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 18th June 2014, 11:46   #26681  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
NVIDIA users should beware that the 340.43 Beta seems to have a bug which causes Custom Resolutions created by previous drivers to be ignored. They must have accidentally or otherwise changed the binary format of the CustomDisplay registry key slightly, making this new driver think the old key is corrupt. You can create them from scratch again, but rather annoying if you had a lot of special custom timings. YMMV.
cyberbeing is offline   Reply With Quote
Old 18th June 2014, 12:38   #26682  |  Link
StinDaWg
Registered User
 
Join Date: Jan 2014
Posts: 216
Quote:
Originally Posted by leeperry View Post
I think StinDaWg & Frechdachs nailed it coz what I don't like with NNEDI is the aliasing they're reporting with AR+LL, will need to reevaluate.
I'm still waiting to hear what madshi has to say about it, just going to use without AR for now.

Quote:
Originally Posted by QBhd View Post
Well if you look at the madVR settings, AR does increase the aliasing bar...
Ya, but it's a HUGE difference, and only happens with downscaling. No AR text looks 99% perfect, AR on text looks really jagged (worse than just Jinc3AR upscaling).

The first image in each is Lanczos4, the 2nd is Lanczos4 AR. Look at letters like M,N,V,A.


http://screenshotcomparison.com/comp...9590/picture:0

Last edited by StinDaWg; 18th June 2014 at 14:41.
StinDaWg is offline   Reply With Quote
Old 18th June 2014, 18:14   #26683  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
StinDaWg, do you see the same when using Spline 3-tap AR upscaling/downscaling, NNEDI3 256 chroma upscaling, NNEDI3 256 luma doubling?

Testing with 256 neurons is to rule out AR enhancing an NNEDI3 artifact. NNEDI3 has a tendency to produce blocky artifacts on high contrast text edges in particular. Only 256 neurons is completely immune.

Testing with Spline 3-tap is again to rule out potential anomalies. madVR's AR algorithm used with all resizers except Jinc was originally tuned to be near optimal (>95% eliminated ringing, without loss of detail loss or sharpness) only with Spline 3-tap. AR has always produced residual ringing artifacts with Lanczos3/4/8, Bicubic75/100, Spline4, and Jinc3/4 which exponentially become more visible with higher taps. The low ringing and soft algorithms on the other hand experience a slight loss of detail and sharpness with AR.


I'd also be curious if you can you reproduce this with madVR v0.87.8, which is prior to madVR correcting NNEDI3's 0.5 pixel shift when performing additional scaling. When pixel peeping I noticed that the text in your AR screenshots appears shifted slightly left & up compared to the non-AR screenshots, which is a bit strange.

Last edited by cyberbeing; 18th June 2014 at 20:52.
cyberbeing is offline   Reply With Quote
Old 18th June 2014, 19:59   #26684  |  Link
seiyafan
Registered User
 
Join Date: Feb 2014
Posts: 161
So is it still recommended to use Catmull-Rom AR LL for downscaling with NNEDI3 or should it be Spine 3?
seiyafan is offline   Reply With Quote
Old 18th June 2014, 23:29   #26685  |  Link
6233638
Registered User
 
Join Date: Apr 2009
Posts: 1,019
FYI I have responded to the discussion of anti-ringing/linear light scaling/NNEDI3, but a moderator has split it off into its own topic here: http://forum.doom9.org/showthread.php?p=1684179

Last edited by Guest; 19th June 2014 at 02:49. Reason: 3
6233638 is offline   Reply With Quote
Old 19th June 2014, 03:01   #26686  |  Link
StinDaWg
Registered User
 
Join Date: Jan 2014
Posts: 216
Quote:
Originally Posted by cyberbeing View Post
StinDaWg, do you see the same when using Spline 3-tap AR upscaling/downscaling, NNEDI3 256 chroma upscaling, NNEDI3 256 luma doubling?

Testing with 256 neurons is to rule out AR enhancing an NNEDI3 artifact. NNEDI3 has a tendency to produce blocky artifacts on high contrast text edges in particular. Only 256 neurons is completely immune.

