Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion.

Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules.

 

Go Back   Doom9's Forum > Hardware & Software > Software players

Reply
 
Thread Tools Search this Thread Display Modes
Old 9th March 2014, 13:26   #24641  |  Link
Audionut
Registered User
 
Join Date: Nov 2003
Posts: 1,281
Looks to me like linear light has less noise, and better preserves edge detail.

edit: Linear (pure power) also has the best overall luminance match.
__________________
http://www.7-zip.org/

Last edited by Audionut; 9th March 2014 at 13:30.
Audionut is offline   Reply With Quote
Old 9th March 2014, 13:26   #24642  |  Link
James Freeman
Registered User
 
Join Date: Sep 2013
Posts: 919
Quote:
Originally Posted by madshi
Quote:
Now you just need to include this into madVR and I think we should all be satisfied.
Which is not what I wanted...
What have you planned for us...

The results are all over the place... Gamma for cyberbeing, Power for 6233638, Rec.709 for nevcairiel, sRGB for iSunrise,
I like sRGB with 2.2 or Power with 1.92.
The wanted power values are also ranging from 1.92 to 2.60.
__________________
System: i7 3770K, GTX660, Win7 64bit, Panasonic ST60, Dell U2410.

Last edited by James Freeman; 9th March 2014 at 13:34.
James Freeman is offline   Reply With Quote
Old 9th March 2014, 13:34   #24643  |  Link
turbojet
Registered User
 
Join Date: May 2008
Posts: 1,840
Quote:
Originally Posted by madshi View Post
Quote:
Quote:
Originally Posted by turbojet View Post
You are probably right. If I set tv to 16-235 in madvr it looks better but still a bit dark but then I get crushed blacks in windowed/fse. nvidia cp setting had no effect, is there any way to fix this here or is it a madvr issue?
The easiest solution would probably be to use the madLevelsTweaker.exe shipping with madVR to force your GPU to output 0-255 in all modes.
Tried that in both user mode and admin mode, which should it be? It had no effect.
__________________
PC: FX-8320 GTS250 HTPC: G1610 GTX650
PotPlayer/MPC-BE LAVFilters MadVR-Bicubic75AR/Lanczos4AR/Lanczos4AR LumaSharpen -Strength0.9-Pattern3-Clamp0.1-OffsetBias2.0
turbojet is offline   Reply With Quote
Old 9th March 2014, 13:35   #24644  |  Link
leeperry
Kid for Today
 
Join Date: Aug 2004
Posts: 3,477
Quote:
Originally Posted by madshi View Post
And what happens if you view images through LAV Splitter?
PotP won't let me select the image source filter AFAIK, it'll always be "Image source". It would be most amazing if you could nail down the name of the still images source filters used in PotP & MPC and provide a picture-viewer mode in mVR that would kill all black frames(that keep coming and going more or less randomly, making it very hard to use) and allow to setup a custom profile with 256 neurons all the way

Quote:
Originally Posted by madshi View Post
you're looking at everything (shadow detail, hues, saturations) except what I wanted you to look at, namely overall brightness and gamma response. But I guess it's my fault for using oppositeColor
I also much prefer GL in all the screenshots you posted(at 100% scaling) but then again OC is not my thang
leeperry is offline   Reply With Quote
Old 9th March 2014, 13:38   #24645  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
Quote:
Originally Posted by nevcairiel View Post
Watching on my PC screen, BT.709 looks the closest watched from a distance for me, although only Gamma has a very obvious difference, for the others its just minimal.
Yeah, makes sense.

Quote:
Originally Posted by Audionut View Post
edit: Linear (pure power) also has the best overall luminance match.
That matches what I see here, but it depends on the display calibration.

