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

[vinnytx]

I hope 10-bit support will be improved.
I mean that it should be available even in windowed mode.
Also, 8-bit videos should not have 10-bit output

[madshi]

Why should 8bit videos not have 10bit output? Videos are YCbCr. Output is usually RGB. The conversion is floating point math, so you have more than 8bit after YCbCr -> RGB conversion.

[vinnytx]

I thought that conversion was lossy.

[huhn]

yes float to integer is lossy.
but float to 8 bit is more lossy than float to 10 bit.

[vinnytx]

I have problems when I play 4k 10-bit videos in fullscreen. I have this error.
https://i.imgur.com/vDWtR2G.jpg

I tried many settings in "Video Renderer" > General/Presentation/Composition and I couldn't solve the problem.


Example
http://www.mediafire.com/file/a3knqi0uaq5ssng/10_bit.mp4

[Shiandow]

You don't have enough memory on your GPU apparently. Are you using any renderscipts? Those could make things worse. I think SSIMDownscaler in particular could end up using quite a lot of memory.

[vinnytx]

Thanks
My video card is 960 gtx with 4GB. I don't use any render script. Have you tested the video I posted?

[ryrynz]

I have problems when I play 4k 10-bit videos in fullscreen. I have this error.


Usually these sorts of problems are related to your graphics card driver installation, uninstall and then reinstall the latest ones.

[vinnytx]

I uninstalled and reinstalled Nvidia driver and nothing has changed.

I found the problem: when I enable DXVA2 copy-back in Lav Filters x86 the problem occurs.
No problem with MPDN x64 and Lav Filters x64

[Shiandow]

Are you sure MPDN is actually part of the problem? Does Lav Filter x86 work when you use a different video player?

[vinnytx]

Yes, I'm sure. Zoom Player and PotPlayer don't have this problem

[stax76]

Any news on the 2880 DPI issue? I've my own player now but still follow development of other players.

[Shiandow]

Any news on the 2880 DPI issue? I've my own player now but still follow development of other players.

I've had a quick look and it looks like MPDN just didn't like it when it had to resize before it had a chance to initialise everything. I've managed to fix it.

Surprisingly most things actually worked. Apart from some small cosmetic problems.

Yes, I'm sure. Zoom Player and PotPlayer don't have this problem

Still not sure why the x64 vs x86 MPDN would behave differently. One possible explanation is that there is a problem in the settings of the x64 version. Also, the error dialog should have a button to copy the detailed information, if you could get the stacktrace from there I might be able to figure out where the problem occurs.

Also, there's still the possbility that you're actually just running out of GPU memory, and that minor changes in settings affect whether you run out completely or not.

[vinnytx]

I have problems with MPDN x86.

I uninstalled x86 version and I installed x64. At the moment I can't do other tests. I tried MPDN x86 portable version and it gives me an error regarding filters that are not properly configured (video graph failed to render). That's not a problem since I configured my pc for x64 video players

[Anime Viewer]

I'm encountering a problem during playback of:
http://tracker.minglong.org/torrents/%5BMezashite%5D%20Aikatsu%20Stars!%20-%20073%20%5BBAE1A1F9%5D.mkv.torrent

At 20:06 (twenty minutes and six seconds) into the video MPDN stops responding. I tried running in debug mode, but it didn't generate a crash dump. My settings are as follows:

MPDN 64-bit edition
version 2.48.0.3603
XySubFilter with reduce aliasing checked
Prefer: Image Quality
Always use new OpenCL Interop API (I tried disabling this and restarted, but it still crashed at the same part, so I have ruled this out)
Presentation API: Driect3D 11
Graphic Adapter: Nvidia GeForce GTX 680M
Video Renderer:
Colormetric: Automatic
Output Range: Full
Video Output bit depth: 10-bit
Script Chain Deband -> SuperChromaRes -> Super-xBR
Ordered dithering
I tried in full screen and in a window, and both crashed but anyway here is what I'm using for scalers:
Luma up: Jinc with anti-ring
Chroma up: Jinc 4 tap with anti-ring
Luma Down: Softcubic 75
Chroma down: Bicubic 75

If you need any more info, or testing with another setting let me know.

BTW: I ran the video with MPC-HC (with MadVR) and didn't encounter the crash/freeze.