Testing with Spline 3-tap is again to rule out potential anomalies. madVR's AR algorithm used with all resizers except Jinc was originally tuned to be near optimal (>95% eliminated ringing, without loss of detail loss or sharpness) only with Spline 3-tap. AR has always produced residual ringing artifacts with Lanczos3/4/8, Bicubic75/100, Spline4, and Jinc3/4 which exponentially become more visible with higher taps. The low ringing and soft algorithms on the other hand experience a slight loss of detail and sharpness with AR.


I'd also be curious if you can you reproduce this with madVR v0.87.8, which is prior to madVR correcting NNEDI3's 0.5 pixel shift when performing additional scaling. When pixel peeping I noticed that the text in your AR screenshots appears shifted slightly left & up compared to the non-AR screenshots, which is a bit strange.
I remember first seeing this issue when NNEDI3 was enabled, so I don't think madVR versions has anything to do with it. I tried NNEDI3 256 neurons/Spline 3AR and it isn't any better, still aliased text. The only thing that removes aliasing on all downscaling algorithms is to not use AR.

Quote:
Originally Posted by seiyafan View Post
So is it still recommended to use Catmull-Rom AR LL for downscaling with NNEDI3 or should it be Spine 3?
I've always wondered why it was recommended here to use Catmull-Rom for downscaling. In various encoding forums where people are obsessed with quality, everyone uses Spline 36 (I guess this is Spline 3 in madVR). Spline is almost as sharp as Lanczos with less ringing/artifacts. Catmull-Rom is too soft IMO. Spline is somewhere in the middle.

If the aliasing is inherent to the algorithm and can't be fixed, I don't think AR should be recommended to use anymore for downscaling (it's still fine for upscaling). One of the greatest benefits from NNEDI3 is near perfect diagonal lines, using AR negates that. I don't see any increased ringing from not using it anyways, at least for 720p->1080p.

Edit: Just took a look at the pictures posted in that linked post and the issue definitely seems to have something to do with the way NNEDI3 doubles. Don't know why it was broken off into another thread either as it's relevant to this discussion we're having.

Last edited by StinDaWg; 19th June 2014 at 03:47.
StinDaWg is offline   Reply With Quote
Old 19th June 2014, 05:27   #26687  |  Link
Guest
Guest
 
Join Date: Jan 2002
Posts: 21,923
The moderating strategy was discussed with and approved by madshi, who is the originator of this thread. Please trust us in our attempts to keep things organized and accessible. It is not just a question of relevance. We have 26000+ posts in this thread! It is a nightmare to find things and very hard for madshi. Please continue the topic of linear light scaling/NNEDI3 in the linked thread. Thank you for your understanding.

Just to remind about the policy: This thread is for support of madVR (bug reports, questions about usage, etc.) and occasional development things when madshi asks for them (help with testing, etc). Everything else including theoretical discussions should go into a separate thread. You can preface the thread with "madVR" so it is recognizable as relevant to madVR.

Last edited by Guest; 19th June 2014 at 06:08.
Guest is offline   Reply With Quote
Old 19th June 2014, 13:02   #26688  |  Link
leeperry
Kid for Today
 
Join Date: Aug 2004
Posts: 3,463
Quote:
Originally Posted by madshi View Post
Yes, please.
So as promised they asked AMD about the copyback nonsense, they replied that their engineering team was aware of the problem but didn't have the time to look at it.