Quote:
Originally Posted by turbojet View Post
Tried that in both user mode and admin mode, which should it be? It had no effect.
Did you reboot after using madLevelsTweaker? If that doesn't help then something else is going on. Make sure you disable (and set to neutral values) all options in the GPU control panel like color/gamma adjustments, dynamic contrast, skin tone adjustments, etc etc. Maybe a driver cleaning and reinstall will help? I'm not sure what else to suggest...
madshi is offline   Reply With Quote
Old 9th March 2014, 13:39   #24646  |  Link
iSunrise
Registered User
 
Join Date: Dec 2008
Posts: 496
If I switch from the sRGB to my Rec.709 calibration target, the BT.709 example suddenly is also closest (which is expected, I guess).

So this really is directly bound to the calibration itself.
iSunrise is offline   Reply With Quote
Old 9th March 2014, 13:43   #24647  |  Link
Werewolfy
Registered User
 
Join Date: Feb 2013
Posts: 137
Quote:
Originally Posted by madshi View Post
My hook/hack should in theory also work in Windows 8, not sure why it's not working for you.
If you have any idea later, I can test or make a debug log for everything you want. I tried different settings, desactivated Reclock, ffdshow,... but it doesn't change anything. I would like to use FSE mode, it's frustrating to see that when the movie begins, madVR switches correctly to 24hz but at the moment I go in full screen it switches to 23hz whatever I do.

To all users on Windows 8 : does it work for you? I can't be the only one... I guess. It's really weird because my setup is very close to 6233638's setup.

Quote:
Originally Posted by madshi View Post
Here's the Avatar dithering comparison again, with added images using different transfer functions for linear light dithering:

|- gamma -|- original -|- linear (pure power 1/0.45) -|- linear (sRGB 2.4) -|- linear (BT.709 1/0.45) -|

Again, make sure you watch these at 100%. Having your browser zoom these will totally screw up everything.

The key point to take from this screenshot comparison is that dithering in gamma light is not correct. I hope everybody can agree with that now? Which image looks best to you will directly depend on how your display is calibrated. The image created with the transfer function nearest to your display calibration should look nearest to the original 8bit image to your eyes.
Personnally I find too that linear light is far closer to the original than gamma light. Between the three, pure power 1/0.45 is the closest one followed by sRGB 2.4 and then BT.709 1/0.45.

Last edited by Werewolfy; 9th March 2014 at 14:04.
Werewolfy is offline   Reply With Quote
Old 9th March 2014, 13:47   #24648  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
Quote:
Originally Posted by madshi View Post
Ok. When did you last test this? Did madTPG already exist then? If not, it might be worth a try testing this again, just to double check the problem still occurs. madTPG changes things a bit because the whole calibration (even 1D lut creation) runs through madVR this way, using high bitdepth rendering. Even when using ArgyllCMS to create GPU gamma ramps, when using madTPG the GPU gamma ramps are never actually used.
I've yet to attempt a calibration with madTPG, as I keep forgetting to enable it. I'll eventually run though everything again, but Argyll still seems to have issues with creating 3DLUTs which fail BTB/WTW clipping tests. I'm not particularly encouraged to spend lot of time testing, until that issue is resolved once and for all.

Quote:
Originally Posted by madshi View Post
Haha, you're looking at everything (shadow detail, hues, saturations) except what I wanted you to look at, namely overall brightness and gamma response.
None of the images you posted thus far are all that close to overall brightness and gamma response of the original when viewed on my display.

Gamma is brighter than the original. BT.709 2.2 is darker than original. The rest are much, much darker.

Quote:
Originally Posted by madshi View Post
Maybe you can test this yourself with an appropriate movie on your display? Using monoColor, in 2-4bit, then in motion toggle linear light dithering on/off, and also switch between 2-4bit and 8bit. I think you will find that linear light dithering is overall nearer to the original, even if it might not be perfect.
What I'd really like to see is a test build which allows me to change the inverse curve used with linear dithering. Please remember that my display is calibrated to a BT.709 curve scaled to ambient light. Essentially it matches BT.709 near-black, and matches L* across the rest of the luminance range. The inverse power curve you are defaulting to with linear dithering, I suspect will always give an extremely inaccurate gamma/brightness result in low bitdepth on my display.

