madVR - high quality video renderer (GPU assisted)

[yukinok25]

Have you tried the tips in this post?

http://forum.doom9.org/showthread.php?p=1724688&highlight=problems#post1724688

Nope! I didn't see that post!Thanks!

I'll try and report back!

Just one question, I deactivated the "how many video frames shall be presented in advance" feature, as I noticed my video was a bit stuttering.

I also disabled the Vsync on the nvidia control panel, and everything worked well after that.

Is the setting "video frames presented in advance" really useful/important, or can I leave it off instead?

[aufkrawall]

Shiandow, do you also take a look at SuperRes' AA/AR with your current SR revision?

Taken my low-res cartoon example, both features can produce some weird artifacts like doubled lines where only one should be.
This happens e.g. with super-xbr luma doubling + SR AR or NNEDI luma doubling + SR AA.
This is why I have them disabled.
Not severe since SR still gives a very decent result when the other parameters are adjusted accordingly, but it'd be nice to have a good anti-ringing filter for super-xbr.

[Telion]

Sharpness algorithms are there to emulate that.
That is why I think anime or cartoons are no indicator or the right way to tweak a sharpness algorithm which WILL be applied on camera captured movies also.
In other words, Anime, Fonts and Test Patterns are NOT good test images for sharpness algorithms.
You are right in your grounds, but your conclusion is not.
Sharpness algorithms are also to compensate for blurring brought by upscaling. This is a really distinct issue with animated content which usually has lots of prominent solid contours encompassing saturated colour gradients. Animation and live action are indeed two very different types of content. So the correct way to tune up sharpness algorithms would be not to sacrifice one of them for the sake of another, but to produce two different sets of corresponding parameters which can be easily switched by toggling some global option with a hotkey.

[madshi]

And *always* responding politely.
A class act.
Thanks... :)

Unlike camera captured material, anime is digitally or hand drawn with ideal and maximum sharpness.
Anime is not blurred (low pass filter or MTF) by the lens or electronics to prevent aliasing or moire that is captured by camera.
Anime or cartoons ARE high contrast material with sharp lines filled with color.

IMO, sharpness algorithms are there so that we can compensate for "less than ideal" camera traits.
This is why super-sampling then dowscaling looks sharper, or capturing with 5K camera and downscaling to HD look WAY better than capturing with HD camera to HD (the Hobbit for example).
Sharpness algorithms are there to emulate that.

That is why I think anime or cartoons are no indicator or the right way to tweak a sharpness algorithm which WILL be applied on camera captured movies also.
In other words, Anime, Fonts and Test Patterns are NOT good test images for sharpness algorithms.
It's true that Anime, fonts and test patterns are quite different to content shot with a camera, but Anime is well loved by many users (including me), so we can't ignore it. Also, Anime sometimes shows artifacts that also occur in normal video content, just Anime shows them stronger. So if we can make Anime content look good, that can help with normal video content, too. Just look at super-xbr: It was original written for enlarging cartoonish computer games. But now it's mutated into a very nice upscaling algorithm for normal video content. But yes, there will be situations where optimizing for Anime may result in different settings than you'd otherwise use. In such cases we need a good compromise.

Dam, it's so cool how 1080p game captures look with high bitrate @WQHD, thanks to FineSharp, super-xbr and SuperRes. I'd almost think it is rendered in realtime in native resolution. :eek:
And 60fps aren't any problem performance wise.
:)

I've been following the discussion a bit and the problem might be that you're trying to do everything in linear light. For a simple unsharp mask filter you tend to get better results if you blur the image in linear light but subtract the unsharp mask in gamma light (or maybe even L*a*b). Unfortunately I couldn't immediately figure out how to apply this to finesharp, but at first glance the RemoveGrain11, FineSharpB parts should be done in linear light, for RemoveGrain4 it doesn't matter, so the LL artefacts are probably caused by either FineSharpA or FineSharpC.
Yeah, you're probably right. Maybe I'll investigate that at some point. But right now I don't have the time, unfortunately.

Just one question, I deactivated the "how many video frames shall be presented in advance" feature, as I noticed my video was a bit stuttering.

I also disabled the Vsync on the nvidia control panel, and everything worked well after that.

