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 25th January 2014, 13:10   #21761  |  Link
MSL_DK
Registered User
 
Join Date: Nov 2011
Location: Denmark
Posts: 137
Is there any way to check an mkv for faults resulting in dropped frames and presentation glitches?

Thanks for the new release!
MSL_DK is offline   Reply With Quote
Old 25th January 2014, 13:14   #21762  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
0.87g has essentially resolved the CPU load issue. It's now close enough that the remaining difference can be chalked up to new features added.

madVR 0.86.11 = 1.52% +/- 0.02% CPU load

madVR 0.87g = 1.63% +/- 0.02% CPU Load

The special 'cyber' build seems slightly worse with 1.67% +/- 0.02% CPU load.
cyberbeing is offline   Reply With Quote
Old 25th January 2014, 13:29   #21763  |  Link
noee
Registered User
 
Join Date: Jan 2007
Posts: 530
Quote:
Originally Posted by madshi
Ok. Can you please try 87g with debanding turned *off*? Does that fix the uploading problem for you? If so, does turning debanding on - and choosing a higher setting for fades bring the problem back?
Quite impressive results with .87g. Yes, it appears to be back to .86 perf levels. In fact, maybe even better, queues are all maxed full.

So, I decided to enable debanding and Error Diffusion, and it's working great with those options too. Maxed out queues, no drops, excellent perf. We'll see what happens with NNEDI, but that might be the line for this GPU (HD6570).
noee is offline   Reply With Quote
Old 25th January 2014, 13:30   #21764  |  Link
PixelH8
Registered User
 
Join Date: Dec 2012
Posts: 5
Oh goodie. I just tried the latest madVR87g build and the swapped colors are definitely gone. But the not-so-little red box that sometimes flashes in the upper left corner is supposed to be there right? To indicate that this is a test build?

I also have a third system with a GTS 250 that I'm in the middle of repairing. As soon as I'm done I'll let you know how the new build works on that system as well.
PixelH8 is offline   Reply With Quote
Old 25th January 2014, 13:34   #21765  |  Link
Siso
Soul Seeker
 
Siso's Avatar
 
Join Date: Sep 2013
Posts: 711
Quote:
Originally Posted by madshi View Post
Now this is weird. Why is it working for you - and for me, too. But not for everyone else? I'm quite confused. BTW, I've already fixed the swapped color channels in the latest test build...
I can confirm the swapped color channels too.
GTX 550 ti, 326.19

MPC-BE latest stable version, lav filters 0.60.1, xy-vsfilter 3.0.0.236

I'm going to try 0.87g

Last edited by Siso; 25th January 2014 at 13:36.
Siso is offline   Reply With Quote
Old 25th January 2014, 13:34   #21766  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
Ooopsi-daisy, forgot to reenable optimization in the MSVC++ compiler settings. Does this one bring 0.87g back to 1.52%, cyberbeing? Please also try enabling debanding with different settings for fade in/out detection, to see how that impacts CPU performance - because in that case madVR will fall back to my SSE2 uploading code, instead of using memcpy. Would be interesting to see how it compares...

http://madshi.net/madVRoptimized.rar (87g, only release mode madVR.ax)
madshi is offline   Reply With Quote
Old 25th January 2014, 13:36   #21767  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
Quote:
Originally Posted by PixelH8 View Post
Oh goodie. I just tried the latest madVR87g build and the swapped colors are definitely gone. But the not-so-little red box that sometimes flashes in the upper left corner is supposed to be there right? To indicate that this is a test build?
Ooops, sorry. That's a debug rectangle which indicates that madVR detected a fade in/out. I sometimes enable that so I can see whether fade in/out detection works or not. I'll disable it again for the next build.

Quote:
Originally Posted by noee View Post
Quite impressive results with .87g. Yes, it appears to be back to .86 perf levels. In fact, maybe even better, queues are all maxed full.

So, I decided to enable debanding and Error Diffusion, and it's working great with those options too. Maxed out queues, no drops, excellent perf. We'll see what happens with NNEDI, but that might be the line for this GPU (HD6570).
Good to hear! I did do some changes which I hoped would improve performance slightly.

Last edited by madshi; 25th January 2014 at 13:38.
madshi is offline   Reply With Quote
Old 25th January 2014, 13:42   #21768  |  Link
SecurityBunny
Registered User
 
Join Date: Jul 2013
Posts: 76
Quote:
Originally Posted by madshi View Post

Can you please retest with 87g?

