madVR - high quality video renderer (GPU assisted)

[oddball]

So does nobody know how to set the display modes so that 1080p content is sent unscaled to the 4K TV (allow the TV to do the scaling) and 4K content is sent @4K? "1080p23, 2160p30" results in 2160p content being downscaled to 1080p30. Just curious why this is occurring. At present I have my brother sending 1080p content scaled @ 2160p (I am guessing?) since that is the res he has the display set at. Basically I removed all 1080p display modes in MadVR's display profile.

[Asmodian]

hy there,
i meant this one:
it was fixed when moving to older driver, wich i use now..
But id like to upgrade and well, maybe someone tried it...

Oh right! I had forgotten about that. I am sure it is a driver bug but I cannot test it as GK110 does not seem to be affected. Nvidia might take a long time to fix something like this.

[StinDaWg]

Nope, it depends on your graphic card. DxVA2 scaling is much better than bilinear in my case (HD7850). It looks similar to bicubic 1.0 in my eyes.
I have a 7850 too. I just checked a test video I have saved for these purposes that I've watched a million times. DXVA2 looks slightly blurrier than bicubic. It may be a bit better than what I was getting on my old 5450, but I wouldn't recommend anyone use it. It's noticeably worse than Lanczos/Spline, and that's not even getting into Jinc/NNEDI3.

[sheppaul]

I have a 7850 too. I just checked a test video I have saved for these purposes that I've watched a million times. DXVA2 looks slightly blurrier than bicubic. It may be a bit better than what I was getting on my old 5450, but I wouldn't recommend anyone use it. It's noticeably worse than Lanczos/Spline, and that's not even getting into Jinc/NNEDI3.

I'm not saying DxVA2 scaling is better than anything you mentioned here except for bilinear in the above post of yours. It's better than bilinear at least. That's the point.

[Asmodian]

Slightly blurrier than bicubic is not good, I hadn't realized AMD cards did not have a good DXVA2 scaler. I wonder why nVidia and especially Intel have better quality DXVA2. I suppose it matters most with Intel as you are more likely to not have the power for the better scalers in madVR; I hope AMD's APUs offer better hardware scaling?

[AMZMA]

hello

how to change OSD font?

thnks

[Asmodian]

You cannot.

[makz]

Hi, I have some problem with MPC-HC and Madvr 3DLut. Trying on my Dell 2312. I create madvr .3dl and all fine, but there is differnce in measurments HCFR between madTPG and MPC-HC+madvr. My problem with rendering 100% color with 3dlut (Dell 2312 has oversatturated native green). Sorry for my bad english. See some pictures. What wrong? Thanks!

[Asmodian]

Hi, I have some problem with MPC-HC and Madvr 3DLut. Trying on my Dell 2312. I create madvr .3dl and all fine, but there is differnce in measurments HCFR between madTPG and MPC-HC+madvr. My problem with rendering 100% color with 3dlut (Dell 2312 has oversatturated native green). Sorry for my bad english. See some pictures. What wrong? Thanks!

It looks like you are using limited range in one test and full range in the other? This has always been the reason when I have gotten different results in HCFR with madTPG and madVR. :o

Make sure you are actually using the same test patches in both places.

edit: HCFR also has a save image option so you can get more reasonably sized images to share. ;)

more explanation: The 3DLUT does not clip when you go out of range, it simply lets the calibration go bad as it cannot be right when over 100%. This means that measuring WtW (or Greener than Green) will give what seems to be uncalibrated colors; over 100% green will go outside the target gamut.

[baii]

If I remember correctly dxva scaler was added because they are cheap. On my win tablet, noting other than dxva scaler will run without drop frame.

Sent from my 306SH

[makz]

more explanation: The 3DLUT does not clip when you go out of range, it simply lets the calibration go bad as it cannot be right when over 100%. This means that measuring WtW (or Greener than Green) will give what seems to be uncalibrated colors; over 100% green will go outside the target gamut.
I do not understand why there is a clip in MadTPG and it is not in MPCHC+MadVR. what is the difference? and sorry for the pictures)

[Remicade]

I don't understand why I see all bars in AVS HD 709 (1- Black Clipping.mp4) when I choose in madVR/devices/properties 16-235 and only 17 and above when I choose 0-255 and raise brightness control at 100. It seems illogical, my graphic card output 0-255.