Last edited by cyberbeing; 9th March 2014 at 14:21.
cyberbeing is offline   Reply With Quote
Old 9th March 2014, 13:47   #24649  |  Link
6233638
Registered User
 
Join Date: Apr 2009
Posts: 1,019
Quote:
Originally Posted by madshi View Post
That matches what I see here, but it depends on the display calibration.
Perhaps the solution is to have it follow the settings chosen in "this display is already calibrated" then, and add a selection when loading 3DLUTs?

1/0.45 looks best to me from the images you have posted, and not by a small margin. The others are all too bright.

I think it's very difficult to judge on larger screens, because the dithering is not as effective there.

Quote:
Originally Posted by iSunrise View Post
If I switch from the sRGB to my Rec.709 calibration target, the BT.709 example suddenly is also closest (which is expected, I guess).
What are you calibrating your screen to? BT.709 content should be viewed using the BT.1886 transfer function (which becomes a flat 2.40 gamma on high contrast displays) and never the BT.709 transfer function, which is intended for capture, not playback.
6233638 is offline   Reply With Quote
Old 9th March 2014, 13:48   #24650  |  Link
James Freeman
Registered User
 
Join Date: Sep 2013
Posts: 919
Quote:
Originally Posted by iSunrise View Post
So this really is directly bound to the calibration itself.
Yep,

Now the question remains what would madshi do about it with madVR.

There should be a Standard setting (don't touch if you don't know what it is).
And Customization options for more advanced users or ones that read the thread.

What is the standard gamma curve of modern LCD TVs?
__________________
System: i7 3770K, GTX660, Win7 64bit, Panasonic ST60, Dell U2410.
James Freeman is offline   Reply With Quote
Old 9th March 2014, 13:49   #24651  |  Link
turbojet
Registered User
 
Join Date: May 2008
Posts: 1,840
Quote:
Originally Posted by madshi View Post
Quote:
Originally Posted by turbojet View Post
Tried that in both user mode and admin mode, which should it be? It had no effect.

Did you reboot after using madLevelsTweaker? If that doesn't help then something else is going on. Make sure you disable (and set to neutral values) all options in the GPU control panel like color/gamma adjustments, dynamic contrast, skin tone adjustments, etc etc. Maybe a driver cleaning and reinstall will help? I'm not sure what else to suggest...
Yes, box is checked. nvidia control panel is all defaults which it looks like it's set to all application controlled. I'll try yet another clean install but I just did that yesterday. This has been something I've been noticing for months but it's starting to aggravate me now. It happens on 2 different computers+tv's too, both nvidia gpus.
__________________
PC: FX-8320 GTS250 HTPC: G1610 GTX650
PotPlayer/MPC-BE LAVFilters MadVR-Bicubic75AR/Lanczos4AR/Lanczos4AR LumaSharpen -Strength0.9-Pattern3-Clamp0.1-OffsetBias2.0
turbojet is offline   Reply With Quote
Old 9th March 2014, 13:53   #24652  |  Link
6233638
Registered User
 
Join Date: Apr 2009
Posts: 1,019
Quote:
Originally Posted by James Freeman View Post
What is the standard gamma curve of modern LCD TVs?
There is no standard. I would use 2.2 gamma (1/0.45) if nothing has been specified in the madVR configuration though, as it is most likely to be used if a display does have a decent out of the box calibration.
6233638 is offline   Reply With Quote
Old 9th March 2014, 13:54   #24653  |  Link
Audionut
Registered User
 
Join Date: Nov 2003
Posts: 1,281
Isn't dithering the final step before rendering? In that case, shouldn't it (luminance) map to whatever the device is calibrated too? Based on the settings in the calibration tab.