Is the setting "video frames presented in advance" really useful/important, or can I leave it off instead?
For windowed or for fullscreen exclusive mode (FSE)? I generally have that setting on, but if playback works better for you with it turned off that's ok, too.

You should set most/all settings in the NVidia control panel to "application control", so madVR can tell the GPU/driver what it wants.

Shiandow, do you also take a look at SuperRes' AA/AR with your current SR revision?

Taken my low-res cartoon example, both features can produce some weird artifacts like doubled lines where only one should be.
This happens e.g. with super-xbr luma doubling + SR AR or NNEDI luma doubling + SR AA.
This is why I have them disabled.
Not severe since SR still gives a very decent result when the other parameters are adjusted accordingly, but it'd be nice to have a good anti-ringing filter for super-xbr.
I don't think trying to fix the ringing after super-xbr is going to work too well. madVR's AR algorithm already tries to make sure that super-xbr doesn't add ringing which wasn't there in the first place. If the original source already has ringing in it, also SuperRes will try its best to keep the ringing in the upscaled image, and to *not* remove it. What we could really use is a good de-ringing algorithm which would run before upscaling and remove ringing from the source. Then super-xbr would work better, and SuperRes, too!

[tFWo]

Some Finesharp settings i found i like on good old live action movies. (for a change :P )


- suggested strength for image enhancement (depends on grain and source quality, 1080p, no upscaling) - 0.4-0.6 low, 0.8-1.0 medium, 1.4-1.6 high (>2.0 ultra)
* thinning - 0.010 - 0.020 seems ok

- strength for upscaling refinement (720p, s-xbr doubling) - 0.6-0.8
* thinning - 0.000 - 0.005 - any noise or blocking is boosted by a large amount even with the lowest settings

- strength for upscaling refinement (SD, s-xbr quading) - <=0.4
* thinning - 0.000 - same as above and even more pronounced



s-xbr observations on same content

- used NNEDI32/64 for image doubling before trying s-xbr, can't find any reason to go back to NNEDI -_- . s-xbr is faster, sharper and i really can't find major drawbacks for image doubling use.

- tried it for chroma uspcaling on SD content and compared to NNEDI64 s-xbr is worse/different??? It's sharper, but there is some strange detail loss in some places.

[mogli]

With D3D11 I do have lower rendering and presentation times. Is this a real performance gain or some pseudo effect like things happening behind madVRs back?

Edit: Clock state stays the same as with D3D9.

[aufkrawall]

You should check if your GPU is running with the same clock as before.
If clock is higher due to higher load, reported rendering times can in fact be lower.

[James Freeman]

Sharpness algorithms are there to emulate that.
That is why I think anime or cartoons are no indicator or the right way to tweak a sharpness algorithm which WILL be applied on camera captured movies also.
In other words, Anime, Fonts and Test Patterns are NOT good test images for sharpness algorithms.

You are right in your grounds, but your conclusion is not.


It's true that Anime, fonts and test patterns are quite different to content shot with a camera, but Anime is well loved by many users (including me), so we can't ignore it.

OK, Agreed.
I can live with a slightly brighter image, but for anime the ringing and haloing is definitely ugly.

After some more testing I can see that above around 1.5, LL shows its bad side around sharp objects.
No good for anything but the most blurry images.
So once again, I definitely can live without LL with a slightly brighter look with higher sharpening strength.

OK, I'll be testing 88.11 Finesharp with the following:
Image Enhancements
LL Off
Repair 1.0
Thinning 0.020
Strength variable

I'll try to match a 480 and 720 downscaled images sharpness to the original 1080 one when they are the same size.
Not HD but surely closer with Sharpening than without.
http://www.mediafire.com/convkey/e1d9/9lza1a8b183f3k9zg.jpg

[Shiandow]

Shiandow, do you also take a look at SuperRes' AA/AR with your current SR revision?

Taken my low-res cartoon example, both features can produce some weird artifacts like doubled lines where only one should be.
This happens e.g. with super-xbr luma doubling + SR AR or NNEDI luma doubling + SR AA.
This is why I have them disabled.
Not severe since SR still gives a very decent result when the other parameters are adjusted accordingly, but it'd be nice to have a good anti-ringing filter for super-xbr.