[carpedee]

I really like this video player, I feel like it puts less strain on my cpu than MPC used with the LAV filters (DX11 helps a lot I guess).

I have a few questions/suggestions regarding some features I didn't find:

- Is it possible to edit the hotkeys like the one to enter in fullscreen mode for example?
- Is it possible to run several instances of MPDN at the same time?
- Is there a hotkey to take a screenshot when playing a video?

[kingpage]

I've just been reading up on Madshi's NGU (NG1). It's very impressive. Is there any plan on bring it to MPDN? Does anyone know if it's easier to run than SSSR (SuperxBR, NNEDI3, etc)?

[aufkrawall]

NGU will remain madVR exclusive and questions regarding it should be discussed in madVR thread. It got different quality settings and the performance costs range goes from relatively close to Jinc to ~NNEDI3 64 neurons.

[Heaud]

Is there a way for MPDN to gain window focus when opening a video? Everytime I open a file, the video loads but does not gain focus. Operating system is Windows 10 64-bit with Creators Update.

[Shiandow]

I've just been reading up on Madshi's NGU (NG1). It's very impressive. Is there any plan on bring it to MPDN? Does anyone know if it's easier to run than SSSR (SuperxBR, NNEDI3, etc)?

Whether NGU will be brought to MPDN, is up to Madshi, and really there's no reason for him to help improve a video player which uses a different renderer. Despite that Madshi has already helped MPDN quite a bit in various ways, but expecting him to implement all of MadVR's functionality in MPDN is just not realistic.

Is there a way for MPDN to gain window focus when opening a video? Everytime I open a file, the video loads but does not gain focus. Operating system is Windows 10 64-bit with Creators Update.

As of now that's the expected behaviour when MPDN is already running when the file opens. It would be fairly easy to change that with a player extension.

You can make MPDN stay on top while playing, which might help. You can enable this in the context menu > Window > Always on Top. And you can limit it to only stay on top while playing in Options > General > Window.

[Zachs]

I have come across a very odd issue. Anytime I select any Upscaler/Downscaler to be Lanczos, MPDN gives an error when trying to render video.

"Specified argument was out of range of valid values."

Even if it's just luma and the video is being played at its resolution(no scaling)

Win10 x64 1703, MPDN x64

This seems to have started after Creator's Update. I think turab's issue (https://forum.doom9.org/showthread.php?p=1804039#post1804039)above may be related.

Ok. Bit of an update regarding this particular bug. It's 100% on Microsoft. If you are up-to-date with your Windows Update, this bug will magically go away.

[Heaud]

As of now that's the expected behaviour when MPDN is already running when the file opens. It would be fairly easy to change that with a player extension.

You can make MPDN stay on top while playing, which might help. You can enable this in the context menu > Window > Always on Top. And you can limit it to only stay on top while playing in Options > General > Window.
Having the Always on Top is not the best option for my use as it will still not gain focus so keyboard shortcuts will not be active until I click on the video frame. Another issue is that clicking on the video frame also pauses the video so to gain focus so that keyboard shortcuts can be used would cause the video to be paused.

You mentioned that it would be easy to remedy this inconvenience with a player extension. Is it possible for someone with basic, but inexperienced understanding of scripting such as myself be able to create an extension that enables automatic window focus upon loading a video file?

[kingpage]

Whether NGU will be brought to MPDN, is up to Madshi, and really there's no reason for him to help improve a video player which uses a different renderer. Despite that Madshi has already helped MPDN quite a bit in various ways, but expecting him to implement all of MadVR's functionality in MPDN is just not realistic.


That's true. Can MadVR be used with MPDN? I can't seem to get it to work.

[ryrynz]

No. It cannot.

[nevcairiel]

That's true. Can MadVR be used with MPDN? I can't seem to get it to work.

No, MPDN can only use its own renderer.

[Zachs]

I've just been reading up on Madshi's NGU (NG1). It's very impressive. Is there any plan on bring it to MPDN? Does anyone know if it's easier to run than SSSR (SuperxBR, NNEDI3, etc)?

MPDN extension is an open source project - madshi's NGU is a closed source / proprietary algo, much like its madVR. It's inherently incompatible with MPDN extension's core concepts - our devs believe in sharing knowledge with the public with appropriate licensing - something the modern world agrees with (.net core, almost a hundred thousand nuget packages etc.).