[Stereodude]

I don't understand why I see all bars in AVS HD 709 (1- Black Clipping.mp4) when I choose in madVR/devices/properties 16-235 and only 17 and above when I choose 0-255 and raise brightness control at 100. It seems illogical, my graphic card output 0-255.
Why's that? Above the selection it says:

"the display expects the following RGB output levels:"

You can pick either PC levels (0-255), TV levels (16-235), or Custom. The content you want to see in YUV/YCbCr video is encoded 16-235. If you pick PC levels MadVR turns YUV 16-235 into RGB 0-255 and levels below 16 and above 235 are truncated. If you pick TV levels YUV 16-235 is turned into RGB 16-235 and nothing is truncated/clipped.

In the latter case your display is expecting 0-255 so it will show you 0-15 and 236-255 (which shouldn't be seen) because you have madVR set incorrectly.

[tjcinnamon]

Is there any chance this gets built directly in the the XBMC/Kodi Windows fork?

[Asmodian]

I do not understand why there is a clip in MadTPG and it is not in MPCHC+MadVR. what is the difference? and sorry for the pictures)

What are you displaying to test in MPC-HC+madVR?

The patches HCFR generates using madTPG are limited range. If you are watching full range video as limited range video in MPC-HC you are getting the expected result.

It is easiest to see using the AVS HD test patterns (http://www.avsforum.com/forum/139-display-calibration/948496-avs-hd-709-blu-ray-mp4-calibration.html). If you watch the white clipping test in basic settings you can see the effect as white goes WtW. The color test patterns included are also limited range so you could use them to do your measurements.

[michkrol]

I see a little discussion going on about DXVA2 scaling.

From my experience, it's only worthwhile on Intel's GPU starting with Ivy Bridge architecture (3rd generation Core i3/5/7/Pentium/Celeron), since it seems to have dedicated hardware for scaling.
Heck, my low power Celeron 1007U can scale 4K@24fps to 1080p with smooth playback, so it is the solution for netbooks and similar.

My Geforce 750Ti seems to use shaders to perform DXVA2 scaling. The GPU load is on par with madVR's own Lanczos4 + AntiRinging scaling, while the quality is far from it, although not really bad.

Is there any chance this gets built directly in the the XBMC/Kodi Windows fork?

MadVR has "support" for being used by a variety of software. You need to ask the developers of the software you want to have support for madVR.

There's little interest from Kobi's developers, since it is multi-platform, while madVR is not. This thread (15s spent on Google) from Kobi's forum might be of interest to you: http://forum.kodi.tv/showthread.php?tid=200527
EDIT: I see you're active in it, so you know better than me how it is.

[leonccyiu]

There is a really informative thread about DXVA decoding on avsforum where Renethx who is the best forum member did extensive testing of patterns to determine what sort of techniques were being used.

http://www.avsforum.com/forum/26-home-theater-computers/1477339-so-you-ve-built-your-htpc-now-what-next-how-get-ultimate-picture-sound-quality-your-htpc-madvr-svp-xbmc-mediabrowser-jriver.html

According to this post, intel has dedicated hardware scalers and provides the best scaling similar to lanczos and ar, while amd and nvidia are using bi-linear and shaders.

http://www.avsforum.com/forum/26-home-theater-computers/1477339-so-you-ve-built-your-htpc-now-what-next-how-get-ultimate-picture-sound-quality-your-htpc-madvr-svp-xbmc-mediabrowser-jriver-6.html#post23847902

[leonccyiu]

Since Chrome and Firefox are based on open source code projects, is it possible that in future, madvr can be implemented as an option as a video renderer since most video is now delivered using html5 and rendered by the browser now.

Right now, almost all web video is scaled using billinear and does not take advantage to dxva scaling, for those with high res monitors, such as even when watching 4k youtube on the 5k imac, it makes the picture very soft and blurred.

This would allow web video such as vimeo, youtube, and any page with html5 video to be rendered with madvr for much better quality.

An obstacle I see though is madvr is 32bit and chrome is moving onto 64bit

[vivan]

Since Chrome and Firefox are based on open source code projects, is it possible that in future, madvr can be implemented as an option as a video renderer since most video is now delivered using html5 and rendered by the browser now.It would be very cool.
madVR is a DirectShow filter and accessible to any application. So it's not related specifically to madVR - to get madVR supported in browsers, those browsers should just start supporting DirectShow.