Last edited by leeperry; 19th June 2014 at 13:07.
leeperry is offline   Reply With Quote
Old 20th June 2014, 06:20   #26689  |  Link
truexfan81
Registered User
 
truexfan81's Avatar
 
Join Date: Nov 2012
Posts: 138
Quote:
Originally Posted by leeperry View Post
So as promised they asked AMD about the copyback nonsense, they replied that their engineering team was aware of the problem but didn't have the time to look at it.
Wow now that is excellent customer service right there.
truexfan81 is offline   Reply With Quote
Old 20th June 2014, 06:31   #26690  |  Link
Mangix
Audiophile
 
Join Date: Oct 2006
Posts: 354
Quote:
Originally Posted by cyberbeing View Post
NVIDIA users should beware that the 340.43 Beta seems to have a bug which causes Custom Resolutions created by previous drivers to be ignored. They must have accidentally or otherwise changed the binary format of the CustomDisplay registry key slightly, making this new driver think the old key is corrupt. You can create them from scratch again, but rather annoying if you had a lot of special custom timings. YMMV.
Can't you just do EDID overrides instead? They're limited to four resolutions but they should work. Well, pixel clock accuracy might be an issue.
Mangix is offline   Reply With Quote
Old 20th June 2014, 09:35   #26691  |  Link
Plutotype
Registered User
 
Join Date: Apr 2010
Posts: 234
FSE blackscreen

Hi folks,
Win7 x64 ( Aero disabled ), madvr 0.87.10, MPC-HC nightly, Reclock, AMD HD7750 ( 14.4 WHQL ), Samsung UE65H8500 ( UHD TV ) connected directly per HDMI. 23.976fps movie played back on 3840x2160p24 resolution.
Playback in windowed mode works fine ( fullscreen as well ), but switching on fullscreen exclusive mode, when going fullscreen causes black screen and the playback stops there.
http://www.mediafire.com/download/59...adVR_-_log.rar

Any hint please what blackscreens the FSE mode?
Thanks
__________________
__________________
System: Intel Core i5-6500, 8GB RAM, GTX960/GTX1060, 55" Sony Bravia KDL-55HX850 @ 1920x1080 24p, Yamaha YSR-1100 surround bar
Setup: Win7/Win10 Pro, madvr/JRiver 24
Plutotype is offline   Reply With Quote
Old 20th June 2014, 14:58   #26692  |  Link
YxP
Registered User
 
Join Date: Oct 2012
Posts: 97
Quote:
Originally Posted by Plutotype View Post
Hi folks,
Win7 x64 ( Aero disabled ), madvr 0.87.10, MPC-HC nightly, Reclock, AMD HD7750 ( 14.4 WHQL ), Samsung UE65H8500 ( UHD TV ) connected directly per HDMI. 23.976fps movie played back on 3840x2160p24 resolution.
Playback in windowed mode works fine ( fullscreen as well ), but switching on fullscreen exclusive mode, when going fullscreen causes black screen and the playback stops there.
http://www.mediafire.com/download/59...adVR_-_log.rar

Any hint please what blackscreens the FSE mode?
Thanks
Try without Reclock. Hasn't happened to me since I got rid of it.
YxP is offline   Reply With Quote
Old 20th June 2014, 16:10   #26693  |  Link
Meulen92
Registered User
 
Join Date: May 2014
Posts: 9
Quote:
Originally Posted by Plutotype View Post
Hi folks,
Win7 x64 ( Aero disabled ), madvr 0.87.10, MPC-HC nightly, Reclock, AMD HD7750 ( 14.4 WHQL ), Samsung UE65H8500 ( UHD TV ) connected directly per HDMI. 23.976fps movie played back on 3840x2160p24 resolution.
Playback in windowed mode works fine ( fullscreen as well ), but switching on fullscreen exclusive mode, when going fullscreen causes black screen and the playback stops there.
http://www.mediafire.com/download/59...adVR_-_log.rar

Any hint please what blackscreens the FSE mode?
Thanks
Sometimes happened to me when i had 'alternative interop hack' enabled in the image doubling settings.
Meulen92 is offline   Reply With Quote
Old 21st June 2014, 05:08   #26694  |  Link
baii
Registered User
 
Join Date: Dec 2011
Posts: 180
What is the cheapest downscale algorithm to use for device like windows tablet?

I try all of them and seem DXVA is the cheapest, am i right?

