madVR - high quality video renderer (GPU assisted)

[toomyzoom]

Yeah, I feel the same way. I have a feeling using HW decoding at all might actually decrease madvr's performance since it would take some load off of the GPU. There's really no point when my CPU is perfectly capable of decoding hi10p without issue. Am I right in thinking this way?

10-bit can't be decoded by hardware unless there is a new GPU designed to decode it.

[DragonQ]

dxva eats less power than cuvid, not sure about quicksync. Also AMD guys have no choice but to use dxva. But IMO all of them is a useless waste of time. Most people can't even tell if any of these things are working unless it is written in some OSD, which only proves its uselessness.
What? All of these hardware acceleration methods reduce CPU usage, which is usually preferable for most PCs and essential for HTPCs.

[mzso]

w00t, :thanks: for the new build!

oh, PS script need to be invoked from within the media player, gonna have to whine to the coder of PotP this time(PS: email sent ^^) :D

PS scripts work in 0-255 in MPC AFAIK, but I might have misunderstood what you wrote.

I'm curious whether support will be added in the next release or not.

Hopefully he didn't reply "use other player" this time. :)

(Like when I tried to explain that the jump to next subtitle jumps to the one after that sometimes. (Apparently when you jump from a subtitleless part of the movie))

[Keiyakusha]

What? All of these hardware acceleration methods reduce CPU usage, which is usually preferable for most PCs and essential for HTPCs.
This is how you want it to be, this is what advertisements teaches you. You need to wake up from your dreams. Very rarely people really benefit from it. Many believe that they do, but in fact they don't. They don't eve know if its working. They have no way to tell, because it is useless. Preferable? Essential? Bullshit. I'm not even saying about people who ask how they can tell if its working, and only after they was told where to get this info, they start to claim about huge improvements in something. Simply LOL.

[DragonQ]

This is how you want it to be, this is what advertisements teaches you. Very rarely people really benefit from it. Many believe that they do, but in fact they don't. They don't eve know if its working. They have no way to tell, because it is useless. Preferable? Essential? Bullshit.

Haha, I can tell it's working - if it isn't on, my HTPC can't play 1080i/25 video smoothly even when not much else is going on, let alone when transcoding for a web client. There's even weaker (and cheaper) systems that can be used for HTPCs thanks to DXVA (e.g. Atom/Zacate).

I don't need or want DXVA on my Core i7 desktop but that's not exactly the norm.

[leeperry]

I'm curious whether support will be added in the next release
yeah, the next PotP build will support PS scripts with mVR, also add "=" on top of the existing "<=" and ">=" framerate conditions in the automatic rules profiles and I'm trying to get him to disable the annoying fade effect on the D3D GUI and to allow CMD_POPUP_USER_MENU to show up because CMD_POPUPMENU_PANSCAN_PRESET / ID_CHAPTER_SEL_MENU / CMD_POPUPMENU_FAVORITE all work perfectly in D3D, so why not USER_MENU huh :scared:

Once PS script support will be operational with mVR, I hope madshi will manage to work his magic once more on the PS script BTB/WTW stuff(so we can get the same colors as with EVR/VMR9) then we'll be all set :cool:

And BTW, we will also be able to use CMD_POPUPMENU_VIDEO_SHADER ;)

But waiting is cool because I just found out that BenQ prolly lie on the 6X CWS spec of the W1070(the very similar W7000 runs it at 6X when using the "dynamic" mode preset but colors are out of control, OTOH it runs it at 4X in "cinema" mode (http://www.hometheaterhifi.com/projectors/projectors-reviews/benq-w7000-projector/page-4-on-the-bench.html)). Also, DLP's require some additional color filtering in order to reach Rec.709 so most of them have their green primary far off Rec.709...the built-in ISF CMS would only help to achieve a lower ΔE on saturations at 25/50/75/100% but this will be completely dependent on the gamut mapping to begin with...and I don't plan on spending a millions years using manual HCFR test patterns in madVR+PotP either, so I'll just wait for mVR to provide the same colors with PS scripts as in EVR/VMR9 and to read more reviews about the W1070, which despite all this still seems to be providing a killer bang/bucks ratio http://forum.slysoft.com/images/smilies/agreed.gif

My german is quite rusty but apparently you can't really expect a proper colorimetry >4X CWS anyway (http://www.cine4home.de/knowhow/DLP-Farbraeder/RGB-Farbraeder.htm). My previous 6 segments HC3100 was posterizing like hell at 5X, but it was just fine at 4X....so when they promise 6X CWS it's the usual commercial bs every manufacturer goes by these days.

