version from GitHub?
I'll probably make another release once you've tested it out.

There's instructions on MPDN Extensions GitHub page on how to use the source version if you are not familiar with it.

** I've released extensions v1.26.1 which should fix the problem.

Let me know how you go.

[cascardian]

Works beautifully now, many thanks!

[Zachs]

Fantastic!

[ryrynz]

Tried updating via the player.

http://i.imgur.com/E3eoGJo.jpg

[Zachs]

@Belphemur any idea what's wrong?

[mecedo]

Can someone explain how to use Belphemur's subtitle drag-and-drop extension? I can't handle subtitles by mouse. I need the possibility to shift subtitles in vertical. The best option will be keyboard shortcuts (maybe it should be implemented in XySubFilter?)

[Shiandow]

Can someone explain how to use Belphemur's subtitle drag-and-drop extension? I can't handle subtitles by mouse. I need the possibility to shift subtitles in vertical. The best option will be keyboard shortcuts (maybe it should be implemented in XySubFilter?)

Are you trying to reposition subtitles by dragging them?

I'm pretty sure the subtitle drag / drop extension simply loads subtitle files when you drop them on MPDN's screen (similar to what you can do with video files and URLs)

I don't think the subtitles can be shifted at the moment, and I'm reasonably sure it can't be added as an extension.

You could try adding it as a feature request in MPDN's bug tracker (https://mpdn.zachsaw.com/bug/view_all_bug_page.php).

[mecedo]

Are you trying to reposition subtitles by dragging them?

Yes, you're right. My misunderstanding :)

[Anima123]

Shiandow, have you improved the soft algorithm within SuperRes in the latest version? I just found it was pleasing to set 'softness' to 0.05, together with the default SuperRes settings.

[Shiandow]

I don't recall any huge changes, but maybe it combines well with some of the other changes (perhaps the Sinc-Jinc prescaler)? Not sure if you're going to see much of a difference with softness set to 0.05 though.

[Belphemur]

Tried updating via the player.

http://i.imgur.com/E3eoGJo.jpg

Any content in particular that easily shows the flickering with Fluid Motion in previous releases?

We switched the MPDN release server to HTTPS, and the downloader included in the installer can't deal with it. I updated the installer as soon as I encountered the problem, but we need to release a new version of the extension to fix the problem.

For now, you need to download both installers manually:
MPDN X64 (https://mpdn.zachsaw.com/Latest/Installers/MediaPlayerDotNet_x64_Installer.exe)
MPDN X86 (https://mpdn.zachsaw.com/Latest/Installers/MediaPlayerDotNet_x86_Installer.exe)

Extensions (https://github.com/zachsaw/MPDN_Extensions/releases/download/1.26.1/MPDN-Extensions_v1.26.1_Installer.exe)

You need to run both of the installer, in any order.

[Anima123]

Use NNEDI3 (both version) together with SSIM result in a blackscreen, ctrl-K says there are error in render chain script.

Super-xBR is fine though. Once I thought it was the chroma algorithm within NNEDI3, but switch to another one doesn't fix the error.

Edit: BTW, I was using the latest render script commit 833bb8d.

Just checked with the latest release 1.26.1, the same error.

Edit2: It seems NNEDI3 is not cooperating with SSIM, other downscaling algorithms (didn't test them all) seem OK.

[aufkrawall]

Wow, the new super-xbr version is a huge improvement. It has far less problems with ringing than the old one, looks much cleaner.

I can confirm blackscreen issues with NNEDI3.

[Shiandow]

Figured out what causes the black screen. And added a method for detecting those kind of problems.

In future you'll see an "Uncompiled Filter." error, which probably doesn't tell you much, but I'll know where to look. This error will also trigger in cases where the problem isn't as visible, which might be a bit inconvenient but will ensure that these kind of problems are captured earlier.

[mlauzon]

I get a weird circular banding on HD live action movies & TV shows where it is colours of the scene, but it doesn't seem to happen on HD anime/cartoons or just SD cartoons.

How can I fix this, if possible? Because it doesn't happen while using other media players.

[Zachs]

I get a weird circular banding on HD live action movies & TV shows where it is colours of the scene, but it doesn't seem to happen on HD anime/cartoons or just SD cartoons.

How can I fix this, if possible? Because it doesn't happen while using other media players.
Are you using any render scripts? What sort of circular banding? Some samples would help.

[mlauzon]

Are you using any render scripts? What sort of circular banding? Some samples would help.

I don't know how to use any of those, everything is default.

I can't provide any examples, I've tried taking a print screen screencap, and all I get after pasting it to MS Paint is a black screen.

[Zachs]

How about a sample so we could try and reproduce the problem?

There's not much to go on with what you described.

[Zachs]

Figured out what causes the black screen. And added a method for detecting those kind of problems.

In future you'll see an "Uncompiled Filter." error, which probably doesn't tell you much, but I'll know where to look. This error will also trigger in cases where the problem isn't as visible, which might be a bit inconvenient but will ensure that these kind of problems are captured earlier.

