madVR - high quality video renderer (GPU assisted) [Archive] - Page 1266

flossy_cake

1st July 2022, 13:02

Regarding power consumption, I'm finding some decent power savings on my GTX 1070 by setting "adaptive" power profile for the media player's process -- in my case mpc-hc64.exe. The amount of saving appears to fluctuate depending on the scene though -- on some scenes I'm getting as much as 40 watts savings, but on a typical scene like 1080p Star Trek episode it's about 20-25 watts savings.

Also tried enabling CPU downclocking & core parking through Windows power settings and found that could save an extra 5 watts but I didn't think it was worth it (i7-4790k).

Finally saved about 100 watts by setting the screen saver to a blank screen and set the taskbar to "auto hide" to prevent burn in.

SirMaster

1st July 2022, 16:13

@SirMaster

I'm curious how this actually works. The "no compression limit" parameter seems to mean that the content remains untouched below it. To achieve this, it's not possible to use a display peak luminance equal to 50. So it seems to me that in this situation, a display peak luminance above that is chosen (say at 150 nits), which guarantees that content below the compression limit is not actually compressed. But then the result must be scaled down to 50 nits, and this final result is expanded to 85 nits by your projector.

For brighter highlights, the original BT.2390 roll-off would be much simpler if it were available.

Edit: I don't know how the result can be scaled down, unless what is sent is not in the full range [0,1]. Can you check if setting a display peak nits higher makes the result different with the rest of your settings? I would say that in this situation it is not used at all.

Setting a higher peak definitely affects the picture. If I set a higher peak, pretty much everything gets dimmer.

NCL is inherently set to DPL, or rather, can't be lower than DPL. So if I put NCL to 0, it would really be 50. But setting it to 85 causes bright things to get brighter, and yet keeps the dark things darker which to my eye increases the dynamic look of the image.

It also allows HSTM to work down to 50 nits rather than down to the real DPL which would be higher.

This makes HSTM have a larger effect. But I still use the conservative HSTM curves and never see any flicker because I am using the the conservative curves that don't seem to cause flicker.

I don't know how the algorithms work or how they combine in this case, but I would say this is not at all an intended way to use these controls. But the point is after much experimentation I still like and even prefer them for my use.

On my 100 nit actual calibrated PC monitor, I often compare these 3:

DPL=100, NCL=0
DPL=50,NCL=0
DPL=50,NCL=100

I find I always prefer the last.

DPL=100 is the "correct setting".
DPL=50, NCL=0 just makes everything brighter and doesn't really look better. Near black gets too gray, etc as everything is rendered brighter than it's "supposed" to be.
DP=50, NCL=100 something "magic" happens. Near black, dark stuff stays dark like it should be, mid tones and bights get brighter like they are with DPL=50. Overall dynamic look is increased to my eye, with no noticeable downside.

I picked DPL 50 because I find that's the lowest I can take it before things start to "fall apart" or look flat, etc in various ways.

And I do keep DTN enabled and that also set at 50, so the target is not locked to 50, but uses the DTN algorithm.

So my conclusion has been, set DPL to 50, and NCL to real display nits. But this only scales to about 100 nits. If I have more than 100 nits, I will start to raise DPL then.

I just don't like how dark the image looks overall when the display has less than 100 nits, and you put in the real DPL which was the whole reason I played with the settings like this.

I don't claim this is correct or best for everyone, just that I know I like it. And friends who have similar setups I have showed them the comparison too and they have come to prefer it as well.

I didn't have a chance to get screenshots, but I will post some tonight.

Sunspark

1st July 2022, 18:19

Finally saved about 100 watts by setting the screen saver to a blank screen

This is a really stupid Windows 10 bug. I could hear the CPU fans whining while it was in screensaver mode. I checked the GPU load and it was spiked up high when the screensaver is active.. even ones that have been around for 25 years. Blank screen is the only one that doesn't spike the power now.

Alexkral

1st July 2022, 20:18

I don't know how the algorithms work or how they combine in this case...