[crotecun]

This is how you want it to be, this is what advertisements teaches you. You need to wake up from your dreams. Very rarely people really benefit from it. Many believe that they do, but in fact they don't. They don't eve know if its working. They have no way to tell, because it is useless. Preferable? Essential? Bullshit. I'm not even saying about people who ask how they can tell if its working, and only after they was told where to get this info, they start to claim about huge improvements in something. Simply LOL.

When I still had an Nvidia card I used to have CoreAVC. It provided a blue icon when it's not in use, and a green icon when CUDA acceleration is on.

If I look at the task manager when the icon is green there is a definite reduction in CPU usage.

And you're telling us this is all fluff...?

[schweinsz]

When will the madVR have the 3D support in your development estimation?

[Qaq]

IMO, HW decoding = no fan noise + playback is smoother (more or less).

[TheLion]

When will the madVR have the 3D support in your development estimation?

Well, it is not so easy as it also highly depends on the infrastructure:

Is there a proper MVC decoder available?

A uniform way to access 3D output on all capable GPUs is needed. Something the DX11.1 is trying to provide.

We also need nevs support with his LAV Splitter.

And 3D playback support of the media player as well.

[mindz]

Software decoding video and audio on my i7 only takes about 3% of the CPU when running madVR, i cant understand the need for DXVA from an i7 POV anyway. And i got an i7 + 5770 in a HTPC case (which is perfectly possible).

The thing im amazed about is how many people still run ancient hardware that needs DXVA and are willing to pay for PQ by doing DXVA scaling, when PC parts these days are so cheap..

[madshi]

Easy answers now. Difficult answers will come in the next couple of hours...

been trying to troubleshoot my problem with some mkv files and madvr (they play fine using EVR/CP). It's mostly some 1080p files but also some 720p files.

Anyway, I just remuxed a problem file to mp4 using mkv2mp4 and mpc-hc with madvr plays it without any problems!? I tried just remuxing the mkv using mkvtoolnix but the new mkv didn't play.

Any suggestions? What more can I test?

(running Win8 Pro, graphics card is radeon 6650M with catalyst mobility 12.10. LAV 0.54/MadVR 0.85.1, MPC-HC b6252)

EDIT: Never mind, looks like the remux only helped on one file...

EDIT2: All files work in ZoomPlayer Free 8.5!
This is really all very weird.

I have found weird issue with native dxva decoding, it seems dxva2n input is not handled same way as dxva2cb.

My output settings are following:
Output format in CCC is set to Limited RGB, madvr outputs TV levels and my TV is set to Limited RGB, basically everywhere is limited RGB.

When using dxva2cb for decoding, WTW and BTB are preserved, after switching to dxva2n, WTW and BTB are cut off, also when watching Black Clipping test video, 17-25 are not flashing anymore.

Ctrl+Alt+Shift+I shows in both cases: "source levels: auto detect (TV)"

When using EVR-CP there is no such difference when changing dxva mode in LAV decoder.

setup: HD 6670, Win7, madvr 0.85.1, LAV filters 0.54.0
Should be fixed in the next build.

When I activate fullscreen together with changing the refresh rate, the start menu icon (from "Classic Start Menu") stays on top of the movie until I exit and enter fullscreen again (without changing the refresh rate).
Of course it only happens with windowed fullscreen, not exclusive mode.
In windowed fullscreen mode, madVR does not do any fancy stuff, so I don't think that's madVR's fault. I don't see what I could do to this change this. But it doesn't happen for me, anyway.

This doesn't seem to happen here, even though LAV says it's using dxva2 native.

/edit:
Or is "[DXVA2]" only supposed to be added to the debug OSD?
Yeah, meant the debug OSD, will fix the changelog.

What I found interesting about DXVA2 scaling is that my GTS450 uses two different scalers for different resolutions.
DXVA2 upscaling for a 720p file looks like bilinear. For SD it looks more like lanczos.
Interesting!

Whenever I play a dvd in MadVR OSD it is wrongly reported "movie 25 fps" even if it's actually 23.976 or 24 fps as correctly reported in real time by both FFdshow raw video filter and reclock. Why?
The "movie ... fps" information is simply what the upstream filter (video decoder or DVD navigator or whatever) is reporting to madVR. It's the fault of the upstream filters, not madVR's fault.