Otherwise, wouldn't it be best to map back to the source profile?
__________________
http://www.7-zip.org/
Audionut is offline   Reply With Quote
Old 9th March 2014, 13:56   #24654  |  Link
James Freeman
Registered User
 
Join Date: Sep 2013
Posts: 919
turbojet,
I think for the last few driver versions Nvidia forced 0-255 with 60Hz, and 16-235 with 59Hz on HDMI output.
So if you set madLevelsTweaker to full range and refresh rate to 60Hz you get double expansion (clipped black and white).
Or changing to YCbCr444/RGB changes the range too in NCP.

It can be my display so don't take my word for it, test yourself.
__________________
System: i7 3770K, GTX660, Win7 64bit, Panasonic ST60, Dell U2410.

Last edited by James Freeman; 9th March 2014 at 14:05.
James Freeman is offline   Reply With Quote
Old 9th March 2014, 14:02   #24655  |  Link
6233638
Registered User
 
Join Date: Apr 2009
Posts: 1,019
Quote:
Originally Posted by Audionut View Post
Isn't dithering the final step before rendering? In that case, shouldn't it (luminance) map to whatever the device is calibrated too? Based on the settings in the calibration tab.
I agree, it should follow the calibration tab. (and 3DLUTs will need a preference to set what their display is calibrated to)

Quote:
Originally Posted by Audionut View Post
Otherwise, wouldn't it be best to map back to the source profile?
If I recall correctly, the end-to-end gamma target for BT.709 content is intended to be around 1.2

1.2/0.51 = 2.35

Though newer standards based upon measurements of mastering-grade monitors (BT.1886) have moved this to 2.40 rather than 2.35

Displays sold today are far more likely to be around 2.2 gamma than 2.35 out of the box though, which is why I would suggest using that value instead if nothing is specified.
6233638 is offline   Reply With Quote
Old 9th March 2014, 14:06   #24656  |  Link
turbojet
Registered User
 
Join Date: May 2008
Posts: 1,840
Clean driver install had no effect. The 250 is connected via s-video to a 59Hz crt with latest drivers. A 650 (327 drivers) is connected to 24/48/50/60Hz LCD via hdmi. I'll go check to see if there is a difference when switching refresh rate on the LCD.
__________________
PC: FX-8320 GTS250 HTPC: G1610 GTX650
PotPlayer/MPC-BE LAVFilters MadVR-Bicubic75AR/Lanczos4AR/Lanczos4AR LumaSharpen -Strength0.9-Pattern3-Clamp0.1-OffsetBias2.0
turbojet is offline   Reply With Quote
Old 9th March 2014, 14:10   #24657  |  Link
iSunrise
Registered User
 
Join Date: Dec 2008
Posts: 496
Quote:
Originally Posted by 6233638 View Post
What are you calibrating your screen to? BT.709 content should be viewed using the BT.1886 transfer function (which becomes a flat 2.40 gamma on high contrast displays) and never the BT.709 transfer function, which is intended for capture, not playback.
Depending on the content I view, I have several choices to go with. I know that Rec.709 is only intended for capturing/monitoring, but BT.709 still is the closest here. BT.1886 didn´t work out to well for me last time I tried. Then again, I´m not using DispCal or the ArgylCMS, I´m using the Eizo Colornavigator for this.
iSunrise is offline   Reply With Quote
Old 9th March 2014, 14:12   #24658  |  Link
Audionut
Registered User
 
Join Date: Nov 2003
Posts: 1,281
Quote:
Originally Posted by 6233638 View Post
(and 3DLUTs will need a preference to set what their display is calibrated to)
What does madVR currently do with a 3dlut selected? Unless I am missing something, there is no option to set a gamma in that case.

If the 3dlut doesn't contain that information, then I agree.

Quote:
Originally Posted by iSunrise View Post
BT.1886 didn´t work out to well for me last time I tried. Then again, I´m not using DispCal or the ArgylCMS, I´m using the Eizo Colornavigator for this.
I found BT.1886 to be extremely useful on my display. It allowed me to effectively crush the blacks with display controls (making black actually look like black) (nothing retarded like setting the black control to 0, but crushed non the less), and the calibration lifts the shadow details on playback.