Me neither, but even though it's clear that HSTM is quite different from BT.2390, it doesn't seem like NCL is really doing what it's supposed to.

SirMaster

1st July 2022, 21:34

Me neither, but even though it's clear that HSTM is quite different from BT.2390, it doesn't seem like NCL is really doing what it's supposed to.

Oh good point.

I know HSTM doesn't work below DPL, but it definitely works below NCL.

madshi did say he could make HSTM work down to 0. I can't remember or not if he actually enabled that option in one of the more recent builds.

Yep, found the changelog for it.

https://www.avsforum.com/threads/improving-madvr-hdr-to-sdr-mapping-for-projector.2954506/page-701#post-61351881

"3) Added contrast recovery option "with dark scenes, as well". By default, contrast recovery (HSTM) activates only if the frame peak is higher than the display peak. But this new option allows you to always turn HSTM on. Doing this will in some way go against director's intent (or at least against what the disc asks for), so I don't recommend using this new option. But you're the boss of your home theater, of course."

70MM

1st July 2022, 22:28

flossy_cake

1st July 2022, 23:07

Dithering is an very cool and awesome technique that can make 8bit as good as 10-12bits, and the dither pattern ("noise") will be totally invisible if it is coloured subpixel dithering and temporally changing every frame. Even non-temporal, if done right, will be practically invisible. You can check this by taking a screenshot with and without dithering. If you open the screenshot in Photoshop and zoom right in and inspect pixel values you should see small fluctuations of +/- 1 steps depending on the pattern.

Asmodian

1st July 2022, 23:17

I prefer non-color dithering and not changing every frame. It is completely imperceptible to me. Changing every frame makes it more noticeable.

SirMaster

2nd July 2022, 06:35

Me neither, but even though it's clear that HSTM is quite different from BT.2390, it doesn't seem like NCL is really doing what it's supposed to.

Here is a simple screenshot comparison to show what I like about the DLP and NCL I use.

I am going to use a 100 DPL baseline assuming your display was 100 nit for example.

https://screenshotcomparison.com/comparison/25235

Here I am comparing:

Default: DPL100 - NCL0 (this is how most would recommend madVR, with correct DPL, NCL=DPL=100 by default here)
Mouseover: DPL50 - NCL0 (DPL half the true value, NCL=DPL=50 by default here)

The obvious result is everything gets brighter, and more compressed?, because DPL is lower of course. This doesn't look any higher dynamic range or anything. This is bad, and I wouldn't use DPL50 - NCL0 like this on a display with 100 real nits.

Now compare these:

https://screenshotcomparison.com/comparison/25236

Here I am comparing:

Default: DPL100 - NCL0 (same as before, the "correct DPL")
Mouseover: DPL50 - NCL100 (DPL50 like before, but this time increase NCL to 100 to match real display nits)

The result is the overall picture is a similar brightness, except now:

The darks are a bit darker (but not clipping or losing shadow detail to my eye).
The mids are about the same as before.
The brights are brighter (but not clipping either).

To me this has a look of a higher dynamic range, and almost like a "veil" is lifted.

So this is how I run my madVR. DPL=50, NCL=real display nits. And this works well in my opinion for displays between 50-100 nits (most projection setups).

For 120 nit projector I might consider something like DPL=(50 + (real nits - 100)) = 70, NCL=120

SirMaster

2nd July 2022, 06:40

SirMaster do you set your video card to RGB Full 8bit or 10bit?
Also I see you tick under dithering Error Diffusion option 2 with "use colour noise" and "change dither for every frame" but both these add noise it says. Why do you "tick" these ( I have never ticked them due to add noise) and what benefits do they actually bring to the table. Many thanks...

I don't use change every frame anymore normally I looked at my HTPC and it's off.

But I used color dithering because it reduces luma noise (which is more noticeable than chroma noise) in theory?

I don't really think I could tell a different either way anyways.

SuperLumberjack

2nd July 2022, 11:05

Hi !