About DXVA Deinterlacing. If I enable the deinterlacing in mVR, will I automatically get the DXVA deinterlacing? And if I get the DXVA deinterlacing does that mean Im getting the Vector Adaptive deinterlacing? I am using ATI 5770. I think i read somewhere mVR will just pick the best deinterlacing algorithm available from the graphics card?
If you activate deinterlacing and don't force film mode then you'll get DXVA deinterlacing. madVR will always pick the best quality algorithm DXVA offers. I'm not sure if the GPU driver control panel options affect which algorithms are offered via DXVA. Maybe.

I just wanted to know that when you enable 10 bit or higher output sometime in a future version, will that work on NVIDIA GeForce cards (or only Quadro cards)?
I don't know.

I just realize that lav dxva2n decoding does not work with the old rendering path option.
Correct, and it probably never will, for some internal technical reasons...

I use AMD HD4550 with Windows7/64bit.
Colors in (2), (3), (4) are the same but different from (1). Especially green is much too oversaturated. (1) stays the same by using copy-back-decoding instead of software.
DXVA-Deinterlacing works as intended, no change in colors.
Interesting: using native decoding instead of copy-back lowers cpu load as expected but as well gpu load! Furthermore dxva2 upscaling has about the same gpu load as Mitchell-Netravali.
Thanks for the feedback.

Hmm, what does it mean if I get a "MadVR reports deinterlacing failed" message when trying to play a 576i/25 video on my Intel i5-430M (Arrandale) when using LAV in DXVA2 Native mode?

EDIT: Doesn't work with DXVA2 Copy-Back or software encoding either. :eek:
It means that the DXVA2 APIs report failure for some reason. I don't know why. And it will be hard to fix without having access to such hardware. The same DXVA2 code in madVR appears to work fine for newer Intel GPUs, so I'm wondering if it's maybe a driver issue. But I don't really know...

When will the madVR have the 3D support in your development estimation?
Replied via PM.

[madshi]

@madshi - I know you can take single screenshots with madVR now, but will you be implementing it s that it can do stepped frames? or should we just use a different render?
might just be MPC-HC then, because it throws up a little box saying it cant do stepped frames and to use a different render
Hmmmmm... I've checked this. It seems to depend on the video and the decoder. After a seek (which is used by MPC-HC for thumbnail creation) madVR currently waits up to 100ms for a decoded video frame to arrive. If the decoder is too slow, madVR will report that frame stepping isn't possible. I've now upped that waiting time to 1 full second. That seems to fix the problem.

[madshi]

Problem 2:
It seems to only be related to DXVA2 decoding, not the scaling. I created a log. (Opened video, seeked once, waited a few seconds, seeked again, waited a few seconds, seeked again, waited a few seconds, then closed the player)
Out-of-order/jerkyness on all three seeks. It looks a bit like this happens until the decoder has caught up after the seek. Not really sure.

log (https://rapidshare.com/files/318478037/madVR_0.85.1_dxva2_out_of_order.7z) (mirror (http://www.mediafire.com/?y3zry4qs4a84znk))
sample (https://rapidshare.com/files/1350258742/worldwarz-p1tt-tlr1_h1080p.mkv) (mirror (http://www.mediafire.com/?etaya7ybam4r0dd))

MPC-HC
Win 7 x64
HD 5850 (Cat 12.11 Beta 8)
I've tested with NVidia 9400, LAV 54.1 and my latest madVR sources on win7 x64 and I can't reproduce any issues here. Seeking works just fine without any issues. Can you please double check with LAV 54.1, and with the next madVR build?

[ajp_anton]

In windowed fullscreen mode, madVR does not do any fancy stuff, so I don't think that's madVR's fault. I don't see what I could do to this change this. But it doesn't happen for me, anyway.It might be that madVR goes to fullscreen before it changes the refresh rate. I found I have no idea how to change the refresh rate manually *after* going to fullscreen to see if that's what's causing it... but it doesn't happen if I just go to fullscreen without a refresh rate change.

edit:
Nevermind haha, found a tool that changes it with a hotkey.
Changing the refresh rate while in fullscreen doesn't cause this. However, if madVR's changer is activated, it will quickly revert the change to whatever madVR wants it to be, which causes the icon to show up again, without even leaving fullscreen.

[pie1394]

Software decoding video and audio on my i7 only takes about 3% of the CPU when running madVR, i cant understand the need for DXVA from an i7 POV anyway. And i got an i7 + 5770 in a HTPC case (which is perfectly possible).

The thing im amazed about is how many people still run ancient hardware that needs DXVA and are willing to pay for PQ by doing DXVA scaling, when PC parts these days are so cheap..