[makz]

What are you displaying to test in MPC-HC+madVR?



AVS HD test patterns
In MPC-HC i watching 16-235 in Grayscale Steps. As I understand it, the range is not affected on clipping

[sheppaul]

According to this post, intel has dedicated hardware scalers and provides the best scaling similar to lanczos and ar, while amd and nvidia are using bi-linear and shaders.

http://www.avsforum.com/forum/26-hom...l#post23847902

Hmm, interesting. BTW, I can confirm that dxva2 scaling on AMD is not certainly bilinear.
A sample to verify: http://forum.doom9.org/showthread.php?p=1705360#post1705360

Bilinear
http://fs1.directupload.net/images/150215/sg39kxf6.jpg

DXVA2
http://fs2.directupload.net/images/150215/3trnrq4p.jpg

[huhn]

Hmm, interesting. BTW, I can confirm that dxva2 scaling on AMD is not certainly bilinear.
A sample to verify: http://forum.doom9.org/showthread.php?p=1705360#post1705360

on my AMD 270 DXVA scaling is clearly bilinear.

[sheppaul]

on my AMD 270 DXVA scaling is clearly bilinear.

This is pretty strange. Change a video renderer to EVR default or try windows media player which is using EVR by default.

I'm using "14.12 AMD Catalyst Omega driver" with HD7850.

[huhn]

This is pretty strange. Change a video renderer to EVR default or try windows media player which is using EVR by default.

I'm using "14.12 AMD Catalyst Omega driver" with HD7850.

EVR is not using bilinear.

so there are 2 possibilities:

madVR has a bug and bilinear is used when DXVA scaling is used. or EVR is not using DXVA scaling.

edit: looks like it is a bug in madVR. I did a small test look here: http://abload.de/img/evrsplita9ubm.png
this is EVR with split demo mode.
would be nice if someone can check madVR again with this sample and see if it is bilinear for you too. after that I will create a bug entry in the bug tracker.

[sheppaul]

EVR is not using bilinear.

so there are 2 possibilities:

madVR has a bug and bilinear is used when DXVA scaling is used. or EVR is not using DXVA scaling.

Are you sure? EVR default seems to use dxva2 scaling while EVR C/P uses bilinear by default. I have a same result with potplayer supporting dxva2 scaling now. Would you open the sample with widnows media player and let me know what it looks like? WMP uses a vanilla EVR and it looks like this in my screen which is obviously not a bilinear and it's the same as dxva2 scaling.

ps. I don't use any post-processing filter in catalyst control center.

http://fs2.directupload.net/images/150215/2hzkqlv9.jpg

[huhn]

can you please show what your results with madVr if madVR uses biliniear for you too it is very clear a bug in madVR.

both EVR in MPC-HC and windows media player don't use bilinear for me but madVR with dxva does.

[michkrol]

edit: looks like it is a bug in madVR. I did a small test look here: http://abload.de/img/evrsplita9ubm.png
this is EVR with split demo mode.
would be nice if someone can check madVR again with this sample and see if it is bilinear for you too. after that I will create a bug entry in the bug tracker.
To my understanding, what your image shows is:
on the left DXVA scaling with video enhancements off,
on the right DXVA scaling with video enhancements on.

The "bug" is probably as simple as madVR disabling video enhancements on your system, but not for some other people and that's the reason you get different results - different driver's settings.

[huhn]

To my understanding, what your image shows is:
on the left DXVA scaling with video enhancements off,
on the right DXVA scaling with video enhancements on.

The "bug" is probably as simple as madVR disabling video enhancements on your system, but not for some other people and that's the reason you get different results - different driver's settings.

the test was made to show that AMD DXVA2 with "enhancements off" is bilinear.

so you didn't get bilinear with DXVA in madVR using a AMD card? this would mean it is a bug my side which i'm totally fine with.

[sheppaul]

can you please show what your results with madVr if madVR uses biliniear for you too it is very clear a bug in madVR.

both EVR in MPC-HC and windows media player don't use bilinear for me but madVR with dxva does.

I've already posted the result with madVR in the two screenshots above. It's pretty strange I have a same result with MPC-HC/BC/WMP.