I have a problem with MadVR, I think since at least 3 months, but I didn't notice it because I didn't watch lots of movies with my computer these last months (I watch lots of things on Amazon Prime Video).

I almost sure that the problem came with the last versions of the Nvidia drivers. It's always getting worse...

In fact, I have dropped frames when I display a movie at 23,976 or 24 Hz. I haven't when I stay with 60 Hz.

But when I use D3D 11 with MadVR, the dropped frames are not showed in the statistics, only when I use D3D 9.

I only have this with MadVR, not with EVR Custom Pres. for example.

I tried to disable LAV Video and use the internal filter of MPC-BE instead, tried to disable the A/V sync, but the problem stays the same.

You can see it in this video: https://mega.nz/file/7pNGySpZ#kr3JRZHdgj9BPr-rWRETCeyUXtSGur7Ki0uKm-lCcpU

Strangely, when I tried to record the problem with ShadowPlay or OBS, it looks smooth as you can see in the 1st part of the video :p Maybe a problem of vsync...

But in fact, I have stuttering like in the 2nd part of the video. And with D3D 9 in MadVR, sometimes I have black frames which are appearing repeatedly (not as long that in my record) or just dropped frames.

I would like to install an older version of my Nvidia drivers, but the problem is the automatic updates of GeForce Experience and the drivers that I can't disable. Thanks Nvidia...! :rolleyes:

Thank you for your help please ! ;)

SuperLumberjack

2nd July 2022, 13:02

The problem is solved!

I had to enable the vsync manually in the Nvidia control pannel. It's stable now.

But it's strange that I have to do this from one day to another...

Damn Nvidia! :angry:

flossy_cake

2nd July 2022, 13:10

I noticed that ticking the setting "scale chroma separately, if it saves performance" is saving me 40 watts on 4k60HDR video.

I prefer non-color dithering and not changing every frame. It is completely imperceptible to me. Changing every frame makes it more noticeable.

Changing it per frame should reduce the visibility of noise, kind of like how flashing a pixel black and white fast enough should start to appear as grey, I think it's called temporal integration or something.

My favourite dithering algo is from CeeJay's SweetFX (part of ReShade) which does randomised noise added to green and subtracted from red and blue:

https://i3.lensdump.com/i/temKJb.png
https://i.lensdump.com/i/tempe7.png
https://i1.lensdump.com/i/tem7yr.png

Alexkral

2nd July 2022, 16:39

To me this has a look of a higher dynamic range, and almost like a "veil" is lifted.

Yes, this clearly looks better. I get the same result on my lcd display by setting the DPL below the real peak nits, AND using the original BT.2390 roll-off.

I made some shaders some time ago to be able to do this, since madVR uses the DisplayCAL roll-off. I'll be sharing them at some point when I have time to work on them a bit and do some tests, but I've been so lazy lately and interested in other things, that I don't even remember how some things worked. Now I'm interested in adding other options (BT.2446), so I hope that this will motivate me to return to the subject.

With the DisplayCAL roll-off, you have to set the DPL much lower to get the same brightness in the highlights, so as you've noticed, the result is more compressed, flat and lacking in contrast. And it has other drawbacks too, like color tints in whites.

With the original roll-off the only drawback in comparison is the loss of detail in the very bright highlights, but I can live perfectly well with that.

Asmodian

2nd July 2022, 19:19

Changing it per frame should reduce the visibility of noise, kind of like how flashing a pixel black and white fast enough should start to appear as grey, I think it's called temporal integration or something.

Try it yourself. The dithering noise is much more obvious when set to change every frame.

Moving noise is more obvious than static noise. Humans are very sensitive to motion.

My favourite dithering algo is from CeeJay's SweetFX (part of ReShade) which does randomised noise added to green and subtracted from red and blue:

That is how madVR's random dithering algorithm with colored dithering noise enabled works.

flossy_cake

2nd July 2022, 22:03

Try it yourself. The dithering noise is much more obvious when set to change every frame.