I have taken a look at those, unfortunately SuperRes and super-xbr have more or less the same problems with images containing ringing. To scale those images "accurately" you'd also need to scale the ringing, but that leads to bad results. I'm beginning to suspect that it might be better to use a separate de-ringing algorithm before scaling the image. Of course you'd then need a good de-ringing algorithm but that might be easier to achieve than trying to make SuperRes not scale something accurately.

At least I've managed to prevent SuperRes from introducing any aliasing, so when combined with super-xbr or NEDI you shouldn't need anti-aliasing, which didn't work too well anyway.

Oh by the way I think I might know why SuperChromaRes didn't preserve dark lines well. You can lose luma information when clipping out of gamut RGB values, normally this doesn't cause much of a problem but each SuperChromaRes pass will do this again, so the luma will drift even more. If that is the problem then it can be easily fixed, and the new SuperRes revision should also help.

[aufkrawall]

Thanks a lot!

[Eyldebrandt]

Don't use too high values for now, though, I still need to fine tune the repair parameter. It seems to need higher repair values if you use higher thinning values.

I've make some tries today, and the repair function clearly seems to improve the aliasing control introduced by xstr.
So, i'll keep digging that way.


Sounds good to me, but I suppose it's not available as a HLSL pixel shader yet? Where can I find that adaptation, in any case?


Nope, unfortunately, there is only an avsi for that.
Here he is https://drive.google.com/open?id=0B4X2WPMIHRQma1hsbUtnMGZlUDg&authuser=0

I hope it will help.

[ryrynz]

Oh, so disabling the "present a frame for every vsync" fixes this issue for you? What's your refresh rate and your movie frame rate? That option should only have an effect when the refresh rate / movie frame rate don't match exactly.


Only the changing the queues does anything to fix the dropped and glitched frames after seeking. Refresh was 60hz and source was 23.976 fps.
I just mentioned the present a frame issue as it's frame dropping / glitching related.

I have taken a look at those, unfortunately SuperRes and super-xbr have more or less the same problems with images containing ringing. To scale those images "accurately" you'd also need to scale the ringing, but that leads to bad results. I'm beginning to suspect that it might be better to use a separate de-ringing algorithm before scaling the image.

Yeah, when I tested super-xbr I can't say I was impressed with the results in this area, sounds like a good idea.

[rack04]

Has anyone else noticed that double chroma resolution is automatically selected when you select double luma resolution? When I unselect it I don't get the option to apply which means the program doesn't know a setting has changed.

[mindbomb]

quick question about image doubling. When you select super xbr or nedi, it becomes selected automatically for chroma, but when I check the stats with ctrl+j, it only shows super xbr/nedi for luma though. Does it only work on luma?

[ryrynz]

Has anyone else noticed that double chroma resolution is automatically selected when you select double luma resolution? When I unselect it I don't get the option to apply which means the program doesn't know a setting has changed.

That's by design for NEDI and super-xbr, both luma and chroma are upscaled and this cannot be changed.

[yukinok25]

For windowed or for fullscreen exclusive mode (FSE)? I generally have that setting on, but if playback works better for you with it turned off that's ok, too.

You should set most/all settings in the NVidia control panel to "application control", so madVR can tell the GPU/driver what it wants.

Thank you very much for your feedback Madshi!

You made me think about that I setup MPC to use exclusively only my nvidia GPU, but not actually MadVR, maybe that could be the reason..

[viewer]

Source: H.264 8bit 4:2:0

1. chroma upscaling: JincAR3
2. chroma upscaling: NNEDI16
3. chroma upscaling: Super-XBR
4. chroma upscaling: JincAR3 with SuperRes
5. chroma upscaling: NNEDI16 with SuperRes
6. chroma upscaling: Super-XBR with SuperRes

SuperRes setting:
passes: 2
strength: 0.75
sharpness: 0.25
softness: 0.10
anti-aliasing: 0.50
anti-ringing: 1.00

1 4
2 5
3 6
http://s5d4.turboimagehost.com/t/23368978_up.png (http://www.turboimagehost.com/p/23368978/up.png.html)

http://s5d4.turboimagehost.com/t/23368979_upx3.png (http://www.turboimagehost.com/p/23368979/upx3.png.html)


--

additional question, why this rule does not working?
if (targetWidth != srcWidth) "A"
else "B"

[ryrynz]

1 4
2 5
3 6

http://s5d4.turboimagehost.com/t/23368979_upx3.png (http://www.turboimagehost.com/p/23368979/upx3.png.html)


5 & 6 are looking best if I go by the removal of the darker outline around the speech bubble which I believe would be the most accurate.