The "bug" is probably as simple as madVR disabling video enhancements on your system, but not for some other people and that's the reason you get different results - different driver's settings.
I have all the video enhancement settings disabled. I don't think madVR did something different here. The only difference I can see is I'm using Windows 8.1.

so you didn't get bilinear with DXVA in madVR using a AMD card? this would mean it is a bug my side which i'm totally fine with.
If then this is not a problem of madVR as I can get the same result with other players. Pretty strange.

[tjcinnamon]

Since Chrome and Firefox are based on open source code projects, is it possible that in future, madvr can be implemented as an option as a video renderer since most video is now delivered using html5 and rendered by the browser now.

Right now, almost all web video is scaled using billinear and does not take advantage to dxva scaling, for those with high res monitors, such as even when watching 4k youtube on the 5k imac, it makes the picture very soft and blurred.

This would allow web video such as vimeo, youtube, and any page with html5 video to be rendered with madvr for much better quality.

An obstacle I see though is madvr is 32bit and chrome is moving onto 64bit

YES YES YES!!!! THIS ^^^

I have UFC fight pass and the "HD" stream is "720p" (which I'd be okay with) if there weren't so much aliasing (jaggies). It's crazy. I started playing around with the HTML to see if I could stream it through to MPC-HC (and madvr).

Does such a product exist to push a videostream to MPC-HC? Perhaps that would be a way to leverage mad-vr to improve most of the highly compressed web-content

[huhn]

YES YES YES!!!! THIS ^^^

I have UFC fight pass and the "HD" stream is "720p" (which I'd be okay with) if there weren't so much aliasing (jaggies). It's crazy. I started playing around with the HTML to see if I could stream it through to MPC-HC (and madvr).

Does such a product exist to push a videostream to MPC-HC? Perhaps that would be a way to leverage mad-vr to improve most of the highly compressed web-content

livestreamer can do this.

of course doesn't work with all pages.

[detmek]

Hmm, interesting. BTW, I can confirm that dxva2 scaling on AMD is not certainly bilinear.
A sample to verify: http://forum.doom9.org/showthread.php?p=1705360#post1705360



I have done a test with your file on Intel Pentium G3220. For some reason DXVA scalling won't activate in madVR if output from decoder is RGB and it defaults to Bilinear. I had to disable RGB output in LAV to test DXVA scalling. Even with that I had a problem - green color lines, right and at the bottom. And EVR has same problem with disabled RGB output. Here is what I got:
EVR:http://www.dodaj.rs/t/3W/RK/4GrGzoGm/evr.jpg (http://www.dodaj.rs/?3W/RK/4GrGzoGm/evr.png) DXVA:http://www.dodaj.rs/t/5/LD/1edhpSkk/dxva.jpg (http://www.dodaj.rs/?5/LD/1edhpSkk/dxva.png) LanczosAR:http://www.dodaj.rs/t/3s/bb/1zFpRDLL/lanczorar.jpg (http://www.dodaj.rs/?3s/bb/1zFpRDLL/lanczorar.png) Bilinear:http://www.dodaj.rs/t/40/8E/bA7CdHs/bilinear.jpg (http://www.dodaj.rs/?40/8E/bA7CdHs/bilinear.png) EVR-disabled RGB: http://www.dodaj.rs/t/z/VP/Qpd3O20/evr-disabled-rgb.jpg (http://www.dodaj.rs/?z/VP/Qpd3O20/evr-disabled-rgb.png)

I am not sure what this means

[sheppaul]

Look at the screen closely and you'll know that there is a pixel shift to left upper in EVR (DXVA scaling). It looks like a genuine bug of EVR and it seems to cause the problem with the dxva scaling of madVR. There is a same shift in EVR and DXVA2 resizer. That's the main reason why I consider both are same.

BTW, the screenshots are from iGPU? EVR and DXVA screenshots look different. :o I'm not sure it's because both are using different resizers or additional processing of madVR affects the result. Would you try the sample with potplayer? I get the exactly same result with EVR and DXVA2 scaling with HD7850 at least.

[ikakun]

On my GPU, EVR-CP(set to bilinear) & madvr DXVA2 scaling looks the same. Plain EVR isn't.

BTW, using NNEDI3 on that test sample, doubling @ 16 neurons + quadrupling @ 16 neurons + jinc3ar image scaling give me better result than increasing the neurons further with or without quadrupling. And here I thought higher NNEDI3 neurons is better. Or is it safe to say that it depends on what source video you're playing? Like anime & live(or should I say real life?) video for example.