Released as v1.26.2. Cheers.

[mecedo]

I've observed the following issue: If computer starts in 23.976Hz mode and we switch to another mode, i.e. 30Hz then every 30fps movie played in windowed mode plays with horrible stuttering. MPDN in Ctrl+J stats shows Composition Rate 23.976Hz. After switching to fullscreen movie plays fluently. But not in windowed. The only way to fix it is to restart operating system. But if we change refreshrate to another mode then problem occurs again and only restart helps.
Did anyone observed a similar issue?

[aufkrawall]

When there is downscaling after SuperRes needed, it uses the scaler selected in MPDN main application:
http://abload.de/thumb/srdownscaleynsjp.png (http://abload.de/image.php?img=srdownscaleynsjp.png)
Is there a way to use a render script shader for this? Maybe add a SSIM checkbox in SuperRes settings?

Btw: I'm still not fully happy with the deband filter. Could someone please port MPV's deband filter to HLSL? It's already ported to ReShade, if this is any help:
https://github.com/crosire/reshade-shaders/blob/master/ReShade/Shaders/JPulowski/Deband.fx

[Shiandow]

When there is downscaling after SuperRes needed, it uses the scaler selected in MPDN main application:
http://abload.de/thumb/srdownscaleynsjp.png (http://abload.de/image.php?img=srdownscaleynsjp.png)
Is there a way to use a render script shader for this? Maybe add a SSIM checkbox in SuperRes settings?


You can change the prescaler to "SuperXbr + SSIM" to achieve that.


Btw: I'm still not fully happy with the deband filter. Could someone please port MPV's deband filter to HLSL? It's already ported to ReShade, if this is any help:
https://github.com/crosire/reshade-shaders/blob/master/ReShade/Shaders/JPulowski/Deband.fx

In what way is MPV's debanding filter better than MPDN's?

[aufkrawall]

You can change the prescaler to "SuperXbr + SSIM" to achieve that.

I can't find it:
http://abload.de/thumb/srnds1m.png (http://abload.de/image.php?img=srnds1m.png)
It's not in the list of all available scalers either.


In what way is MPV's debanding filter better than MPDN's?
It's working very well for me in games, while MPDN's deband filter often misses quite some banding in videos compared to madVR. So I thought it might not hurt to have the option to try it.

Btw: Now that super-xbr has far less problems with ringing, it might make sense to add it also as a chroma scaler?

[lanzorg]

Online streaming, pipe input, RAR archives, DVD playback


Does anyone know how to launch MPDN from Firefox or Chrome with streaming websites like YouTube?

[Shiandow]

I can't find it:
http://abload.de/thumb/srnds1m.png (http://abload.de/image.php?img=srnds1m.png)
It's not in the list of all available scalers either.

No, you'd have to add it; if you press "modify list" you can add whatever prescaler you want, which can be as complicated as you like. Since there are increasingly more options it's not really doable to add every single possible method, but I'm always open to better options to put in the default list.


It's working very well for me in games, while MPDN's deband filter often misses quite some banding in videos compared to madVR. So I thought it might not hurt to have the option to try it.

Well, the method is quite simple so I can give it a try. I might try to improve a bit on the method. FWIW I currently see the biggest banding problems when the source is pretty much 'broken', MadVR might be a bit more aggressive in those cases, but in exchange MPDN's method can be used at pretty much full strength without completely obliterating the image.


Btw: Now that super-xbr has far less problems with ringing, it might make sense to add it also as a chroma scaler?

I'd have to see if that's possible, but for now I'd rather focus on the new reconstructive Bilateral method, which seems rather promising (still need some kind of 'limiter' to prevent artefacts, but other than that the quality is impressive).

[aufkrawall]

No, you'd have to add it; if you press "modify list" you can add whatever prescaler you want, which can be as complicated as you like. Since there are increasingly more options it's not really doable to add every single possible method, but I'm always open to better options to put in the default list.

I can add SSIM downscaler there, but then SuperRes isn't used at all according to OSD and it's still downscaled with MPDN internal downscaler.


Well, the method is quite simple so I can give it a try. I might try to improve a bit on the method. FWIW I currently see the biggest banding problems when the source is pretty much 'broken', MadVR might be a bit more aggressive in those cases, but in exchange MPDN's method can be used at pretty much full strength without completely obliterating the image.

I think this would definitely make sense, since even "HQ" sources often suffer severe banding.
When you set the threshold of your deband filter to a very high value, this kind of banding isn't reduced as much as with madVR deband filter, but the image gets a softer look in general too.



I'd have to see if that's possible, but for now I'd rather focus on the new reconstructive Bilateral method, which seems rather promising (still need some kind of 'limiter' to prevent artefacts, but other than that the quality is impressive).
Looking forward to it. :)
With cartoons, there are currently often errors which are still easy to spot.

[foozoor]

I use my laptop/ultrabook very often and sometimes I watch some videos.
Could you share a preset for the Intel HD 530 integrated GPU?