[MistahBonzai]

Re: 31176 by Viewer.

Talk about pixel peepin' :) My display is 10-bit 4:4:4 and after crawling all over the test images the only artifacts that really stand out to me is the ringing or whatever on the straight edges in 3&6. It's very noticeable in 3, not so much in 6 (but still there). I guess 1&4 look the cleanest to me. In anycase spotting the differences in 'real life' would take a very practiced and focused eye - after all this is chroma processing. BTW: Great test/images.

[kiwijunglist]

Hi,

I was hoping some one could recommend me some settings.

My setup (dedicated htpc):

PC: i3-2100 / HD7750 2GB / Win7x64 / HDMI out
Receiver: Pioneer Elite SC-LX87 - When running as Full RGB it clips (disgards) 0-16 and 235-255 levels.
Display: Samsung ua60h6400 - 1080P Series 6 TV. TV Accepts 8bit/10bit/12bit colour as input. Panel is likely 8bit.
Source material - YUV 4:2:0 8bit

With madVR i have to either set 16-235 levels with RGB Full in CCC or/ 0-255 levels with RGB Limited in CCC otherwise I get crushed blacks. Strangely my AVR reports Limited RGB whether I set CCC to Full or Limited. The AVR is set to pass through and performs no video conversion.

[B]I don't need to worry about desktop colour being washed out, I don't play any games I only watch mkv on the htpc.

Can someone suggest what settings i should run (Source from LAV DXVA2 copyback is YUV 4:2:0 8bit)
MadVR 0-255 vs 16-235?
MadVR: 8 bit vs 10+ bit colour?
CCC: 8 or 10 or 12 bit colour depth. ITC on or off?
CCC: YCrCb 4:2:0 / YCrCb 4:2:0 / RGB Full / RGB Limit?

I was imaging it would be best to do:
Mad VR 16-235 / 10+ bit / 12 bit / ITC on / RGB Full ?

Also I notice that if I set colour to 12bit in CCC, and 10bit in MadVR, when i press ctrl+j it says "D3D11 exclusive (10bit)", not 12bit.

Thanks
Kiwi

[x7007]

Is super-xbr has better quality for 1080p-720p movies than NNEDI3 ? or when something last on the list is always better ?

Should I need Super-Res filter for 1080p-720p ?

[MSL_DK]

In older versions rendering times were averaged "forever". That means you got nicely stable times after a while. *However*, directly after switching algos or changing zoom factors or things like that, rendering times are usually much higher than a couple of seconds later. This means that old way of long-time averaging often started too high and needed a looooong time to settle down to the correct values. The new method averages rendering times over just 1 second of playback. So naturally the rendering times can jump quite a bit. However, if the rendering times over multiple consecutive measurement periods are relatively near to each other, madVR averages them together again. So as long as the jumps are not too large from one measurement second to the next, the times should calm down over time, too, and then get much quicker to the correct end result compared to before.

Yes, but try to compare these two recordings of what is going on. Do not know if it has anything to do with the latest versions have not made use of "reduced banding"

One with reduced banding low, high (https://www.dropbox.com/s/b7dga0dcc4h0nxe/fil%2017-06-15%2013.08.02.mov?dl=0)

One without (https://www.dropbox.com/s/kgpxvgma9siolpd/fil%2017-06-15%2013.09.21.mov?dl=0)

Let me add that I avoid dropped frames by raising present queue to 14 also 14 at render.

[aufkrawall]

Is super-xbr has better quality for 1080p-720p movies than NNEDI3 ? or when something last on the list is always better ?

Should I need Super-Res filter for 1080p-720p ?
As a simple rule, I'd suggest the following:
-With low scaling factors, even NNEDI3 is hardly better than Jinc.
-for medium scaling factors, super-xbr gives a similar quality to NNEDI3 with much lower performance costs
-for extreme scaling factors, there is no way around NNEDI3, best used with 32 neurons or 64 (or more, if performance wouldn't be an issue)

720p -> 1080p is scaling factor 1.5, that's not nothing and there might be content with which super-xbr can already look superior to Jinc. I think you can use it if performance isn't an issue, but it wouldn't be a dealbraker if performance wasn't sufficient and you'd have to stick to Jinc.