[ryrynz]

Or is it safe to say that it depends on what source video you're playing? Like anime & live(or should I say real life?) video for example.

Yes, that's it exactly.

[detmek]

Look at the screen closely and you'll know that there is a pixel shift to left upper in EVR (DXVA scaling). It looks like a genuine bug of EVR and it seems to cause the problem with the dxva scaling of madVR. There is a same shift in EVR and DXVA2 resizer. That's the main reason why I consider both are same.

BTW, the screenshots are from iGPU? EVR and DXVA screenshots look different. :o I'm not sure it's because both are using different resizers or additional processing of madVR affects the result. Would you try the sample with potplayer? I get the exactly same result with EVR and DXVA2 scaling with HD7850 at least.

Yes, I noticed the shift myself but there is a difference between EVR and madVR DXVA scalling as EVR is a bit blurry. Also, on RGB output DXVA scalling won't work for madVR and EVR. Both renders default to Bilinear.

It's iGPU, I don't have discrete GPU.

Here are screenshots with PotP. I used default settings, only changing renders.
EVR:http://www.dodaj.rs/t/3P/jb/2Y0uBi/potevr.jpg (http://www.dodaj.rs/?3P/jb/2Y0uBi/potevr.png) madVR DXVA: http://www.dodaj.rs/t/8/m/5Lmdwk2/potmadvrdxva.jpg (http://www.dodaj.rs/?8/m/5Lmdwk2/potmadvrdxva.png) madVR Bilinear: http://www.dodaj.rs/t/E/Hw/2rFlsn8T/potmadvrbilinear.jpg (http://www.dodaj.rs/?E/Hw/2rFlsn8T/potmadvrbilinear.png)

EVR does use DXVA scalling as I got same bug and same difference in sharpness. Maybe a driver problem.

[sheppaul]

on RGB output DXVA scalling won't work for madVR and EVR. Both renders default to Bilinear.
DXVA2 Scaler seems not working with RGB output. It is understandable as DXVA2 decoder also doesn't work with RGB output.

[madshi]

DXVA2 Scaler seems not working with RGB output. It is understandable as DXVA2 decoder also doesn't work with RGB output.
That is correct for AMD, but not for Intel. I noticed this problem with AMD about a week ago. It seems I'll have to go back to scaling NV12 -> NV12, when using DXVA. A pity, really.

[detmek]

madshi, then why DXVA scaler won't work with RGB on my Intel GT1 GPU (Pentium G3220)?

[Asmodian]

DXVA2 decoding and deinterlacing outputs NV12 surfaces. Unfortunately pixel shaders can't use them directly. So madVR has to convert the NV12 surfaces somehow to make them pixel shader compatible. There are 3 different ways madVR can use to do that. Solution (1) is copyback (downloading the NV12 data to CPU RAM, then re-uploading it to GPU in a different format). Solution (2) is conversion/processing via OpenCL, introduced in v0.87.0. And solution (3) is a copy operation on the GPU, which can be lossless with AMD GPUs, but produces a slightly blurred chroma channel with NVidia and Intel GPUs. There's a trade-quality-for-performance option to switch between (1) and (3). And there's a new option in the "rendering -> general settings" to enable/disable (2).

Does this mean that AMD users should enable the trade quality for performance options "don't use 'copyback' for DXVA deinterlacing (Intel, NVidia)" and "don't use 'copyback' for DXVA decoding (Intel, NVidia)"? And the reason it doesn't mention AMD is that no quality is lost? :confused:

[nevcairiel]

Does this mean that AMD users should enable the trade quality for performance options "don't use 'copyback' for DXVA deinterlacing (Intel, NVidia)" and "don't use 'copyback' for DXVA decoding (Intel, NVidia)"? And the reason it doesn't mention AMD is that no quality is lost? :confused:

Those options should not do anything at all on AMD, thats why it has the vendor tags behind it.

[Asmodian]

Those options should not do anything at all on AMD, thats why it has the vendor tags behind it.

Thanks. So on AMD copyback is never used?

[Warner306]

For anyone using LAV Filters and DXVA2 Copyback, the latest LAV Filters (0.64) contains improvements in DXVA2 decoding. I downloaded the new filters and discovered I was able to turn up my madVR settings while actually lowering the rendering queue. I consider this a considerable performance upgrade.