I don't know if Core2Duo P8600 + nVidia ION-LE (on AOpen GP7A-U) is an ancient hardware. This is my 24hr-online mini-server at home and also the regular daily usage PC. Although ION graphics performance is not as great as high-performance GPU, it is still on pair with the Intel HD4000 on i7-3770K. Yet The Intel IGP occupies the biggest market share on PC's graphics chipset...

This is not a joke! I have been used several recent-yeared Intel graphics platform like HD2000, HD3000. All of them are much slower than nVidia ION or AMD/ATI HD5450m, and the Intel graphics driver is much less features. It also has somewhat unstable driver stability on video playback functions.

My HTPC is also an aged hardware. But it is still the fastest one among all my own. C2D E8400@3.6G + MSI Lightning GTX260+ (GT200, 216SP, 1792MB).

With the madVR Luma Lanczos3 + Anti-ring + CUVID, I found the GTX260+ could be overloaded for H.264 1280x720 & MPEG2 HDTV 1440x1080 / 1920x1080 contents displayed at 1920x1080. I need to turn off Luma's anti-ring, or use SW MPEG-2 decoding + processing. But SW MPEG-2 decoding sacrifices some MPEG-2 quality enhancement by this nVidia GPU processing. With madVR previously, it is only available by using CUVID to let GPU process the entire decoding + post-processing (de-ring + de-blocking deinterlacing) which the video image needs to be moved twice between GPU RAM and MB RAM.

Since DXVA input pin is supported in madVR 0.85.x now, the same GPU processing quality can be obtained with reduced GPU loading and I can use Luma Lanczos3 Anti-Ring with most 1280x720, 1440x1080 contents on this GPU.

Even the nVidia old GPU (VP1, VP2 engine) and AMD/ATI HD5xxx (or older series) and Intel old GPU (?) which are not capable of DXVA2 MPEG2_VLD, but only MPEG2_IDCT / MC, it is still able to take advantage of either quality post-processing or performance-optimized MPEG2 video decoding.

For such HW, use Microsoft DTV-DVD Video Decoder to handle MPEG2 video and output decoded image to madVR via DXVA2 path.

[pie1394]

It might be that madVR goes to fullscreen before it changes the refresh rate. I found I have no idea how to change the refresh rate manually *after* going to fullscreen to see if that's what's causing it... but it doesn't happen if I just go to fullscreen without a refresh rate change.

edit:
Nevermind haha, found a tool that changes it with a hotkey.
Changing the refresh rate while in fullscreen doesn't cause this. However, if madVR's changer is activated, it will quickly revert the change to whatever madVR wants it to be, which causes the icon to show up again, without even leaving fullscreen.

I also get a weird PAL DVD. I read the log, and somehow it is started with NTSC 720x480 4:3 stream. No wonder why madVR's refresh rate changer logic decides to use 1920x1080i30 mode. But it leaves the full-screen mode when the first video menu comes out.

Yet all video menus and contents are actually PAL 720x576 16:9. I did the verification by playing all individual .vob files and all of them get switched to 1920x1080i25 mode.

madVR - log[DVD_PAL].rar (http://www.datafilehost.com/download-85bde401.html)

[italospain]

But SW MPEG-2 decoding sacrifices some MPEG-2 quality enhancement by this nVidia GPU processing.


Can you explain the enhancements exactly.

thanks.

[madshi]

The bars are gone, nice & thanks.

But now I have another problem:
madVR fails to correctly enter FSE if there's a display mode change at the same time when DXVA2 downscaling is selected.
It's reproducible to me. Doesn't happen with other downscale methods.
Can't reproduce that here. Please try again with the next madVR build, maybe something has changed. Maybe not, don't know for sure...

[madshi]

I also get a weird PAL DVD. I read the log, and somehow it is started with NTSC 720x480 4:3 stream. No wonder why madVR's refresh rate changer logic decides to use 1920x1080i30 mode. But it leaves the full-screen mode when the first video menu comes out.

Yet all video menus and contents are actually PAL 720x576 16:9. I did the verification by playing all individual .vob files and all of them get switched to 1920x1080i25 mode.
madVR just listens to what the upstream filter says. If that's wrong, it's not madVR's fault.