I think SuperRes always makes sense if upscaled, as long as it isn't set too aggressive.
I chose less aggressive values for myself to make it look good with any scaling factor and instead I additionally also rely on FineSharp as an image enhancement. It makes blocking worse, but in the end low quality sources never look really acceptable (especially not without deblocking filter) and I can live with the result and don't have to switch settings for different kinds of source quality.

[Werewolfy]

Thanks for the feedback. Yes, I've also seen artifacts from SuperRes AR. Personally, I do like to use FineSharp for upscaling refinement, provided you use a good quality source and a good image doubling algorithm. E.g. try a very sharp downscaled image (preferably one which doesn't have ringing artifacts), quadruple resolution to 400%, then enable FineSharp in upscaling refinement. Try turning sharpness down to 0 and play with just the thinning. Then do it the other way round. That should give you a good feeling for what either of these parameters do.


Ok good idea, I will try that this week end.


How about the test build?



I tried it and you were right, it's definitely between LL on and off. The picture dosen't get a lot brighter like LL off and the dark contouring on anime is less pronounced. If you don't want to offer a LL option in Fine Sharp, your test build represents a decent compromise but still not a perfect one.

But honestly I find very optimistic to think you can set 3 presets for sharpness... My TV has 4 options for sharpness, 2 of them have 100 levels. I agree that there is too many options atm but to reduce that at only 3 presets is maybe a little bit too extreme. What about just a strength setting for each sharpness filter? For Fine Sharp, you could for example set the strength and the thinning for each level. I don't say we have to find 100 levels of course but it will be a more flexible solution I think. You only have one parameter that controls everything but remains very flexible like you find on a TV.

[MSL_DK]

Yes, but try to compare these two recordings of what is going on. Do not know if it has anything to do with the latest versions have not made use of "reduced banding"

One with reduced banding low, high (https://www.dropbox.com/s/b7dga0dcc4h0nxe/fil%2017-06-15%2013.08.02.mov?dl=0)

One without (https://www.dropbox.com/s/kgpxvgma9siolpd/fil%2017-06-15%2013.09.21.mov?dl=0)

Let me add that I avoid dropped frames by raising present queue to 14 also 14 at render.

A question regarding banding. Occurs it from the player to the screen or is it a part of the source? Is the function at all necessary when playing 1: 1 bluray? I can easily find examples where banding occurs and where the function overcomes this but at the expense of detail, prefer still using reduced banding.

[huhn]

this is a source problem and comes from bad mastering and from encoding.

[Asmodian]

A question regarding banding. Occurs it from the player to the screen or is it a part of the source? Is the function at all necessary when playing 1: 1 bluray? I can easily find examples where banding occurs and where the function overcomes this but at the expense of detail, prefer still using reduced banding.

Banding can occur both when rendering the video to the screen or it can be part of the source. With madVR visible banding is from the source. Lots of blurays have bad banding in the source.

[TheElix]

I have been waiting for someone to report what I'm seeing but I guess I'm the only one. Since the introduction of DX11 version, whenever I choose it I get flickering video 30 second into the playback. And it's not disappearing until I restart the player or revert back to DX9. By flickering I mean alternating video frames with black frames. Like suddenly the video goes to 3D mode and shutter glasses activates over the screen. Please, tell me if this issue has been encountered already. I have a Radeon HD7990 with 15.5 Beta Drivers (Win 7 x64).

[leeperry]

Yeah, when I tested super-xbr I can't say I was impressed with the results in this area, sounds like a good idea.
Yeah, I've run quite a lot of tests and in combination with SR I still prefer NEDI for DVD/720p and NNEDI3 for tiny videos.

NEDI looks far more natural, I prefer to set sharpness using SR. The "non-double defaults" SR presets with a sharpness of 28 looks fantastic to me. XBR looks sharper but far too digital to my taste atm and not up to NNEDI3's eagle eye edge detection for 360*288 hopeless videos, so I dearly hope that NEDI won't be ditched http://forum-images.hardware.fr/images/perso/zigg.gif

[Hyllian]

I just updated Super-xBR shaders (for Retroarch. I don't know how and if madshi will introduce it on madVR).