Moving noise is more obvious than static noise. Humans are very sensitive to motion.

That is how madVR's random dithering algorithm with colored dithering noise enabled works.

That's weird cause I've implemented CeeJay's algo into my own temporal dithering shader for my calibration tools app where those screenshots are taken, and I can toggle between temporal and static in real time and view the results, and also play around with frame rate limter RTSS to see how many fps needed for temporal to become invisible. Same observation when toggling between static/temporal dithering in the NVidia driver.

edit: just checked MadVR on an AVSHD709 gradient ramp with it set to Random Dithering, "use colored noise" and toggling "change dither for every frame" and cannot observe a difference. Maybe your monitor uses its own temporal dithering which is somehow clashing/interacting with it? Maybe the slower pixel response time of LCD makes the patterns blur together or something?

edit: here (https://drive.google.com/file/d/1WdtpXFWGsa1X5Fbn2t-AOuYrJWYmaSgG/view?usp=sharing) is the gradient ramp test pattern. When I turn off dithering in MadVR I get obvious banding so it's definitely working, I just can't see any diff between temporal or static.

edit: well the description "hides dither patterns, but subjectively increases noise" doesn't make sense to me since hiding the dither pattern is what reduces noise.

edit: speculation: maybe it's because the temporal dithering is animated at the monitor refresh rate, so if you're in 24hz display mode then you would only get 24fps dithering which I'm looking at now using RTSS to cap at 24fps and yes I can see it is slightly noisier with my nose up to the screen but still not as noisy as static at the same strength. Maybe OLED's instant pixel response makes temporal look even noisier? From memory I could see static dither patterns in the dark tones when I used to have Panasonic OLED and I was wondering why they didn't use temporal, maybe that's why?

Sunspark

3rd July 2022, 21:59

I am the same as Asmodian for dithering. I also like coloured off and change frame off. I just use ordered. The extra power of error diffusion doesn't seem worth it on my hardware and random is noisier.

flossy_cake

4th July 2022, 05:22

https://i1.lensdump.com/i/tnRFQv.png

Hellfire1985

4th July 2022, 20:41

Hi everyone! ive been a member now since 2016 and have been using Madvr for a bit longer than that. i haven't posted one post and i am by far a not an expert but i thought i would share some things that have helped me to get more performance out of my windows 10 pc.

I have found a program called Wtools by WagnardSoft (im not connected with this developer in anyway) and it has a feature to turn on called "Message Signal Interrupt" (Nvidia cards only) which i have found has dropped my rendering times down a fair bit. does anyone know much about this? i have only read that it has something to do with the I/O of the graphics card so it doesn't have to compete with other components??? but any information about this would help and hopefully this could help someone else to get more out of their GPU!?!?

I have also turned off Hardware-accelerated GPU scaling in Windows 10 as well as a few other things, i would like to know if there are any other tweaks that anyone knows of to boost it more? other than the standard Nvidia control panel tweaks? I haven't seen many posts about tweaks so I thought to myself you can only ask the question!

Also too, i have a awesome but old 1080p RGBY 70 inch Sharp Aquos TV that i brought in 2012 which has a great picture and colours and im trying to keep it for as long as i can so i am using 4x DSR to upscale to 4k then downscale to 1080p with Nvidia control panel options either off, application controlled or running the least amount of filtering possible, which does make everything look alot better than it is without using DSR, i noticed that there really is much information in the way of using DSR so any thoughts or suggestions on my setup would be appreciated?

I am running a 3900x @4.5Ghz, x570 Aurous pro wifi, 16gb 3600mhz 16-16-16-32 CR1 @XMP, Gigabyte Xtreme Gaming GTX-1080 @2100mhz all OC's completely stable and cool with zero stuttering

Thank you in advance!

Hellfire1985

4th July 2022, 21:51

Continuing on from my previous post.

As i stated my Sharp TV is RGBY not just the standard RGB, i know it is only adding a yellow sub-pixel but if anyone knows how that works in a display and also would it, if any have an effect on the RGB conversion or anything in Madvr?