Random dithering with smooth motion no longer flashes the screen. OpenCL error diffusion (unchecked use random dithering) still causes a black screen unfortunately. I'll take your advice and stay away from OpenCL for the time being. Thanks for the smooth motion fix.
SecurityBunny is offline   Reply With Quote
Old 25th January 2014, 13:49   #21769  |  Link
pie1394
Registered User
 
Join Date: May 2009
Posts: 212
Quote:
Originally Posted by madshi View Post
P10 and SmoothMotion making problems: Does this still occur with 87g? Even with debanding turned off?
This version is actually worse on the ION-LE platform. None of these two issues is fixed. Instead it introduces a new issue:

- Video screen becomes totally black when the 1st subtitle arrived from XySubFilter -- if SmoothMotion logic is NOT engaged. It happens to any 24-frames or 60-fields contents regardless of resolution. Debanding option ON/OFF does not matter, either.
pie1394 is offline   Reply With Quote
Old 25th January 2014, 13:52   #21770  |  Link
Soukyuu
Registered User
 
Soukyuu's Avatar
 
Join Date: Apr 2012
Posts: 169
Quote:
Originally Posted by madshi View Post
Now this is weird. Why is it working for you - and for me, too. But not for everyone else? I'm quite confused.
I smell driver issue. Everyone with failing openCL features, including me, are on one of the newest driver versions. Maybe nVidia broke something in the update?
Soukyuu is offline   Reply With Quote
Old 25th January 2014, 13:53   #21771  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
Quote:
Originally Posted by madshi View Post
Ooopsi-daisy, forgot to reenable optimization in the MSVC++ compiler settings. Does this one bring 0.87g back to 1.52%, cyberbeing?
That build is even better, ~1.26% CPU load.


Quote:
Please also try enabling debanding with different settings for fade in/out detection, to see how that impacts CPU performance - because in that case madVR will fall back to my SSE2 uploading code, instead of using memcpy. Would be interesting to see how it compares...
I'll look into this.
cyberbeing is offline   Reply With Quote
Old 25th January 2014, 13:55   #21772  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
Quote:
Originally Posted by pie1394 View Post
This version is actually worse on the ION-LE platform. None of these two issues is fixed. Instead it introduces a new issue:

- Video screen becomes totally black when the 1st subtitle arrived from XySubFilter -- if SmoothMotion logic is NOT engaged. It happens to any 24-frames or 60-fields contents regardless of resolution. Debanding option ON/OFF does not matter, either.
Hmmmm... I've tried all of this on my NVidia 9400 GPU which had problems with black rectangles when using XySubFilter in the past. But no problems here, whatsoever. How can I reproduce these issues? Maybe you can provide me with small samples to test with, in case it has something to do with the samples somehow?

Quote:
Originally Posted by cyberbeing View Post
That build is even better, ~1.26% CPU load.
Cool!

Quote:
Originally Posted by Soukyuu View Post
I smell driver issue. Everyone with failing openCL features, including me, are on one of the newest driver versions. Maybe nVidia broke something in the update?
That is quite possible. Anyone willing to downgrade to the driver versions used by PixelH8 to check if that fixes the OpenCL issues?
madshi is offline   Reply With Quote
Old 25th January 2014, 13:57   #21773  |  Link
James Freeman
Registered User
 
Join Date: Sep 2013
Posts: 919
87g with Debanding, 4K 60fps video (downscaled to HD with Catmull-Rom) works like charm without frame skipping (although buffers not fully full).
Without Debanding the buffers completely full with this video.

"1 frame drop every t" & "clock deviation %" works better than 86.11, its a lot more stable.
Performance up to par with 86.11 or even better, Perfect.
__________________
System: i7 3770K, GTX660, Win7 64bit, Panasonic ST60, Dell U2410.

Last edited by James Freeman; 25th January 2014 at 14:23.
James Freeman is offline   Reply With Quote
Old 25th January 2014, 14:02   #21774  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,140
madVR v0.87.1 released

http://madshi.net/madVR.zip

Code:
* fixed: CPU consumption was unnecessarily high
* fixed: auto profile switching didn't fully work
* fixed: uploading was very slow, resulting in performance problems
* fixed: chroma channels were swapped when using error diffusion with NVidia
* fixed: some OpenCL crashes when using OpenCL 1.2 DLL with NVidia
* fixed: smooth motion FRC produced black image in some situations
* fixed: video mode deinterlacing was completely broken
There may still be some problems left. But there's definite improvement over v0.87.0, so I've decided to release v0.87.1 now.