Now the user can tweak the filter weights with just one knob. It's a param that can vary from 0.0 to 2.0:

0.0 means bilinear filtering.
0.65 is equivalent to cubic filtering.
1.29 is sinc filtering (the default value). This is the biggest sane value.
Above 1.29 is insanely sharp, which can introduce other artifacts or make the image artificial. Anyway, I think it can please some users.

[Warner306]

Some help from experienced madVR users wanted...

Below is a link to a chart I created ranking madVR's processing features on its ability to influence and improve picture quality. This chart is to be part of an online guide.

The madVR settings hierarchy is based on my personal experiences with madVR. Each setting is ranked in a defined order of importance. The details are explained here:

Chart Removed

Any comments on its accuracy? Is the order correct?

[huhn]

10 bit output is just wrong.

10 bit output on a 10 bit display just has less noise with less banding nothing else. nothing gets more vivid nothing has more CR.

image enhancement works on all sources not only upscaled ones.

[Warner306]

10 bit output is just wrong.

10 bit output on a 10 bit display just has less noise with less banding nothing else. nothing gets more vivid nothing has more CR.

image enhancement works on all sources not only upscaled ones.

The difference in 10-bit output appears much improved on my display (a 1080p plasma). The picture appears more saturated to my eyes. Where would you put this feature, instead?

[JarrettH]

Any preference between xbr and superres here?

bicubic75 AR chroma in all

jinc
https://farm1.staticflickr.com/402/18904396025_2782c711d3_o.png

xbr (catmull downscaling with LL and AR on)
https://farm1.staticflickr.com/453/18281846934_f9aabca1df_o.png

superres
https://farm1.staticflickr.com/462/18716710950_180c4121c8_o.png

superres has less halos in these pics. i also feel the changes in superres are more faithful to the jinc upscaled original.

[Asmodian]

Some help from experienced madVR users wanted...

Below is a link to a chart I created ranking madVR's processing features on its ability to influence and improve picture quality. This chart is to be part of an online guide.

The madVR settings hierarchy is based on my personal experiences with madVR. Each setting is ranked in a defined order of importance. The details are explained here:

Guide: madVR Image Settings Hierarchy (http://forum.kodi.tv/showthread.php?tid=223175&pid=2030806#pid2030806)

Any comments on its accuracy? Is the order correct?

I would also delete the (detail) in the image upscaling and SuperRes sections. It is false to conflate "luma" with "detail" even if most of the fine detail is in the luma plane.

10-bit output should be the least important, shouldn't it? Unless you have a display with extremely high contrast differentiating dithered 10-bit and dithered 8-bit is almost impossible.

Anyway the idea of ranking based on impact on image quality is... not very accurate; I see what you mean but there are many situations where your order is completely wrong (even after ignoring 10-bit). The recommended uses do not make sense to me. Your SuperRes advantage description is wrong too, it is not doing calculations to decide which pixels to selectively sharpen.

Debanding is not more or less important depending on the output bitdepth; it is for removing banding in the source. Debanding is run long before dithering to the output bitdepth and the banding debanding addresses is not the same banding >8-bit output helps with.

Image upscaling is less important when using image doubling, not more.

Why is chroma scaling more important for watching videos in their native resolution?

Dithering doesn't require colored noise and it isn't really adding noise to provide an illusion of higher bitdepth. It is distributing the error to nearby pixels in such a way that much of the higher bitdepth information is preserved. This increases the apparent noise in the image. In a way dithering exchanges (apparent) precision in the spacial resolution for an (apparent) increase in bit-depth.

[Warner306]

I would also delete the (detail) in the image upscaling and SuperRes sections. It is false to conflate "luma" with "detail" even if most of the fine detail is in the luma plane.

10-bit output should be the least important, shouldn't it? Unless you have a display with extremely high contrast differentiating dithered 10-bit and dithered 8-bit is almost impossible.

Anyway the idea of ranking based on impact on image quality is... not very accurate; I see what you mean but there are many situations where your order is completely wrong (even after ignoring 10-bit). The recommended uses do not make sense to me. Your SuperRes advantage description is wrong too, it is not doing calculations to decide which pixels to selectively sharpen.