My tv also has a option for wide colour gammit and i do use it because when i turn it off the colours wash out, yet windows 10 can tell when a display has WGC but the TV doesn't show up as WCG, so im thinking the TV is too old for Windows 10 to recognize the option.

What doesn't help that Sharp didn't provide many of the TV's spec's not even in the the owners manual, could anyone please help me with giving me more information about this TV?
It is a Sharp Aquos LC70LE735X

Sunspark

4th July 2022, 21:57

I find that ordered not coloured is a tiny bit warmer, you can see that in the 4th stripe from the right.

I also imagine having change pattern every frame off also makes it less noisy. You can see the difference with it on and off if you define your display as 2-4 bit there.

This pairs well with Mitchell-Netravali AR. Has a creamier cinematic tone. With coloured noise on, it feels heavier.

Very subjective I know.

huhn

4th July 2022, 22:28

Hi everyone! ive been a member now since 2016 and have been using Madvr for a bit longer than that. i haven't posted one post and i am by far a not an expert but i thought i would share some things that have helped me to get more performance out of my windows 10 pc.[ /quote]
i will only answer to madVR relate this because this is the madVR thread

[QUOTE]I have found a program called Wtools by WagnardSoft (im not connected with this developer in anyway) and it has a feature to turn on Message Signal Interrupt (Nvidia cards only) which i have found has dropped my rendering times down a fair bit. does anyone know much about this? i have only read that it has something to do with the I/O of the graphics card so it doesn't have to compete with other components??? but any information would help and hopefully this could help someone else to get more out of their GPU!?!?
it supposed to be used by the 3000 series by default.
so there is something behind it why it isn't used on other card do it at your own risk.

I have also turned off Hardware-accelerated GPU scaling as well as a few other things, i would like to know if there are any other tweaks that anyone knows of to boost it more other than the standard Nvidia control panel tweaks? I haven't seen many posts about tweaks so I thought to myself you can only ask the question!
pretty sure madVR doesn't care about this.

Also too, i have a awesome but old 1080p RGBY 70 inch Sharp Aquos TV that i brought in 2012 which has a great picture and colours and im trying to keep it for as long as i can so i am using 4x DSR to upscale to 4k then downscale to 1080p with nvidia control panel options either off, application controlled or running the least amount of filtering possible, which does make everything look better than without using DSR, i noticed that there really is much information in the way of using DSR so any thoughts or suggestions on my setup would be appreciated?
but why would you do that just supersample in madVR to do the same and you will also get proper downscaling not the one provided by nvidia which is not that good...

As i stated my Sharp TV is RGBY not just the standard RGB, i know it is only adding a yellow sub-pixel but if anyone knows how that works in a display and also would it have any effect on Madvr?
i know exactly how that works there is no reason it would create an more accurate image. just pray it doesn't do anything bad.

My tv also has a option for wide colour gammit and i do use it because when i turn it off the colours wash out, yet windows 10 can tell when a display has WGC but it does show up, so im thinking the TV is too old for Windows 10 to recognize the option.
because having a WCG doesn't mean it has a bt2020 designated WCG which windows is talking about.
having a WCG is a standard feature. it's not rare for a TV to loose gamut over the years maybe that's why it is washed out without it.

flossy_cake

5th July 2022, 08:09

I find that ordered not coloured is a tiny bit warmer, you can see that in the 4th stripe from the right.

If you inspect the pixel values in photoshop you will see all pixels are grey and there is no colour in the image. If it looks warm I can't explain that really, maybe it's due to some greyscale nonlinearity of your display causing the coloured dithering to be not quite as uniform as it should be.

el Filou

5th July 2022, 12:08

MSI lowers interrupt latencies but I doubt it has such a big effect on madVR. Anyway it can't be bad and you'll see immediately if it works or not, the device will have immediate big obvious problems if it doesn't.