The only thing I haven't been able to open source is the core player itself, mainly due to the strict copyright laws here where I live.

That said, it's unlikely madshi would ever make NGU compatible with open source licensing, even if we somehow know how it's done.

[aufkrawall]

I really like MPDN and especially the open render scripts, so please don't stone me to death. :o
However, there's also mpv, which is 100% open source, cross platform and constantly developing at a fast pace.
Sure MPDN still has its unique features, but I think it would be more helpful for more users to directly offer shaders, knowledge etc. to the mpv project in one or the other way (even though open source scene atmosphere might be a bit difficult, but dunno).

[Zachs]

I believe there's no reason why contributing to MPV wouldn't directly benefited MPDN or vice versa. That's the whole point of the open source initiative.

[Shiandow]

You mentioned that it would be easy to remedy this inconvenience with a player extension. Is it possible for someone with basic, but inexperienced understanding of scripting such as myself be able to create an extension that enables automatic window focus upon loading a video file?

Should be pretty doable. If you're up for it I would strongly recommend trying it yourself. Some things will be easier if you download Visual Studio (the community edition is free for personal use) but it's not necessary (it'll just be slightly harder to figure out what went wrong). You might want to go to Options > General > Debugging and enable the debugging dialog, because otherwise you're not going to get any useful information if it fails.

Anyway, here's a rough guid to what you'd need to do. If you're unsure about anything you can probably google the answer, or if not come back and ask in this thread.

To start I'd recommend basically copying the GoTo (https://github.com/zachsaw/MPDN_Extensions/blob/master/Extensions/PlayerExtensions/GoTo.cs) script. Just copy that file somewhere and open it up in your favorite text editor.

That extension already checks when a new file opens in order to enable / disable its item in the menu (try to figure out how it does this), so we can just reuse that to do something else (like bring the window to the front). The rest we don't need, you can remove everything but the "Descriptor" and the "Initialize", "Destroy" and "PlayerStateChanged" methods.

Then you need to fill out the descriptor with new information (most of this is optional, but you do need to change the GUID to another one, otherwise MPDN can't tell it apart from the GoTo extension). And be sure to rename the GoTO class and anything else that says "GoTo" to something more sensible.

After that bit of bookkeeping is over you can change the PlayerStateChanged method to actually bring the window into focus. You'll probably need to start with checking if the player started playing, using something like "if (e.NewState == PlayerState.Playing)". And then bring the window to the front. Some quick googling suggests calling "Activate()" on the window form (which is "Player.ActiveForm"), but not everyone got this to work apparently. There are a lot of other suggestions on google or stackoverflow on how to get window to the front in C# so you should probably try one of those if the simple method doesn't work.

After that you can copy the file containing your extension to MPDN / Extensions and it should work (although don't be too disheartened if it doesn't, try the debugging dialog and see if you can figure out what's wrong). Best of luck!

I really like MPDN and especially the open render scripts, so please don't stone me to death. :o
However, there's also mpv, which is 100% open source, cross platform and constantly developing at a fast pace.
Sure MPDN still has its unique features, but I think it would be more helpful for more users to directly offer shaders, knowledge etc. to the mpv project in one or the other way (even though open source scene atmosphere might be a bit difficult, but dunno).

I might be a tad biased, but in MPV their equivalent to "renderscripts" is some weird combination of GLSL with some 'special comments'. In MPDN you can use the whole of C# as your 'special comments' and use that to manipulate HLSL / OpenCL / DirectCompute shaders. Sure their API does allow people to port scripts to MPV, but it doesn't give you a lot of support for actually developing them. It also makes it very hard to tell what's happening, which isn't great for sharing ideas.

It's possible there's a more advanced API somewhere that I'm not aware of.

[pirlouy]

That's the whole point of the open source initiative.
Exactly: split/fork as soon as possible, in order to lose potential users, instead of trying to have one unique project, easier for newcomers. Total Linux Fork users < 1%. Sorry for this grievance.

I don't know why you're still mad at Madshi. Sometimes you don't want someone to tell you what to do, you want to do it your way, so closed source is an option. Or another example, imagine some company wanna buy his NGU algorithm. You have billion reasons to want closed-source project, the same as open-source. As good as your player is, whatever the reasons, it is still closed-source, and nobody will blame you for that.