Debanding is not more or less important depending on the output bitdepth; it is for removing banding in the source. Debanding is run long before dithering to the output bitdepth and the banding debanding addresses is not the same banding >8-bit output helps with.

Image upscaling is less important when using image doubling, not more.

Why is chroma scaling more important for watching videos in their native resolution?

Dithering doesn't require colored noise and it isn't really adding noise to provide an illusion of higher bitdepth. It is distributing the error to nearby pixels in such a way that much of the higher bitdepth information is preserved. This increases the apparent noise in the image. In a way dithering exchanges (apparent) precision in the spacial resolution for an (apparent) increase in bit-depth.

Thanks for the input. I will rethink the chart.

I will start by revisiting 10-bit output on my own display. I didn't think it was placebo but I will recheck.

[huhn]

The difference in 10-bit output appears much improved on my display (a 1080p plasma). The picture appears more saturated to my eyes. Where would you put this feature, instead?

the saturation should be totally the same double check your for black clipping and white clipping. plasma are "1 bit" displays anyway. that doesn't mean it is worseless on them.

the more a processing a display does the more useful 10 bit can be even for 8 bit displays.

in the end it reduces dither noise and still doesn't add banding and that's way ED is pretty "useless" for it.

[Warner306]

the saturation should be totally the same double check your for black clipping and white clipping. plasma are "1 bit" displays anyway. that doesn't mean it is worseless on them.

the more a processing a display does the more useful 10 bit can be even for 8 bit displays.

in the end it reduces dither noise and still doesn't add banding and that's way ED is pretty "useless" for it.

I still think my display looks better in 10-bits, but I'll remove it from the pyramid as I had overstated its importance.

Talking to the two posters above always keeps me from spreading misinformation, so thanks.

[ryrynz]

I just updated Super-xBR shaders

Now the user can tweak the filter weights with just one knob.

I think it can please some users.

For sure, let's hope it makes it's way into madVR.

[MysteryX]

If anyone is interested, I just released the software I had been working on.

It has a SVP Setup Wizard that makes it easy for anyone to configure SVP and madVR to their optimal settings (although you guys already know how to do it)

It also has a Media Encoder that allows enhancing SD videos in just a few clicks, using NNEDI3, EEDI3 (which is way too slow for doing in real-time), and you can even increase the frame rate to 60fps. I'm using it to upscale noisy 288p VCDs into 720p, and from there madVR further upscales it for my display. The difference is stunning.

Btw, thanks Madshi for the idea of applying sharpening after every double. Although I still can't use SuperRes in AviSynth, I've split the final sharpening into two sharpenings applied at different stages and it does allow me to get a crisper image while keeping the result natural. AviSynth is also a pain in the *ss to use. This tool make it accessible to anyone. Or make it practical to process 50 videos in a row.

You can get the Natural Grounding Player here on Source Forge (http://naturalgroundingplayer.sourceforge.net/) (it's open source). The application itself has a different purpose, but the SVP Setup Wizard and Media Encoder parts of it can be very useful on their own.

[MSL_DK]

Yes, but try to compare these two recordings of what is going on. Do not know if it has anything to do with the latest versions have not made use of "reduced banding"

One with reduced banding low, high (https://www.dropbox.com/s/b7dga0dcc4h0nxe/fil%2017-06-15%2013.08.02.mov?dl=0)

One without (https://www.dropbox.com/s/kgpxvgma9siolpd/fil%2017-06-15%2013.09.21.mov?dl=0)

Let me add that I avoid dropped frames by raising present queue to 14 also 14 at render.

I have found out what happens. It's pretty simple but equally annoying. My AMD graphics card switches on and off "High Performance Mode"

this is a source problem and comes from bad mastering and from encoding.

Banding can occur both when rendering the video to the screen or it can be part of the source. With madVR visible banding is from the source. Lots of blurays have bad banding in the source.

Thanks for the information.

[huhn]

I still think my display looks better in 10-bits, but I'll remove it from the pyramid as I had overstated its importance.
and that is totally possible.

[JustinChase]

If anyone is interested, I just released the software I had been working on.

It has a SVP Setup Wizard that makes it easy for anyone to configure SVP and madVR to their optimal settings (although you guys already know how to do it)

It also has a Media Encoder that allows enhancing SD videos in just a few clicks, using NNEDI3, EEDI3 (which is way too slow for doing in real-time), and you can even increase the frame rate to 60fps. I'm using it to upscale noisy 288p VCDs into 720p, and from there madVR further upscales it for my display. The difference is stunning.