[bcec]

I have a basic question.

If I have am doing 1080->2160 upscaling, it is basically perfect image doubling right? If I have image doubling enabled, which uses NNEDI3 with16 neurons, I don't quite understand why image upscaling settings also affect the performance.

I thought if I have perfect image doubling, only image doubling setting would be used, and it doesn't matter what I select in image upscaling section. So why does image upscaling selection affect the performance, when I have image doubling enabled and doing 1080p->2160p?

[Asmodian]

I have a basic question.

If I have am doing 1080->2160 upscaling, it is basically perfect image doubling right? If I have image doubling enabled, which uses NNEDI3 with16 neurons, I don't quite understand why image upscaling settings also affect the performance.

I thought if I have perfect image doubling, only image doubling setting would be used, and it doesn't matter what I select in image upscaling section. I guess I don't quite understand how image doubling and image upscaling play along with each other.

If you are only doubling the luma then the chroma also needs to get bigger somehow. Chroma scaling only applies for the conversion from 4:2:0 or 4:2:2 to 4:4:4. Image scaling is used to bring the chroma up to the target resolution after doubling the luma.

[bcec]

If you are only doubling the luma then the chroma also needs to get bigger somehow. Chroma scaling only applies for the conversion from 4:2:0 or 4:2:2 to 4:4:4. Image scaling is used to bring the chroma up to the target resolution after doubling the luma.

ah, that makes sense. thanks for the explanation.

Is there a noticeable difference between using Jinc3AR vs Lancsoz3AR for upscaling the chrome portion, when NNEDI3 was used to upscale luma? I see quite a bit performance difference between the two.

[Asmodian]

ah, that makes sense. thanks for the explanation.

Is there a noticeable difference between using Jinc3AR vs Lancsoz3AR for upscaling the chrome portion, when NNEDI3 was used to upscale luma? I see quite a bit performance difference between the two.

Short answer: No. Even Catmull-Rom looks great for chroma to me. If doing a perfect 2x upscale I think going down to catmull-rom for chroma and image upscaling is perfectly reasonable.

edit: I should add that madshi has stated that Jinc is probably a waste if using image doubling, even if not at a perfect 2x scaling.

[bcec]

Short answer: No. Even Catmull-Rom looks great for chroma to me. If doing a perfect 2x upscale I think going down to catmull-rom for chroma and image upscaling is perfectly reasonable.

edit: I should add that madshi has stated that Jinc is probably a waste if using image doubling, even if not at a perfect 2x scaling.

Thanks a lot mate!

[madshi]

So you can't replicate madVR lowering render time after exiting out of, and then back into fullscreen?
Couldn't reproduce it with a quick test.

I have in MadVR http://i58.tinypic.com/dqr89e.jpg and when I start a .wtv with 25fps I get display at 59.9Hz, then I press alt+tab to go on Windows Media Center (Win8.1) and then again alt+tab on MPC-HC and this time I get display at 50Hz and the image is more saturated.
What should I do?
Alt+Tab should make no difference. Are you sure that your TV/GPU supports 1080p25 output? Often only 1080p50 is supported. If you enter an invalid mode, the madVR behaviour can be "weird". You should only enter modes which are actually available by the OS/GPU driver.

I'm jus thinking about something. Why do the video profiles have to be configured manually? Thinking about it, it could be auto-configured pretty easily.

Let's say I'm running with SVP and madVR has 16.6ms to render each frame. What I would manually do is open each resolution of video (288p, 360p, 480p, 720p, 1080p) in Widescreen, press CTRL+J and create a profile that brings the rendering time as close to 16.6ms as possible.

madVR could do these tests and build these profiles itself.

or... perhaps even better, it could try various settings until it gets close to its maximum allowed rendering time, and cache these results. If there are dropped frames, it would automatically lower the settings, if rendering time gets lower, it would automatically higher the settings.

Is something like this in the road-map?
Dynamically switching between different profiles, depending on how high the rendering times are, is a "dangerous" thing to do. Basically it could happen that madVR would then switch back & forth between two different profiles all the time. And the switching itself costs a bit performance, as well.

I have all sorts of ideas and plans for madVR, but I'm not ready to talk about it now.