The only thing concern here is color management, other than that, don't mind turning down everything else/use any option to lower the load.
baii is offline   Reply With Quote
Old 21st June 2014, 11:52   #26695  |  Link
6233638
Registered User
 
Join Date: Apr 2009
Posts: 1,019
Quote:
Originally Posted by baii View Post
What is the cheapest downscale algorithm to use for device like windows tablet?

I try all of them and seem DXVA is the cheapest, am i right?

The only thing concern here is color management, other than that, don't mind turning down everything else/use any option to lower the load.
It depends on the hardware. That's possibly true for Intel hardware, but probably not with Nvidia.

Check the render stats to see. (hit CTRL+J when a video is playing, shorter times are better)
6233638 is offline   Reply With Quote
Old 21st June 2014, 13:56   #26696  |  Link
somy
Registered User
 
Join Date: Dec 2009
Posts: 177
Hi, I'm considering NUC for my next HTPC, does the i3 CPU with Intel® HD graphics 4400 good enough for MadVR (not extreme settings) with 1080P content?
Thank you in advance!
somy is offline   Reply With Quote
Old 21st June 2014, 22:18   #26697  |  Link
ryrynz
Registered User
 
ryrynz's Avatar
 
Join Date: Mar 2009
Posts: 3,230
Quote:
Originally Posted by somy View Post
Hi, I'm considering NUC for my next HTPC, does the i3 CPU with Intel® HD graphics 4400 good enough for MadVR (not extreme settings) with 1080P content?
Thank you in advance!
Should be fine providing you're displaying on a ~1080 screen.
ryrynz is offline   Reply With Quote
Old 21st June 2014, 23:03   #26698  |  Link
tjcinnamon
Registered User
 
Join Date: Nov 2011
Posts: 44
Quote:
Originally Posted by cyberbeing View Post
StinDaWg, do you see the same when using Spline 3-tap AR upscaling/downscaling, NNEDI3 256 chroma upscaling, NNEDI3 256 luma doubling?

Testing with 256 neurons is to rule out AR enhancing an NNEDI3 artifact. NNEDI3 has a tendency to produce blocky artifacts on high contrast text edges in particular. Only 256 neurons is completely immune.

Testing with Spline 3-tap is again to rule out potential anomalies. madVR's AR algorithm used with all resizers except Jinc was originally tuned to be near optimal (>95% eliminated ringing, without loss of detail loss or sharpness) only with Spline 3-tap. AR has always produced residual ringing artifacts with Lanczos3/4/8, Bicubic75/100, Spline4, and Jinc3/4 which exponentially become more visible with higher taps. The low ringing and soft algorithms on the other hand experience a slight loss of detail and sharpness with AR.


I'd also be curious if you can you reproduce this with madVR v0.87.8, which is prior to madVR correcting NNEDI3's 0.5 pixel shift when performing additional scaling. When pixel peeping I noticed that the text in your AR screenshots appears shifted slightly left & up compared to the non-AR screenshots, which is a bit strange.
What kind of beastly machine do you have where you can run 256 neurons (with no dropped frames)?!?!?
tjcinnamon is offline   Reply With Quote
Old 21st June 2014, 23:37   #26699  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
With a higher-end GPU, 256 neurons on SD content should be within reason.

Though the intention of that post was only to take screenshots of the end result to rule out NNEDI3 artifacts, not actually watching the video in motion.

Last edited by cyberbeing; 21st June 2014 at 23:56.
cyberbeing is offline   Reply With Quote
Old 22nd June 2014, 05:35   #26700  |  Link
tjcinnamon
Registered User
 
Join Date: Nov 2011
Posts: 44
Quote:
Originally Posted by cyberbeing View Post
With a higher-end GPU, 256 neurons on SD content should be within reason.

Though the intention of that post was only to take screenshots of the end result to rule out NNEDI3 artifacts, not actually watching the video in motion.
I suppose I could run it too so long as it was screenshots.

with 1080p content do I even need to use LUMA image doubling NNEDI3 if my monitor resolution is max 1080p?
tjcinnamon 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 15:39.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.