I've tested hardware accelerated GPU scheduling with a 1050 Ti but I couldn't see any difference with madVR even though it's said that feature helps with lower end systems and I have an ancient CPU.

My [Sharp Aquos LC70LE735X] also has a option for wide colour gammit and i do use it because when i turn it off the colours wash out. As i stated my Sharp TV is RGBY not just the standard RGB, i know it is only adding a yellow sub-pixel but if anyone knows how that works in a display and also would it, if any have an effect on the RGB conversion or anything in Madvr? The only way to use a non-standard native gamut with madVR is this:
- set your TV gamut to native/widest
- profile it and create madVR-compatible 3DLUTs for it with a colourimeter for all the standard gamuts, and use the LUTs with madVR.
IMO the only way that TV's gamut is wider than BT.709 is in the yellows and maybe greens, but you would see very little difference in practice. Not worth the trouble for me.

SirMaster

5th July 2022, 14:32

SirMaster do you set your video card to RGB Full 8bit or 10bit?

I forgot to answer this part.

I use 8-bit as it just works overall more simply than 10-bit and on a projector with around 100 nit luminance range, 8 bits are plenty for smooth gradients, especially with madVR's high quality dithering.

shaolin95

5th July 2022, 20:00

Hello guys.
Suddenly I have lost the extra settings from the Beta version. I tried downloading the latest full version then unpacking the latest beta on it as usual and clicked the install bat but when it finished I STILL only see a few option on the HDR section.
How can I fix this please?
Thanks

Sunspark

6th July 2022, 02:02

If you inspect the pixel values in photoshop you will see all pixels are grey and there is no colour in the image. If it looks warm I can't explain that really, maybe it's due to some greyscale nonlinearity of your display causing the coloured dithering to be not quite as uniform as it should be.

Not coloured is 81,81,81 in the stripe I pointed out as a good place to look, and coloured is 80,81,80 or variations of 80 with 81. So there is a luminance difference however small it might be.

jkauff

6th July 2022, 16:50

djsolidsnake86

6th July 2022, 23:54

hello i have a little problem, i'm using madvr with mpc and lav filters, i was always using dxva2 copyback, now i switched to d3d11 because someone say that it use less energy on the vga/cpu (actually o have a 3060ti)
my problem is this: i use triggercmd with alexa for vocal commands on mpc (using ID commands and autohotkey), with d3d11 many times the vocal commands not work, or work in retard
anyone know why happen this? i have to switch back to dxva2?

Sunspark

7th July 2022, 00:55

shaolin95

7th July 2022, 02:25

Try deleting the registry key (madshi) and your settings.bin file before you re-install.

Thanks.
Tried that but no luck.

jkauff

7th July 2022, 18:24

Thanks.
Tried that but no luck.
Did you re-install the last official version of madVR? The betas only replace a subset of those files.

shaolin95

7th July 2022, 20:14

Did you re-install the last official version of madVR? The betas only replace a subset of those files.

Yes but I will start again just in case with the latest from this thread. Then unpack the beta 159 inside that folder after that and report back :)

Sunspark

8th July 2022, 16:56

I know you guys will deny, but I can see the difference.

On my system with LAV and MadVR in D3D9 overlay mode, I see a visual difference between None (software decoding) and DXVA2-copyback.

With copyback, blacks look a bit blacker, colours look a tiny bit more saturated. With none (this applies to quicksync as well) mid-tones seem slightly brighter, colours are slightly duller. We're not talking a huge difference, but it is a difference.

I think it is a gamma difference. What I don't understand is, why would this happen? Shouldn't all decoders produce identical output? If the answer is yes, then where in the chain would all the remaining pieces interpret the data differently that gamma would be affected? Video driver? Video renderer? Video player?

Why would the gamma be different?

nevcairiel

8th July 2022, 17:16

Seeing is nothing, if there is a difference capture it with screenshots of the same frame, maybe make a histogram to see if colors are different, or disable dithering to see if pixel values actually change.