In essence, I now see excellent shadow detail, with actual blacks.

Previously, I would either have crushed blacks, or grey shadows. It was impossible (for me), to get the display to show a pure black with shadow detail.
__________________
http://www.7-zip.org/

Last edited by Audionut; 9th March 2014 at 14:24.
Audionut is offline   Reply With Quote
Old 9th March 2014, 14:22   #24659  |  Link
iSunrise
Registered User
 
Join Date: Dec 2008
Posts: 496
Quote:
Originally Posted by leeperry View Post
PotP won't let me select the image source filter AFAIK, it'll always be "Image source". It would be most amazing if you could nail down the name of the still images source filters used in PotP & MPC and provide a picture-viewer mode in mVR that would kill all black frames(that keep coming and going more or less randomly, making it very hard to use) and allow to setup a custom profile with 256 neurons all the way
PotP just needs to support a way to select LAV as the source filter for tiff, png, jpeg, tga and bmp images (the PotP author only really needs to add these, because it already has checkboxes for several file formats that LAV supports), since nevcairiel already added that in 2012. In that case, madVR would take care of the rest.

Last edited by iSunrise; 9th March 2014 at 14:24.
iSunrise is offline   Reply With Quote
Old 9th March 2014, 14:49   #24660  |  Link
leeperry
Kid for Today
 
Join Date: Aug 2004
Posts: 3,477
Quote:
Originally Posted by madshi View Post
The trick is to do a 1D gamma ramp calibration first via ArgyllCMS, but then to "merge" the final 1D gamma ramps into the 3dlut.
And the 3DLUT+ArgyllCMS headaches will begin, otherwise you're stuck with a banding .cal file applied in 8bit by the GPU drivers after mVR's dithering....that's why I specifically wanted a display with proper calibration options(and applied in 10/12bit at that) so this would all be like waking up from a bad dream.

Quote:
Originally Posted by iSunrise View Post
PotP just needs to support a way to select LAV as the source filter for tiff, png, jpeg, tga and bmp images (the PotP author only really needs to add these, because it already has checkboxes for several file formats that LAV supports), since nevcairiel already added that in 2012. In that case, madVR would take care of the rest.
And what would that change? Can mVR already run a picture-viewer mode when using LAV? Anyway, feature requests are closed so I'll keep playing hide & seek with the random black frames if I find the patience.

Quote:
Originally Posted by cyberbeing View Post
None of the images you posted thus far are all that close to overall brightness and gamma response of the original when viewed on my display.

Gamma is brighter than the original. BT.709 2.2 is darker than original. The rest are much, much darker.
Exactly what I'm seeing as well.

Quote:
Originally Posted by cyberbeing View Post
my display is calibrated to a BT.709 curve scaled to ambient light. Essentially it matches BT.709 near-black, and matches L* across the rest of the luminance range.
Personally, I chose the most acceptable gamma preset in my TV(that goes from -3 to +3, I picked -2 which is pretty dark..around 2.4), then I maxed out the contrast(not clipping in HCFR, everything's smooth there) and I set the brightness setting to my taste. I'm spot-on D65, blacks are slightly crushed but PQ is stunning and my CRT's always came with crushed blacks, I'm not all that interested in what happens in the 3 first levels(out of 256) tbh.

Things might be different if there were proper gamma settings/10 points white balance in my F5000 Sammy TV but only their "6" serie does that and anything higher that their "5" serie can't do BFI without FI duh...and I think I like slightly crushed blacks anyway. Definitely not going back to Argyll, especially after whining to Graeme for the Nth time about BT1886 looking funky and him sending me yet another possible bug fix....20 mins of flashing colors to end up with a crappy gamma curve are beyond my patience.

Last edited by leeperry; 10th March 2014 at 00:02.
leeperry is offline   Reply With Quote
Reply

Tags
direct compute, dithering, error diffusion, madvr, ngu, nnedi3, quality, renderer, scaling, uhd upscaling, upsampling

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +1. The time now is 21:57.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.