[AndreaMG]

@Madshi, don't know if already mentioned but deinterlacing features via shortcut keys are broken.:)
Regarding the dvd fps issue, correct me if I'm wrong, MadVR is the last in the chain, so if FFdshow Raw Filter shows me the correct result and also Reclock this means that Lav video is telling something wrong to MadVR or could it be Raw Filter out?:confused:

[madshi]

@Madshi, don't know if already mentioned but deinterlacing features via shortcut keys are broken.:)
Regarding the dvd fps issue, correct me if I'm wrong, MadVR is the last in the chain, so if FFdshow Raw Filter shows me the correct result and also Reclock this means that Lav video is telling something wrong to MadVR or could it be Raw Filter out?:confused:
More details about the deint shortcut keys, please. Don't forget that IVTC (= film mode) won't work when using DXVA decoding.

I don't know which upstream filter is reponsible for the wrong FPS information. You can look at the pin connection information in MPC-HC. The key information is: "AvgTimePerFrame: 417083". You can calculate the FPS from this by doing: "fps = 10000000 / AvgTimePerFrame".

[madshi]

Madshi:
The screen-capture cosmetic glitch happens when when paused (ZP always pauses before taking a screenshot).
For a split-second, whatever was in the frame prior appears. If I just run ZP play a video and press Alt+F, a black screen will appear, but if I open the navigation interfaces that appear on-top of the video and then close the navigation interface, then play a bit of video and take a screenshot, the navigation interface will pop for a split second. It appears like windows that don't get the 'wmpaint' event.
On a quick check I can't seem to reproduce the problem. Does it depend on the decoder? Does it only occur in fullscreen exclusive mode? Or also in windowed mode? Which navigation interface should I open up to reproduce this? Maybe you could create a detailed step-by-step guide to reproduce the problem? Thanks.

[omarank]

With the madVR Luma Lanczos3 + Anti-ring + CUVID, I found the GTX260+ could be overloaded for H.264 1280x720 & MPEG2 HDTV 1440x1080 / 1920x1080 contents displayed at 1920x1080. I need to turn off Luma's anti-ring, or use SW MPEG-2 decoding + processing.

This is really strange. I am too using a GTX 260 in an old system having Pentium D Processor. I also use CUDA for decoding videos. My system is able to handle all kinds of commercially available content with the following settings: chroma upscaling - Jinc 4 taps AR, image upscaling - Jinc 3 taps AR and image downscaling - Catmull Rom AR with Linear Light. My system fails (drop frames) only while playing 1080p 60 fps videos, 4k videos etc. For these files also, it is the hardware decoder which is not able to cope up. Do you have Antialiasing and Anisotropic filtering settings in NVIDIA Control Panel not set to "use 3D Application setting"?

[dansrfe]

You're using a GTX 260? :eek: I thought only >= GTX 660 Ti could handle madVR max settings...?

[ajp_anton]

Could you add a keyboard shortcut for manually activating/deactivating the most appropriate refresh rate for the currently playing video?

[pirlouy]

You're using a GTX 260? :eek: I thought only >= GTX 660 Ti could handle madVR max settings...?
Because you read extreme cases...
Most sources are 24,25 or 30fps. Most ouput are 1080p.
And plenty of GPU can handle Jinc3 easily in these cases. My HD5770 for example, despite not being a "powerful" video card.
Another example: It isn't in comparative anymore, but I can play Crysis 2 without problems with good quality.

But if you take extreme scenarios, you will find 720p 96fps sources, for 1440p output with 3 sharpening filters, 2 denoise, etc.

[bjd]

Because you read extreme cases...
Most sources are 24,25 or 30fps. Most ouput are 1080p

..and if you check the footer you will see a 1080p tv listed so I would assume the scaling is actually nearest neighbour (no scaling)

[AndreaMG]

Could you add a keyboard shortcut for manually activating/deactivating the most appropriate refresh rate for the currently playing video?
I use this for everything and works like a charm (give it a try)::p

http://micromighty.com/Products/Screen_Mode_Switch.aspx

[AndreaMG]

More details about the deint shortcut keys, please. Don't forget that IVTC (= film mode) won't work when using DXVA decoding.
.
Sorry, I cannot reproduce it anymore, but last time it showed while playing a dvd (with Lav Video Decoder set to NVIDIA CUVID HW decoding) and trying to manually (via shortcuts keys) toggle either deinterlacing on/off or change the deinterlacing content type, an OSD message appeared that it couldn't be done because... (don't remember...:p)

[ajp_anton]

I use this for everything and works like a charm (give it a try)::p

http://micromighty.com/Products/Screen_Mode_Switch.aspxBut then it won't auto-detect the movie fps, it won't go back automatically when I close the media player, and it won't help me figure out what's causing the problem I had a few posts up. Plus it requires yet another external program instead of extending a feature that's already there.