Ahah it's fun, it was a long time since we had a discussion about Open/Closed Source. :)
I hope I didn't piss you off. I let you have the final word in the end.

@Shiandow: maybe you could ask a request on MPV tracker. I'm sure they would be interested in a "shader coder" (sorry if it's an approximate description).

[madshi]

I'm not sure if Zachs is mad at me. FWIW, there was a time when I wasn't very friendly to him, and I kinda regret that. Sadly, in forums one often writes things without really thinking it through, and then it's hard to take it back, because it's all "written in stone", so to say.

[jkauff]

There are pluses and minuses to both open and closed source applications (Richard Stallman notwithstanding). Microsoft, for example, recently has been open sourcing a lot of their work, but you're not going to get the source code for Windows or Office anytime soon. Too many salaried people working on those products.

madshi, bless him, has been sharing the fruits of his labor for many years now for free--and in fact won't take donations. For my part, I wouldn't want madVR to be a community project. It's nice to have one person in charge, especially one who takes as much user feedback as madshi does. I look forward to the day when I can give him my money.

There's plenty of room for madVR with its large base of supported media players, MPDN, mpv, and any other innovative video solution that comes along. The real shame is that most people still use Windows Media Player or VLC because they don't know or care that there are superior alternatives.

[aufkrawall]

It's possible there's a more advanced API somewhere that I'm not aware of.
I must confess that I got a bit lazy and got satisfied with good ol' Jinc scaling (and mpv's ewa_lanczossharp looks quite sharp for downscaling if scaling factor isn't too high).
But there also is e.g. NNEDI3 available as a usershader, doesn't seem like GLSL is a showstopper (I gues there is nothing with OGL without annoyance :p ).
Like I said, MPDN still has its very own unique features, which is totally fine and great.
It's just that, afterall, it's only about watching videos. Refining scaling quality etc. is cool, but it's just annoying how hard it is to use MPDN or madVR (+ players) for e.g. online streaming. It just doesn't meet my media habits anymore. With mpv + youtube-dl I basically can watch anything which is available for free by just pointing the URL to the player (YouTube, YouTube live, YouTube HDR, general HLS, files located online...).

[Anime Viewer]

After updating to the most recent version I'm getting the following error message:

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

Specified argument was out of the range of valid values. (MediaPlayerDotNet)

------------------------------
Error Type = System.ArgumentOutOfRangeException
Error Message = Specified argument was out of the range of valid values.
Error Source = MediaPlayerDotNet
Error Site = Mpdn.IScaler C(Mpdn.IScaler)
Error occurred = at a.C.C(IScaler )
at a.C.b(IScaler )
at Mpdn.D3D9VideoRenderer.VideoRenderer.ag()
at Mpdn.D3D9VideoRenderer.VideoRenderer.aG()
at Mpdn.D3D9VideoRenderer.VideoRenderer.e(Boolean )
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(String , FrameBufferInputFormat , Int32 , Int32 , Int32 , Int32 , InterlaceFlags , Double )
at B.M.A(String , FrameBufferInputFormat , Int32 , Int32 , YuvColorimetric , InterlaceFlags , PointF )
at B.M.A(IMedia )
at MediaPlayerDotNet.MainForm.A(IMedia , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.OpenMedia(IMedia , Boolean , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.OpenMedia(String , Boolean , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.P(Object , EventArgs )
at System.EventHandler.Invoke(Object sender, EventArgs e)
at System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
at System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
at System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
at System.Windows.Forms.Control.WndProc(Message& m)
at System.Windows.Forms.ToolStrip.WndProc(Message& m)
at System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
at 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 = 15.9 GB
Available Physical Memory = 11.2 GB
Date Time = 9/30/2017 11:50:13 AM

I've checked and applied the one Windows Update I didn't have, but even after applying that and restarting the computer I still get the error.

[Shiandow]

But there also is e.g. NNEDI3 available as a usershader, doesn't seem like GLSL is a showstopper (I gues there is nothing with OGL without annoyance :p ).

GLSL isn't the problem. It's just that MPV's usershaders aren't much more than GLSL with some extra metadata. In MPDN the metadata is replaced with C#, which is a heck of a lot more powerful.