[Shiandow]

I can add SSIM downscaler there, but then SuperRes isn't used at all according to OSD and it's still downscaled with MPDN internal downscaler.

The OSD system might be having some trouble (making it more failproof has been on my todo list for a while) or something is wrong. As far as I can tell the this script (http://pastebin.com/1BJZJ4jK)should do what you want.


Looking forward to it. :)
With cartoons, there are currently often errors which are still easy to spot.

That goes both ways by the way, if you know where to look 'non-reconstructive' methods cause errors that are quite easy to spot as well ;). One of the easiest to spot is chroma bleeding (http://screenshotcomparison.com/comparison/164757), but sometimes it's simply loss of detail (http://screenshotcomparison.com/comparison/164756).

[aufkrawall]

The OSD system might be having some trouble (making it more failproof has been on my todo list for a while) or something is wrong. As far as I can tell the this script (http://pastebin.com/1BJZJ4jK)should do what you want.

Oops, mea culpa. I totally didn't consider that a script chain can be added as a prescaler option.
Works well and OSD reports it correctly.


That goes both ways by the way, if you know where to look 'non-reconstructive' methods cause errors that are quite easy to spot as well ;). One of the easiest to spot is chroma bleeding (http://screenshotcomparison.com/comparison/164757), but sometimes it's simply loss of detail (http://screenshotcomparison.com/comparison/164756).
Agreed. However, imho such kind of errors is less obtrusive than e.g. bright colored dots between areas of different color, like in cartoons.

[Shiandow]

Agreed. However, imho such kind of errors is less obtrusive than e.g. bright colored dots between areas of different color, like in cartoons.

Those kinds of artefacts should have stopped mostly with the new version (1.26.2), if not then something is wrong. The remaining artefacts are mostly colour not following edges correctly (usually when there's a dark edge next to a red area).

[aufkrawall]

The remaining artefacts are mostly colour not following edges correctly (usually when there's a dark edge next to a red area).
Yep, that describes it well.

[huhn]

Well, the method is quite simple so I can give it a try. I might try to improve a bit on the method. FWIW I currently see the biggest banding problems when the source is pretty much 'broken', MadVR might be a bit more aggressive in those cases, but in exchange MPDN's method can be used at pretty much full strength without completely obliterating the image.

so the debanding algo is changed now?

in the past it was powerful but it could easily remove a huge deal of details unlike madVR f3kd based deband filter.

[Zachs]

so the debanding algo is changed now?

in the past it was powerful but it could easily remove a huge deal of details unlike madVR f3kd based deband filter.
You might've remembered it the wrong way around.

[huhn]

You might've remembered it the wrong way around.

i did the test my self and posted it here so now. it was at the time the deband filter was still in part of madVR that some time ago! i can search for the post or at least the disc if you like.

[Zachs]

i did the test my self and posted it here so now. it was at the time the deband filter was still in part of madVR that some time ago! i can search for the post or at least the disc if you like.
I can't remember the earliest implementation of the filter but it's been like this for a long time i.e. preserves more details but not as strong.

[huhn]

found the post.

http://forum.doom9.org/showpost.php?p=1725141&postcount=30693

it extremely terrible in motion.

i can cut a sample later but this will take time i don't even know which season this was.

[Shiandow]

found the post.

http://forum.doom9.org/showpost.php?p=1725141&postcount=30693

it extremely terrible in motion.

i can cut a sample later but this will take time i don't even know which season this was.

That does seem to be an earlier implementation of the algorithm.

Might want to give it another try with the new version on default settings. Although that image looks like it could be 'broken' in the sense that you have to choose between representing the source accurately, or removing the banding.

[huhn]

got this crash after first time starting with a complete reset of all setting.

TITLE: SharpDX Error
------------------------------

An unexpected error 'SharpDX.SharpDXException' has occurred.

------------------------------
ADDITIONAL INFORMATION:

HRESULT: [0x887A0005], Module: [SharpDX.DXGI], ApiCode: [DXGI_ERROR_DEVICE_REMOVED/DeviceRemoved], Message: The GPU device instance has been suspended. Use GetDeviceRemovedReason to determine the appropriate action.
(SharpDX)

------------------------------
BUTTONS:

&Abort
------------------------------


it was most likely running out of vram still this is no reason to crash i guess.

[huhn]

That does seem to be an earlier implementation of the algorithm.

Might want to give it another try with the new version on default settings. Although that image looks like it could be 'broken' in the sense that you have to choose between representing the source accurately, or removing the banding.

yes it is clearly changed.
and it doesn't harm the truck anymore.

this is with 1.0
http://abload.de/img/truckxoad8.png

[aufkrawall]

MPDN's deband with a very high strength looks a bit like a deblocking filter. I notice this when e.g. watching a news video, the face of the newsspeaker becomes a bit too blurry.

[Shiandow]

FYI I think the 'power' setting doesn't really do anything at the moment. Should probably fix that at some point.