Yes! there is an option in the "Adjust Desktop Color Setting" menu for Limited and Full ranges.

How cool is that?
Took'em 10 years but still.... thanks.
Finally! :)

anyone please kindly help how to fixed the error?
This is a crash in the NVidia OpenCL driver. Not sure why it crashes. You could try reinstalling the NVidia driver or something. If you can't get it to work, you may have to disable NNEDI3, which will then probably "fix" this crash.

When I open a video file in windowed mode, the file is always same resolution than default, nothing change.
Madvr calibration tab I set 1080p24 and 1080p60.
I think normal if I open a 480p file, it need to open with 1080p resolution for properly upscalling, no?
The media player decides in which size the video is rendered. madVR does what the media player wants. So you need to look for an option to start in fullscreen (or something similar) in your media player.

Has anyone else had any persistent issues with UAC pop-ups upon running a video asking if I want to allow madHcCtrl.exe to make changes to my computer?
Strange. madHcCtrl doesn't really do anything funny, except maybe trying to find madVR instances running somewhere on your (W)LAN, if you have network functionality enabled. You could try disabling network functionality and see if that removes this weird Windows behaviour.

I found that enabling the 3Dlut help with the color and gamma, but introduces a large number of artifacts when watching shows with large amounts of constant flat color for instance Anime (Sky, faces, buildings, ...), and close ups of faces. For now I have disabled it while looking for the cause and a solution.
I used the default which contained about 1800 samples to create the 3Dlut, I'm thinking of trying with more samples, but also wanted to ask is this expected?
Try updating to the latest Calman Beta/RC. They've added some bugfixes to create better 3dluts. The issue is probably in the 3dlut itself, now in madVR.

please give an option to disable 'on top' for the madVR settings dialog.
Ok, will add it to my to do list.

what an amazing video renderer this is! :)

i just discovered it 2 weeks ago just when i had bought the new Pioneer BDP LX58 blu ray player.
that blu ray player was a waste of money and i sold it right away when i compared the picture quality with madvr.

with a calibrated 9200 patches 3D LUT the picture quality is amazing.
the LX58 player doesnt even come close in details and color sharpness in the picture.

so you know that you have the best picture quality money can buy with this solution :)
Glad you like it! The other users did make some good suggestions to optimize your settings, though.

My video is starting in full screen, it is not an issue. But it is not starting in exclusive mode. When i use alt+j, it show me i am in windowed mode, not exclusive. Also the blue seek bar of the exclusive mode is not there in mpc.
Most probably "something" (e.g. some other window) is in front of the video rendering area. I can say for sure if you create & upload a debug log with the OSD (Ctrl+J) turned on.

Do you have "delay playback start until render queue is full" and "delay playback start after seeking, too" enabled?

I get very variable audio sync without having them selected.
i hear about sync issue with madVR quite a lot.
Really? madVR carefully present frames synced to the graph master clock. I don't see how madVR could produce any audio sync issue, regardless of any settings. FWIW, the user who reported this issue a few posts ago in the meanwhile found out that there isn't really any audio sync issue between Jinc and NNEDI3 for him, after all, and that the audio sync issue is limited to one specific video file.

Believe I found a bug in madvr. It does not appear to be releasing its system timer (tick rate) hold on unload.
Correct. If you want this fixed, could you please create a bug tracker report? Personally, I don't find this very important because the 16ms timer is idiotic, anyway. Maybe it made sense 10 years ago. But with today's PCs 16ms timer resolution simply makes no sense, whatsoever. Of course that's only my personal opinion. I do consider it "not nice" that madVR doesn't restore the original timer resolution, though.

Can madVR handle MVC MKV 3D files that are made with Makemkv?
Simple remuxes? Or side-by-side encodings? "No" for simple remuxes.

Was this implemented/fixed already? I'm having a lot of issues with this lately, and not only with screenshots.
What do you mean "with this, and not only with screenshots"? Your quote is specific to screenshots. So I don't understand the "not only with screenshots" part. Anyway, nothing has changed so far. It's on my to do list, but not with an ultra high priority.

Did you ever managed to get madVR as video processor work with Stereoscopic Player?
madVR is not meant to be used as a video processor, at least not as this time.