The only reason it's unfortunate that they use GLSL (although portability kind of forced that decision on them) is that this means you can't just automatically convert MPDN's renderchains into usershaders.


It's just that, afterall, it's only about watching videos. Refining scaling quality etc. is cool, but it's just annoying how hard it is to use MPDN or madVR (+ players) for e.g. online streaming. It just doesn't meet my media habits anymore. With mpv + youtube-dl I basically can watch anything which is available for free by just pointing the URL to the player (YouTube, YouTube live, YouTube HDR, general HLS, files located online...).

You don't like the 3DYD Youtube filter?

For what it's worth, I have my old player extension that downloads a stream using youtube-dl and loads it into MPDN. I didn't work on it much because I didn't think it was too useful when there's a nice DirectShow filter for youtube urls. But if there's a demand for it I could see if I could polish it up a bit.

[aufkrawall]

You don't like the 3DYD Youtube filter?

It doesn't work reliably for me with 4k 60fps video, playback often stalls after some time because it's not streaming fast enough.
So far I can only achieve stable playback with mpv + youtube-dl (and browsers, of course).


For what it's worth, I have my old player extension that downloads a stream using youtube-dl and loads it into MPDN. I didn't work on it much because I didn't think it was too useful when there's a nice DirectShow filter for youtube urls. But if there's a demand for it I could see if I could polish it up a bit.
I already made a fuss about it in the MPC HC thread. :devil:
It has several advantages like expanding general HSL functionality, special media site functionality or allowing streaming of YouTube live.
I'd think if more users knew about this, they'd find it quite useful.

Though I'm not sure if some basic youtube-dl integration alone is the holy grail. I also tried piping youtube-dl's output to MPC HC and it's still limited in terms of e.g. seeking compared to mpv.

[Shiandow]

Though I'm not sure if some basic youtube-dl integration alone is the holy grail. I also tried piping youtube-dl's output to MPC HC and it's still limited in terms of e.g. seeking compared to mpv.

Out of interest, how exactly do you use youtube-dl with mpv? Do you just pipe the output from youtube-dl into mpv?

[aufkrawall]

You drop its executable into mpv's folder (or install it as a dependency on Linux) and it's automatically used when opening URLs with mpv.
It's not used for the actual streaming but for source discovery. At least that's my conclusion from seeing its process closing after playback start.

[Cy4n1d3]

Ok. Bit of an update regarding this particular bug. It's 100% on Microsoft. If you are up-to-date with your Windows Update, this bug will magically go away.
I am fully updated using latest Win10 (1703) and still get this error.
After downgrading the player from 2.49 to 2.48 (still using 1.31 renderscripts) the error goes away and everything is working as before.
So even if this error / bug is somehow occuring due to Windows malfunctioning, some code change inbetween 2.48 and 2.49 seems to trigger it.

Regards

[mringis]

After updating to the most recent version I'm getting the following error message:

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

Specified argument was out of the range of valid values. (MediaPlayerDotNet)

------------------------------
Error Type = System.ArgumentOutOfRangeException
Error Message = Specified argument was out of the range of valid values.
Error Source = MediaPlayerDotNet
Error Site = Mpdn.IScaler C(Mpdn.IScaler)
Error occurred = at a.C.C(IScaler )
at a.C.b(IScaler )
at Mpdn.D3D9VideoRenderer.VideoRenderer.ag()
at Mpdn.D3D9VideoRenderer.VideoRenderer.aG()
at Mpdn.D3D9VideoRenderer.VideoRenderer.e(Boolean )
at Mpdn.D3D9VideoRenderer.VideoRenderer.A(String , FrameBufferInputFormat , Int32 , Int32 , Int32 , Int32 , InterlaceFlags , Double )
at B.M.A(String , FrameBufferInputFormat , Int32 , Int32 , YuvColorimetric , InterlaceFlags , PointF )
at B.M.A(IMedia )
at MediaPlayerDotNet.MainForm.A(IMedia , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.OpenMedia(IMedia , Boolean , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.OpenMedia(String , Boolean , Boolean , Boolean , Boolean )
at MediaPlayerDotNet.MainForm.P(Object , EventArgs )
at System.EventHandler.Invoke(Object sender, EventArgs e)
at System.Windows.Forms.ToolStripItem.RaiseEvent(Object key, EventArgs e)
at System.Windows.Forms.ToolStripMenuItem.OnClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleClick(EventArgs e)
at System.Windows.Forms.ToolStripItem.HandleMouseUp(MouseEventArgs e)
at System.Windows.Forms.ToolStrip.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.ToolStripDropDown.OnMouseUp(MouseEventArgs mea)
at System.Windows.Forms.Control.WmMouseUp(Message& m, MouseButtons button, Int32 clicks)
at System.Windows.Forms.Control.WndProc(Message& m)
at System.Windows.Forms.ToolStrip.WndProc(Message& m)
at System.Windows.Forms.ToolStripDropDown.WndProc(Message& m)
at 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 = 15.9 GB
Available Physical Memory = 11.2 GB
Date Time = 9/30/2017 11:50:13 AM