Btw, thanks Madshi for the idea of applying sharpening after every double. Although I still can't use SuperRes in AviSynth, I've split the final sharpening into two sharpenings applied at different stages and it does allow me to get a crisper image while keeping the result natural. AviSynth is also a pain in the *ss to use. This tool make it accessible to anyone. Or make it practical to process 50 videos in a row.

You can get the Natural Grounding Player here on Source Forge (http://naturalgroundingplayer.sourceforge.net/) (it's open source). The application itself has a different purpose, but the SVP Setup Wizard and Media Encoder parts of it can be very useful on their own.

Sounds promising, but "Whoops, we can't find that page."

[ThurstonX]

Sounds promising, but "Whoops, we can't find that page."

If you google "Natural Grounding Player" the first hit should be:

https://www.spiritualselftransformation.com/blog/conscious-business/conscious-ressources/natural-grounding-player-v1-0-is-here/

Perhaps there's a naming issue to be worked out. While that's happening, I think madshi should look into a method for doubling or even quadrupling a video's native vibrations as the next enhancement... :D

[MysteryX]

If you google "Natural Grounding Player" the first hit should be:

https://www.spiritualselftransformation.com/blog/conscious-business/conscious-ressources/natural-grounding-player-v1-0-is-here/
The whole project disappeared from Source Forge. I contacted them to know what is going on and am waiting for an answer.

The v1.0 doesn't have the Media Encoder feature.

Perhaps there's a naming issue to be worked out. While that's happening, I think madshi should look into a method for doubling or even quadrupling a video's native vibrations as the next enhancement... :D
WOW that's a GREAT idea!

Did you know you can take 5 videos with a certain vibration, and combine them into a production that has a higher vibration than any of its individual video because of the synergy between them and the state of the producer?

That's a perfect example of that. I prefer the remix than the original.
https://www.youtube.com/watch?v=KOffsOHOCX0

[aufkrawall]

Upcoming Catalyst 15.20 driver is supposed to have improved OpenCL drivers (page 3):
http://www.moorinsightsstrategy.com/wp-content/uploads/2015/06/AMD-Brings-New-Value-to-Radeon-by-Moor-Insights-and-Strategy.pdf

[MysteryX]

I think madshi should look into a method for doubling or even quadrupling a video's native vibrations as the next enhancement... :D
Here are a few ways. The only way to alter the vibration of a video is to do a remix. A remix can be done with a higher level of synergy between the elements, of from a higher state of consciousness. Or, you can literally add a silent audio in the background carrying a high vibration. Someone also took a guided meditation, removed all the negative words from it, and ended up with a new version that had a higher vibration.

Another way is to change the way you relate to it. Enhancing the image quality does make it more real and thus you relate more to it.

SVP does greatly alter the way you relate to the vibration of the video. When you view a video in 24fps, the brain knows it is a film and processes it with its frontal cerebral lobes. When viewing the same video at 60fps, however, it believe it to be real and processes it with the cortex in the back of the head. Thus, with SVP, you relate with the vibration of the video at a much more primal level and it has a deeper effect on your psyche.

[leeperry]

Upcoming Catalyst 15.20 driver is supposed to have improved OpenCL drivers (page 3):
http://www.moorinsightsstrategy.com/wp-content/uploads/2015/06/AMD-Brings-New-Value-to-Radeon-by-Moor-Insights-and-Strategy.pdf
Finally a replacement for 13.12 or does swapping the OpenCL DLL's actually work out equally well?

When viewing the same video at 60fps, however, it believe it to be real
It takes at least 72fps to fool the human brain AFAIK. Easy to achieve with a deinterlaced NTSC video accelerated to 72Hz :)

[huhn]

Finally a replacement for 13.12 or does swapping the OpenCL DLL's actually work out equally well?

adding the DDL to the mpc-hc folder worked really good but i didn't used this longer than a couple of mins. it just a hack to my eyes.

[noee]

Finally a replacement for 13.12 or does swapping the OpenCL DLL's actually work out equally well?


I would not recommend this, it did not help at all on my Win7Ult setup with old HD6570.

But I guess it could depend on h/w.