[AndreaMG]

But then it won't auto-detect the movie fps, it won't go back automatically when I close the media player, and it won't help me figure out what's causing the problem I had a few posts up.
I'm not following you:DYou said you wanted madshi to add a feature to MANUALLY change the refresh rate and I simply suggested a program that does exactly that not only under MadVR but under everything, of course it does not detect the actual fps of what is going on screen, for that just use the built-in MPCHC autochange fullcreen monitor mode which in my case gives always the correct results even when MadVR fails:p

[DragonQ]

It means that the DXVA2 APIs report failure for some reason. I don't know why. And it will be hard to fix without having access to such hardware. The same DXVA2 code in madVR appears to work fine for newer Intel GPUs, so I'm wondering if it's maybe a driver issue.

I'm sure I've used MadVR before on this machine (with software decoding and hardware deinterlacing) and have never seen this error before, plus EVR does deinterlacing fine. I'm using the latest drivers.

Anything I can do? Provide a sample, or let you access my laptop remotely perhaps?

[secvensor]

Problems occured:
1.At adjustment the active mark on Jink in image upscaling vanishes.
2.The new screenshot option through madvr does not keep influence of external active filters.

[ajp_anton]

I'm not following you:DYou said you wanted madshi to add a feature to MANUALLY change the refresh rate and I simply suggested a program that does exactly that not only under MadVR but under everything, of course it does not detect the actual fps of what is going on screen, for that just use the built-in MPCHC autochange fullcreen monitor mode which in my case gives always the correct results even when MadVR fails:pI said "activating/deactivating the most appropriate refresh rate for the currently playing video". I don't want to need to know the framerate of the video, I was just asking for a way to manually change the refresh rate to the correct one chosen by madVR whenever I want.

[omarank]

You're using a GTX 260? :eek: I thought only >= GTX 660 Ti could handle madVR max settings...?

Well, I think cards >= GTX 260, >= GTX 460, >= GTX 560 and >= GTX 660 should all be able to handle these settings, at least for commercially available content. My TV has native resolution of 1360x768, so madVR always has to do either upscaling or downscaling. I have tested these settings on a variety of content and haven't encountered any serious drop in performance while rendering. GTX 260 has VP2 hardware decoder. It gives up on high bitrate (>100 Mbps) H264 videos, some VC1 videos and of course those fancy 1080p 60 fps and 4k videos.

[pie1394]

Can you explain the enhancements exactly.

thanks.

Just like what I have mentioned, the de-ring and de-block processes with per-pixel-adaptive algorithms. It often needs high computation power, but old-day CPUs were not powerful enough. That's why almost no one design the CPU software-based solutions for MPEG2 decoder.

About the noise reduction and image sharpening (like Darbee), I think most people here don't like it unless it does not kill the details. Fortunately the nVidia Pure Video does a good job about these parts, too.

To my eyes that I can see nVidia's Pure Video MPEG-2 decoder indeed does better jobs on handling interlaced MPEG2 video of most DVD and HDTV contents. For very high bit-rate MPEG-2 (>= 20Mbps) and most H.264 HDTV contents, however, I agree that IQ difference is not so much between software and Pure Video decoders.


This is really strange. I am too using a GTX 260 in an old system having Pentium D Processor. I also use CUDA for decoding videos. My system is able to handle all kinds of commercially available content with the following settings: chroma upscaling - Jinc 4 taps AR, image upscaling - Jinc 3 taps AR and image downscaling - Catmull Rom AR with Linear Light. My system fails (drop frames) only while playing 1080p 60 fps videos, 4k videos etc. For these files also, it is the hardware decoder which is not able to cope up. Do you have Antialiasing and Anisotropic filtering settings in NVIDIA Control Panel not set to "use 3D Application setting"?

You are right about Jinc3 AR mode on GTX260+ partially. I have retested some contents and just realized the exact cases which GTX260+ is overloaded. For progressive contents, GTX260+ has the enough computation power.

The frame dropping only happens to interlaced contents regardless of video content's format and resolution (720x480 ~ 1920x1080 4:2:0 30fps). So it looks like the pixel-adaptive deinterlacing algorithm on GTX260+ is very computation power demanding.

ps: You can set the card to output 1920x1080 resolution to check this. My TV is 1024x1080 interlaced PDP. But I still need to set it to send out 1920x1080 signal to have best IQ. The Hitachi engineers did different tricks on handling different signal resolution + refreh rates to utilize the panel's best characteristics for this TV.