I've checked and applied the one Windows Update I didn't have, but even after applying that and restarting the computer I still get the error.

same with me :scared:

I'm on win 7 64. go back to 2.48 for now

[Zachs]

Hmm I'll investigate. I'm no longer getting the error but I did at some stage.

Have you guys tried uninstalling MPDN fully and reinstalling? Make sure you backup your settings before you do that. And try doing a windows restart after reinstalling.

[Anime Viewer]

Hmm I'll investigate. I'm no longer getting the error but I did at some stage.

Have you guys tried uninstalling MPDN fully and reinstalling? Make sure you backup your settings before you do that. And try doing a windows restart after reinstalling.

I did try a complete uninstall and reinstall of MPDN as well as the extensions, rebooted, reinstalled (clean install of both MPDN and extensions), rebooted again, and then launched MPDN. The problem was still present.

[thewildsun]

I'm having a problem with video play speed and subtitle sync. If I set the rate to 0.75x, the subtitles are too slow, and if I set the rate to 1.25x, the subtitles are too fast. Is this a bug?

I'm on 2.48 since I get the "Specified argument was out of the range of valid values. (MediaPlayerDotNet)" error on 2.49, on Win10.

[Heaud]

I have an odd issue in which MKV muxed from a MP4 is not playing at the intended 60FPS. The source is from a recording of a game which plays fine from the MP4 container, but I get lots of repeated frames and a reported average video rate of 39.9Hz based on the renderer overlay (https://img.warywolf.net/k0VYm.png). I am using MPDN 2.49.0.3606 and the latest version of MKVToolNix to mux.

Any ideas as to what might be going on? YouTube was able to render a video upload at 60FPS with the same MKV file so I am guessing the muxing isn't causing a problem with playback.

[huhn]

do you have a short sample?

[Heaud]

do you have a short sample?

Here is a link to the sample (http://www113.zippyshare.com/v/d8BtOYNO/file.html) (warning: there is audio) which is only a 18 second highlight of a match.

[huhn]

ok that a little bit strange it seems like the file is VFR with relative stable 40 FPS. maybe the file is stable 40 FPS and the game was using something like adaptive V-sync to stay stable at 40.
but the file claims to be constant 60 FPS in media info.

i'm not really sure but this line for audio : Frame rate : 46.875 FPS (1024 SPF)
but that's not something i know alot about.

[mikezilla2]

so why should i use this over MPC HC BE / modern codecs ?

[TaurezAG]

These two features will be very useful

1. Screenshot
A screenshot feature will be very useful, even more so if before/after processing(like scaling/RenderScript) screenshots can be taken separately.

2.
Additional input file format support for the 3DLUT RenderScript.
Currently, it only supports the .3dlut format which from what I've seen is supported only by a few software, and also just by calibration software. I would like to use 3DLUT for color correction. I'm using DaVinci Resolve. This is the format specification for the .cube files generated by Resolve: Link (https://forum.blackmagicdesign.com/viewtopic.php?f=21&t=40284)
I've attached a sample 3DLUT file generated by Resolve.

[huhn]

the 3D LUT implementation never worked correctly.

DisplayCAL supports both IRIDAS and 3D LUTs.
so you just needs a profiling of your screen (it may even work with profiling from a different program) and you can create a LUT for both programs in a very short time with the "same" results.