DXVA-CopyBack is proven to provide the same pixel data as software decoding, and both deliver it the same way to madVR, so it wouldn't even know the difference.

huhn

8th July 2022, 18:09

VQMT results
frame,value
0,1.#INF00
1,1.#INF00
2,1.#INF00
3,1.#INF00
4,1.#INF00
5,1.#INF00
6,1.#INF00
7,1.#INF00
8,1.#INF00
9,1.#INF00
10,1.#INF00
11,1.#INF00
12,1.#INF00
13,1.#INF00
14,1.#INF00
15,1.#INF00
16,1.#INF00
17,1.#INF00
18,1.#INF00
19,1.#INF00
20,1.#INF00
21,1.#INF00
22,1.#INF00
23,1.#INF00

VQMT.exe "C:\Users\might\OneDrive\Desktop\RawScreenshots\New folder\software.yuv420p8le" "C:\Users\might\OneDrive\Desktop\RawScreenshots\New folder\copyback.yuv420p8le" 2160 3840 300 1 results PSNR SSIM MSSSIM VIFP

djsolidsnake86

10th July 2022, 11:18

You need to verify the power consumption on your own system. Different setups will function differently. GPU-Z is good for taking a look at the GPU power consumption in watts and GPU load %.

I think you might be the first person using Alexa to control this stuff, but if it works better in dxva2 mode then use that.

Test out software decoding "None" while you're at it to see if it works better with Alexa.

ok i solved the problem, was a server problem by the skill, i contacted the author
dx11 or dxva is the same in this case
however i don't ever imagine that i was the first that use mpc commands with alexa, is very very useful

bob88

14th July 2022, 21:40

Hello,
Could you please help me. I am getting this very bad color transition. When there is shades of white or black its very uneven and not smooth. I am watching this on a digital projector with MPC-HC and madVR.
I tried to take a screenshot but when I looked at the screenshot on my other 8bit SDR monitor it did not appear to have the same uneven transition so I guess you wont see it either.
I made 2 screenshots with OS HDR on and off. Dont know if it has any relation. Also I dont think its HDR when OS HDR is off even though madVR is set to pass HDR content to the display.
Here is the screenshot. I can see the bad transition on the white guy:
https://i.ibb.co/2kbdHtv/Untitled.jpg (https://ibb.co/yQ3q7y6)
https://i.ibb.co/B4Hn4g7/Untitled.png (https://ibb.co/bHjbHry)

Here is a photo of the bad transition that I took with my phone:
https://i.ibb.co/drSFgSQ/20220714-233450.jpg (https://ibb.co/JKNVQNz)

This is the projector info:
https://i.ibb.co/Pcr6pDV/Screenshot-2022-07-14-224512.png (https://ibb.co/v3cj5XM)
https://i.ibb.co/tJdZQYW/Screenshot-2022-07-14-231144.png (https://ibb.co/pLmWb49)

This is what I have in the Nvidia control panel. Color depth does not go higher than 8 bit...
https://i.ibb.co/dLLLTMN/Screenshot-2022-07-14-234259.png (https://ibb.co/1dddgQC)

This is the video info:
Video
ID : 1
Format : HEVC
Format/Info : High Efficiency Video Coding
Format profile : Main 10@L5@High
HDR format : SMPTE ST 2086, HDR10 compatible
Codec ID : V_MPEGH/ISO/HEVC
Duration : 8 min 52 s
Bit rate : 16.0 Mb/s
Width : 3 840 pixels
Height : 2 160 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 24.000 FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 10 bits
Bits/(Pixel*Frame) : 0.081
Stream size : 1 017 MiB (95%)
Language : English
Default : Yes
Forced : No
Color range : Limited
Color primaries : BT.2020
Transfer characteristics : PQ
Matrix coefficients : BT.2020 non-constant
Mastering display color primaries : BT.2020
Mastering display luminance : min: 0.0000 cd/m2, max: 1000 cd/m2

Any help would be appreciated. Thank you.

huhn

15th July 2022, 00:43

try to disable dxva2 decode and just see if that already fixes it.

bob88

15th July 2022, 01:04

Yes I went to LAV video decoder and changed DXVA2 (native) to (copy-back) or D3D11 and both seem to fix the issue. Not sure which one of the two would be proper to use though. But thank you! :D

huhn

15th July 2022, 01:11

depends on the use case.
with dxva2 copyback everything will work d3d11 is faster but has a limited feature set.