Current theory is that older NVidia drivers might make OpenCL work with madVR, and that newer drivers might have broken something. Drivers 306.97 are reported to be working with GTX 560 Ti. And drivers 310.90 are reported to be working with GTX 460. This needs to be double checked.
madshi is offline   Reply With Quote
Old 25th January 2014, 14:07   #21775  |  Link
burfadel
Registered User
 
Join Date: Aug 2006
Posts: 2,229
I tried the MadVR optimised one, and it didn't even load! (MPC-HC used a different renderer). I then went back and realised the previous version I had was 0.87E or 0.87F, so I copied over the files from 0.87G, then used the MadVR optimised version and it works fine .
burfadel is offline   Reply With Quote
Old 25th January 2014, 14:13   #21776  |  Link
antonyfrn
Registered User
 
Join Date: Nov 2012
Posts: 17
I just installed 87G and tried with MadVR Optimised and both give me the below message if I go from full screen to windowed mode. (I used the restore defaults bat to clear out any settings)

resetting Direct3D device failed (80070005)
creating Direct3D device failed (80070005)
__________________
Windows 10 Pro 64Bit
i7 6700k + H110i GT @4.6Ghz
16GB Corsar Vengance DDR4
EVGA GTX 970 FTW ACX 2.0
ASUS Maximus VIII Hero
ASUS Xonar D2X
antonyfrn is offline   Reply With Quote
Old 25th January 2014, 14:14   #21777  |  Link
DarkSpace
Registered User
 
Join Date: Oct 2011
Posts: 204
Quote:
Originally Posted by madshi View Post
Lanczos3AR in only one dimension (Y, IIRC) to shift the image 0.5 pixels.
Yes, should be y dimension (x needs the shift, after all). Thanks for answering!
Now for the (I think) last curious question: Even though it's probably insignificant, do you have any plans to let the user choose the shifting algorithm?

Last edited by DarkSpace; 10th May 2014 at 00:12. Reason: closed my brackets
DarkSpace is offline   Reply With Quote
Old 25th January 2014, 14:21   #21778  |  Link
mark0077
Registered User
 
Join Date: Apr 2008
Posts: 1,106
Quote:
Originally Posted by madshi View Post
madVR v0.87.1 released

http://madshi.net/madVR.zip

Code:
* fixed: CPU consumption was unnecessarily high
* fixed: auto profile switching didn't fully work
* fixed: uploading was very slow, resulting in performance problems
* fixed: chroma channels were swapped when using error diffusion with NVidia
* fixed: some OpenCL crashes when using OpenCL 1.2 DLL with NVidia
* fixed: smooth motion FRC produced black image in some situations
* fixed: video mode deinterlacing was completely broken
There may still be some problems left. But there's definite improvement over v0.87.0, so I've decided to release v0.87.1 now.

Current theory is that older NVidia drivers might make OpenCL work with madVR, and that newer drivers might have broken something. Drivers 306.97 are reported to be working with GTX 560 Ti. And drivers 310.90 are reported to be working with GTX 460. This needs to be double checked.
Cheers for the new version. All working fine except for OpenCL as reported. My driver version is 332.21 WHQL with GTX295 incase anyone else has older driver versions with the same card and it works. Do you have a contact in Nvidia or do you need any of us to post on their forums regarding the OpenCL issues?
mark0077 is offline   Reply With Quote
Old 25th January 2014, 14:25   #21779  |  Link
antonyfrn
Registered User
 
Join Date: Nov 2012
Posts: 17
@madshi just some info for you posted by an NVidia Rep ManuelG maybe something improvements driver side or a headache I think they are off the new Quadros 334.67

Quote:
For those who don't like to mod their drivers or risk corrupting their driver installation, we will be releasing an update Geforce driver soon.
creating Direct3D device failed (80070005) message with 0.87.1 you just posted
__________________
Windows 10 Pro 64Bit
i7 6700k + H110i GT @4.6Ghz
16GB Corsar Vengance DDR4
EVGA GTX 970 FTW ACX 2.0
ASUS Maximus VIII Hero
ASUS Xonar D2X

Last edited by antonyfrn; 25th January 2014 at 14:28.
antonyfrn is offline   Reply With Quote
Old 25th January 2014, 14:27   #21780  |  Link
Soukyuu
Registered User
 
Soukyuu's Avatar
 
Join Date: Apr 2012
Posts: 169
Quote:
Originally Posted by madshi View Post
Anyone willing to downgrade to the driver versions used by PixelH8 to check if that fixes the OpenCL issues?
331.81 -> 310.90:
- openCL dither works
- NNEDI doubling works (slideshow)
- luma NNEDI upscaling gives a green tint to the whole image for 10bit h264, 8bit is fine.

I am going to update versions until I hit the blackscreen again. This is with 87.1 btw.
__________________
AMD Phenom II X4 970BE | 12GB DDR3 | nVidia 260GTX | Arch Linux / Windows 10 x64 Pro (w/ calling home shut up)

Last edited by Soukyuu; 25th January 2014 at 14:30.
Soukyuu 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 22:17.


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