I have a question concerning the Display Switcher: I have made four custom resolutions by using CRU Utility to play videos with fps of 23.976, 24.000, 25.000 and 29.970. Windows is silly so there is no possibility to distinguish between the 23.976 and 24.000 preset
Ideally those should be mapped to 23, 24, 25 and 29 Hertz. That way madVR would be able to properly assign them to the correct framerates.

What would help me is the implementation of this logic (C style):

if (framerate == 23.976fps) display mode = 720p48;
if (framerate == 24.000fps) display mode = 720p72;
This is currently not possible, except maybe by using file name tagging for those 24.000fps movies. Some day I'm going to change the display mode switcher logic. But for now it is what it is.

BTW time to convince madshi to accept donations so he can focus more on madvr, earning the money with this product - no more need to focus on other things
That would be awesome, but not very realistic. My commercial projects produce a nice income. Getting the same income from donations is next to impossible.

Turned out to be every 15 minutes to the second. Apparently the current version of raptr (AMD Gaming Evolved) is doing something that causes that stutter every 15 minutes.
Ouch!

So does nobody know how to set the display modes so that 1080p content is sent unscaled to the 4K TV (allow the TV to do the scaling) and 4K content is sent @4K? "1080p23, 2160p30" results in 2160p content being downscaled to 1080p30. Just curious why this is occurring.
When looking for the best matching display mode, madVR puts the highest priority on matching refresh rates. The reason for that is that a mismatch in the frame rate is a bigger problem than having to up/downscale. So if the movie is 24fps, 1080p23 is a better match than 2160p30, because when using 2160p30 you'd get 3:2 pulldown judder. Use "1080p23, 2160p23", then madVR should switch modes just the way you want it.

Hi, I have some problem with MPC-HC and Madvr 3DLut. Trying on my Dell 2312. I create madvr .3dl and all fine, but there is differnce in measurments HCFR between madTPG and MPC-HC+madvr. My problem with rendering 100% color with 3dlut (Dell 2312 has oversatturated native green).
Weird. How did you test this with MPC-HC+madVR? HCFR can only use madTPG to show test patterns, not MPC-HC+madVR. So you had to use different test patterns for MPC-HC+madVR, right?

So you get the correct/expected results when using madTPG? And incorrect results with MPC-HC+madVR? Is there any difference between enabling/disabling the 3dlut in MPC-HC+madVR? Maybe the 3dlut is simply not applied at all when using MPC-HC+madVR? Or is it applied, but somehow incorrectly? That is important to find out.

I don't understand why I see all bars in AVS HD 709 (1- Black Clipping.mp4) when I choose in madVR/devices/properties 16-235 and only 17 and above when I choose 0-255 and raise brightness control at 100. It seems illogical, my graphic card output 0-255.
Those numbers "16-235" and "0-255" tell you which color codes are used for black and white. When using "16-235", black is sent with 16, and white with 235. In this setup, because black is 16, you can send colors which are darker than black. This is called BTB ("blacker than black"). When using "0-255", black is sent as 0. So it's impossible to send any BTB information. BTB would be negative values, which isn't supported by DVI/HDMI. Those bars from 0-15 in "Black Clipping.mp4" are BTB information. So they can only be sent to the display when using 16-235.

The patches HCFR generates using madTPG are limited range.
madTPG expects the calibration software to always send TV levels = limited range. So basically HCFR has to send "black" as 16. This is just a communication rule between the calibration software and madTPG, though. madTPG uses the madVR settings to actually display those test patterns. So if madVR is configured to output PC levels, madTPG will output "black" as 0, just like madVR does. Which means that madTPG and madVR (+ any media player) should produce identical results.

madshi, then why DXVA scaler won't work with RGB on my Intel GT1 GPU (Pentium G3220)?
We were talking about RGB *output*, not *input*. DXVA scalers don't support RGB input. But some support RGB output.

Thanks. So on AMD copyback is never used?
Correct.

[madshi]

madVR v0.87.14 released

http://madshi.net/madVR.zip

* madTPG: fixed "madVR_Get/SetSelected3dlut" API
* madTPG: added "madVR_Enable3dlut" API
* madTPG: network search is a bit more extensive/agressive now
* fixed: madVR network functionality didn't always take WLAN into account
* target rect related change for DVBViewer
Again, only changes for calibration software (ArgyllCMS, Calman, LightSpace) and for DVBViewer. Maybe I'll find some more time for madVR development soon, but I can't say for sure right now...