HDMI 2.0 is limited to 8 bit for 4k 60 RGB.
and good luck with your landmine win 11 "nice" to see other user reaching 2 digits in fake ids for displays.

a PJ with 1500 nits sure windows.

bob88

15th July 2022, 11:56

Well my cable is HDMI 2.1. It literally says 8k on it. Bought it for a lot of money.
https://www.ebay.com/itm/381457457295?hash=item58d0a2e08f%3Ag%3A16gAAOSwi8pfLSr5&LH_BIN=1
Are you saying its not really v2.1 because it gives me only 8bit? Should I be getting 10bit with this configuration?
Not sure what fake ids you are talking about. As for the projector its also a pretty expensive one:
https://www.samsung.com/us/televisions-home-theater/tvs/4k-laser-projectors/120-the-premiere-4k-uhd-hdr-smart-tv-2020-sp-lsp7tfaxza/
Specs on the site say the peak nits are 2200.

huhn

15th July 2022, 12:42

looks like the PJ is not HDMI 2.1.

nit is measured in 2 CM^2
lumen is meter^2

bob88

15th July 2022, 14:42

Do you mean to say that the projector does not support HDMI 2.1. Or that the cable is not really 2.1?
The projector supports eARC HDMI which I think is only on HDMI 2.1. so that suggests it is 2.1. Furthermore isnt it suppose to allow 10bit even if it is 2.0?

el Filou

15th July 2022, 16:10

The official stance of the HDMI Forum is that there is no "HDMI version" for devices, only capabilities and bandwidth.
In other words, a device can support a capability that was added with revision 2.1 of the spec, say eARC, but not support the higher bandwidth FRL mode (up to 48 Gbps) that was also added with HDMI 2.1.

18 Gbps TMDS HDMI ("HDMI 2.0") supports 4K 10/12-bit at 24 Hz but not 60 Hz and your display is set to 60 Hz.

bob88

15th July 2022, 16:33

OK... and also from the info that we see in madVR it says that the display supports 10bit and 12bit.
Does this mean that I should look for the fault of it being 8bit in the cable?
Could it be a software problem? I have installed the latest video driver. Not sure what to do more than that...

PS:
It seems you are correct. Taking it down to 30Hz allowed me to select 12bit. However everything seems quite stuttering at 30Hz. Isnt a movie at 24 fps supposed to play smooth on 30Hz?
Does this prove that my cable is HDMI 2.0?

https://i.ibb.co/zQzX5C4/Screenshot-2022-07-15-183826.png (https://ibb.co/jbSy5qr)
https://i.ibb.co/v3KrNPG/Screenshot-2022-07-15-183906.png (https://ibb.co/R3VMX0m)

JaylumX

15th July 2022, 17:09

OK... and also from the info that we see in madVR it says that the display supports 10bit and 12bit.
Does this mean that I should look for the fault of it being 8bit in the cable?
Could it be a software problem? I have installed the latest video driver. Not sure what to do more than that...

PS:
It seems you are correct. Taking it down to 30Hz allowed me to select 12bit. However everything seems quite stuttering at 30Hz. Isnt a movie at 24 fps supposed to play smooth on 30Hz?
Does this prove that my cable is HDMI 2.0?

Do not select 4k x 2k, 3840 x 2160 (native) from the Ultra HD, HD, SD section.
Scroll down and look for 3840 x 2160 under the PC section. It should provide you with more output options. If you are using an HDMI 2.1 cable (or displayport) for Samsung TV's you have to enable "Input Signal Plus" in order to improve the refresh rate, output, resolution and HDR options. Other display brands have their version.