[madshi]

Could you add a keyboard shortcut for manually activating/deactivating the most appropriate refresh rate for the currently playing video?
Why? You're the first/only one requesting this.

I'm sure I've used MadVR before on this machine (with software decoding and hardware deinterlacing) and have never seen this error before, plus EVR does deinterlacing fine. I'm using the latest drivers.

Anything I can do? Provide a sample, or let you access my laptop remotely perhaps?
Anything you can do? Yes, send me your laptop... :D

Ok, if you say older madVR builds might have worked, you could try different older madVR builds to find out which exact version worked and with which exact build it stopped working. Then maybe I can do something.

Problems occured:
1.At adjustment the active mark on Jink in image upscaling vanishes.
This has been reported already, but I can't reproduce it on my PC. I need a detailed step-by-step instruction to reproduce the problem.

2.The new screenshot option through madvr does not keep influence of external active filters.
I'm not sure what you mean?

I've been killing time until PotP would support PS script with mVR, comparing screenshots between EVR/mVR/HR with the 6240 MPC build using the test pattern attached to this post(mediafire refuses to accept it and they've killed all my previous uploads of it, lemme know where else I could upload it if need be).

I could never get the same colors in mVR when using YV12 so I decided to output RGB32HQ from ffdshow, I've also disabled all color conversions in mVR.

That's without processing:

-EVR http://thumbnails103.imagebam.com/22180/9cf827221791316.jpg (http://www.imagebam.com/image/9cf827221791316) -HR http://thumbnails102.imagebam.com/22180/39a292221791319.jpg (http://www.imagebam.com/image/39a292221791319) -mVR http://thumbnails103.imagebam.com/22180/ef090a221791303.jpg (http://www.imagebam.com/image/ef090a221791303)

They're all identical :)

And that's with the stock "nightvision script" set on post-scaling, and this is:

-EVR http://thumbnails104.imagebam.com/22180/337c0a221791307.jpg (http://www.imagebam.com/image/337c0a221791307)
-mVR with my display set to 0-255 http://thumbnails106.imagebam.com/22180/705403221791312.jpg (http://www.imagebam.com/image/705403221791312)
-mVR with my display set to 16-235 http://thumbnails101.imagebam.com/22180/341b17221791315.jpg (http://www.imagebam.com/image/341b17221791315)

I might be doing something wrong! But if that's not the case, improving on the existing PS script support sounds like a great idea but atm getting the exact same colors as with EVR would prove to be extremely useful if any possible please.
I've looked at the nightvision script and found it to be badly written. It actually uses the alpha channel of the renderer output to calculate its colors, and it seems the EVR alpha channel has different data than the madVR alpha channel. The alpha channel is really totally nonsense and should not be used by any sane custom shader script. The current nightvision script has this line:

float c = dot(tex2D(s0, tex), float4(0.2, 0.6, 0.1, 0.1));
Replace this line with the following to remove the incorrect alpha channel use:
float c = dot(tex2D(s0, tex), float4(0.25, 0.65, 0.1, 0.0));

[thedragon1]

hi madshi,
i'm new
i have a laptop (with Nvidia GF GT540M) and try to use madVR. When i played the video , madVR always ran on intel graphic card (HD 3000) and sometime it loaded full (100%) then crashed. Please tell me , how to use madVR for Nvidia card ?
Thanks.
p/s i tried to set high - performace Nvidia processor (Gobal setting) in Nvidia control panel but it's not work
win 8 pro x64

[wanezhiling]

i have a laptop (with Nvidia GF GT540M) and try to use madVR. When i played the video , madVR always ran on intel graphic card (HD 3000)
http://forum.doom9.org/showpost.php?p=1570847&postcount=10533

[pirlouy]

..and if you check the footer you will see a 1080p tv listed so I would assume the scaling is actually nearest neighbour (no scaling)
Sorry but I don't get your message. :/

[thedragon1]

http://forum.doom9.org/showpost.php?p=1570847&postcount=10533

it's work ^^ thanks for helping ^^

[ajp_anton]

Why? You're the first/only one requesting this.Right now, for debugging the problem I'm having. Also you could more easily switch back and forth to see the difference (some screens allow non-60Hz input while still displaying at 60).

But for something actually useful - changing the refresh rate takes time and sometimes messes with background programs. When watching random short videos, I'd rather just stick to 60Hz, even in fullscreen.
IMO it fits nicely with the other on/off keyboard shortcuts already present.

[leeperry]

I've looked at the nightvision script and found it to be badly written. It actually uses the alpha channel of the renderer output to calculate its colors, and it seems the EVR alpha channel has different data than the madVR alpha channel.
oops, will try :thanks:

But as you can imagine the only script that REALLY matters to me at this point is the one that does gamut mapping from AVS (http://www.avsforum.com/t/912720/color-correction-with-a-htpc-simpler-solution-and-now-it-really-works/0_100), I will embed it within PotP automatic profiles based on frame rate and resolution, save PotP's configuration into an .ini file, make it read-only and roll it when rolling displays = masterplan to success http://forum-images.hardware.fr/images/perso/alphat.gif

Its colors when using EVR/VMR9 match pretty much exactly ddcc() even though they both use completely different code AFAIK. There are only very slight differences in the <16 and >235 regions so if you don't crush BTB and WTW anymore as you explained earlier, they might very well end up perfectly identical I guess.

The other PS scripts that would matter to me would be mirroring (but nobody bothered writing one apparently) and levels conversion.....and that's me http://forum-images.hardware.fr/images/perso/zigg.gif that my custom levels/gamut/matrix filename tag feature request will be implemented at some point :cool:

PS: so, when using the following gamut mapping script:
sampler s0 : register(s0);
float4 p0 : register(c0);

static float4x4 r2r =
{
0.879235699565134,0.10128068131873,0.0194836191161357,0,
0.0448339014270118,0.915215520228331,0.0399505783446596,0,
0.0154998582994896,0.0971141863242463,0.887385955376264,0,
0, 0, 0, 0
};

float4 main(float2 tex : TEXCOORD0) : COLOR
{
float4 c0 = tex2D(s0, tex);
c0 = pow(c0, 1/0.45);
c0 = mul(r2r, c0);
c0 = saturate(c0);
c0 = pow(c0, 0.45);

return c0;
}

With the 6240 build of MPC, feeding RGB32HQ(dithering disabled in ffdshow) and this script set for post-resizing:

1) untouched:

EVR CP: http://thumbnails103.imagebam.com/22349/96ca0d223487747.jpg (http://www.imagebam.com/image/96ca0d223487747)

mVR(dithering and all color conversions disabled): http://thumbnails105.imagebam.com/22349/ef090a223487748.jpg (http://www.imagebam.com/image/ef090a223487748)

= identical :cool:

2) aforementioned script processed:

EVR CP: http://thumbnails106.imagebam.com/22349/e47260223487742.jpg (http://www.imagebam.com/image/e47260223487742)

mVR(dithering and all color conversions disabled): http://thumbnails101.imagebam.com/22349/4cee37223487745.jpg (http://www.imagebam.com/image/4cee37223487745)

= not identical :o

[madshi]

The other PS scripts that would matter would be mirroring (but nobody bothered writing one apparently)
Mirroring is extremely easy:

sampler s0 : register(s0);

float4 main(float2 tex : TEXCOORD0) : COLOR
{
return tex2D(s0, float2(1 - tex.x, tex.y));
}

[DragonQ]

Anything you can do? Yes, send me your laptop... :D

Ok, if you say older madVR builds might have worked, you could try different older madVR builds to find out which exact version worked and with which exact build it stopped working. Then maybe I can do something.
Hmmm...I can't find a previous build that works, even back to 0.61.

That leaves two possibilities: a problem with the latest driver, or MadVR deinterlacing has never worked on Arrandale. :(

[madshi]

I think it probably never worked. Don't know why...

[leeperry]

Mirroring is extremely easy:

sampler s0 : register(s0);

float4 main(float2 tex : TEXCOORD0) : COLOR
{
return tex2D(s0, float2(1 - tex.x, tex.y));
}
oh my oh my, very cool stuff! :thanks:

I think someone should create a separate thread with easy access to PS scripts in the OP, much like the one they have on HCFR.

[bjd]

Sorry but I don't get your message. :/

The point I was making if you are watching 1080p source on a 1080p screen and have MadVR set to use Jinc, it will default to nearest neighbour as there is no scaling to perform so rendering times are a lot lower.

On my GT430, Jinc4 upscaling (576p/720p24 -> 1080p24) renders in around 35-40ms which is too slow really.

[pirlouy]

It won't "default nearest neighbour", it will just do no image scaling (but still chroma scaling). But even if I take 576p or 720p, there's no noticeable difference in rendering time.
Indeed, your GT430 is not able to handle Jinc (but it's a very low market GPU), the same goes for Intel HD3000. My point was not to tell all GPU can handle Jinc, but no need to get latest expensive GPU to handle current cases (except if you're a perfectionist which adds several filters).