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 14th May 2015, 23:10   #29981  |  Link
MS-DOS
Registered User
 
Join Date: Sep 2012
Posts: 76
FSE mode works fine here in both 8 and 10 bit modes. 2 hours of playback without a single drop\glitch, "present a frame for every Vsync" unchecked.
All queues are full in windowed mode, while present queue is at 6-7\8 in FSE.
MS-DOS is offline   Reply With Quote
Old 14th May 2015, 23:16   #29982  |  Link
huhn
Registered User
 
Join Date: Oct 2012
Posts: 6,149
Quote:
Originally Posted by MysteryX View Post
Something is confusing.

SuperRes can be applied either in Upscaling Refinement or in Chroma Upscaling. What's the difference?
the superres from Upscaling Refinement isn't used on chroma upsacaling.
don't forget chroma upscaling does only scale to RGB and nothing else

Quote:
FineSharp and LumaSharpen can either be applied in Image Enhancements or Upscaling Refinement. What's the difference?
the "Image Enhancements" is applied before scaling AFAIK.
huhn is offline   Reply With Quote
Old 14th May 2015, 23:29   #29983  |  Link
madshi
Registered Developer
 
Join Date: Sep 2006
Posts: 9,137
Quote:
Originally Posted by omarank View Post
I still have the same opinion as before: images look better with madVR’s old debanding algorithms irrespective of the debanding strength. Instead of replacing the old “high” preset with Shiandow’s algorithm, I believe it would be better to modify the old “high” preset algorithm such that it is able to remove an equivalent amount of banding as Shiandow’s algo. However, I have no issues if you even replace the “high” preset as Shiandow’s debanding has really improved after he made those modifications.
Thanks for the feedback! Seems there'll be another Shiandow deband iteration coming. <sigh> Of course it's all good, the algorithm getting better all the time. But I hope you users/testers won't be bored at some point, retesting revised algos all the time?

Quote:
Originally Posted by omarank View Post
By the way, you haven’t added that trade quality option for disabling Error Diffusion dithering for 10 bit output?
It's on my list.

Quote:
Originally Posted by detmek View Post
Madshi, render queue gets filled with this new version on my Intel IGPU in full screen windowed mode. Now I only have a second or two frame out-of-order during switching Full Screen - Windowed mode and vice versa. Same with switching Windowed - FSE. It maybe better for me to stay with D3D9 for now.
If you see no advantage for D3D11 than there's no problem at all staying with D3D9. I think I didn't answer to your previous comment: If "use a separate device for presentation" causes the same problem with D3D9 then that points to a driver issue. I've had several of those with Intel already. Many of them have to do with sharing textures over multiple devices, sadly.

Quote:
Originally Posted by zveroboy View Post
I tried both (D3D11 enabled & D3D11 disabled). Still not working.
For a moment, turned off desktop composition, and immediately turned back on.

0.88.7 - the same behavior.
Did you restart the media player after disabling D3D11? If so, did previous madVR versions not have this problem? If so, could you please check which exact madVR version introduced the problem for you? That would help me figuring out which change may have broken this for you. You can find the old madVR builds here:

http://www.videohelp.com/software/ma...sions#download

Quote:
Originally Posted by huhn View Post
ok it found the problem it only works with "refine the image after every ~x2 upscaling step".
"refine the image only once after upscaling" doesn't work.
Ah, I see. But that does make sense: If you refine the image only once after upscaling is complete, octupling won't bring any benefit over e.g. Jinc3 AR upscaling. Luma doubling brings a nice quality improvement. Quadrupling only a small improvement. Octupling will be invisible, unless you sharpen the image in between the doubling steps.

Quote:
Originally Posted by Dogway View Post
Can't you image double in YUV space? Currently there are 2 extra color conversions (RGB->YCbCr->RGB).

This is worrying:

Luma and Chroma doubling should use the same kernel otherwise mismatches will arise specially on aliased borders.
Artifacts from different algorithms will be rather low, because the chroma channel is usually very soft. But if you're worried about this kind of problems, you can use both luma+chroma doubling, then this problem should be solved.

Multiple colorspace conversions are not "nice", but I'm using 16bit integer textures which have a LOT of precision, and I'm not cutting away BTB/WTW, either, so there's really not much to worry about. The colorspace conversions in madVR are not 100% lossless, but they should be nearly lossless.

Quote:
Originally Posted by James Freeman View Post
I like Shiandow's debanding, it does an excellent job and it is very versatile with only 2 settings.
I love how the Threshold determines the strength and Detail can fine tune everything in between.

I find that both algorithms are doing what they're meant to do perfectly... so it is a tough choice.
A debanding filter is a "set it and forget it" for some people and some will want a more in depth control.
I think that Shiandow's debanding is great for both types of people, it can do exactly what the default debanding does with everything in between (and way more if you want) and only with 2 knobs to twist.

Some people like a setting above High (I read Ultra somewhere), and some will want lower than Low.
Personally I find the custom mode with the default debanding algo too complicated so I never used it, but with shiandow's algo a single Detail knob does just that.

Again, it's a though choice but I would go with the one that can be fine tuned easily, and comes out as a one size fits all.
There will be no fine tuning (by users). After all is said and done, there will be low/med/high, just as it is now. No other controls. Regardless of which algorithm will be chosen. So "configurability" is not something I'm interested in hearing your feedback about. What I want to hear about is quality differences, maybe quality vs performance considerations, nothing else. Thanks!

(Maybe, in some far away time, there might be an advanced options page with more controls, but this is not something that is of any decision making interest to me right now. I only care about the algo which provides the best quality (and performance) right now. Everything else simply DOESN'T MATTER.)

Quote:
Originally Posted by MysteryX View Post
I'm trying the new version. DirectX 11 presentation path isn't working on my computer. The audio plays fine but there is no video (black). I had the same issue when testing Media Player.NET + DirectX 11

As for image doubling, it still isn't working on my computer. Applying NNEDI3 frame quadrupling doesn't affect rendering times. The option "use alternative interop hack" isn't helping either. However, NEDI doubling works.

I have a Radeon HD 7670M with Windows 7 x64.
Do you have the Windows 7 Platform Update installed? If yes, this sounds like a driver issue to me. Don't know what else it could be.

Quote:
Originally Posted by huhn View Post
can you please test d3d11 10 bit FSE? you are on AMD right?
would be nice to know if my system is just on a rampage.
Works fine on two different Windows 7 PCs with two different GPUs for me. And works also fine on my Windows 8.1 x64 dev PC. All in both 8bit and 10bit.

Quote:
Originally Posted by James Freeman View Post
FSE mode fails to engage. Black Screen in 8bit. "exclusive mode failed" message in 10bit.
Nvidia Windows 7 64
Works fine here. Is that a new problem for you? If so, could you please try to find out which exact madVR build introduced this problem for you?

Quote:
Originally Posted by leeperry View Post
My bad, then what's confusing is that whenever luma NEDI is checked, ideally NEDI chroma should also be automatically ticked
Exactly that should already happen, as huhn already reported.

Quote:
Originally Posted by leeperry View Post
I also wish I could enable 2X for >1.01 and 4X for 2.01, I kinda find 1.2 & 2.4 too high and pretty arbitrary.
Isn't "always - if upscaling is needed" exactly the same as 1.01/2.01? I'm confused.

Quote:
Originally Posted by MysteryX View Post
Something is confusing.

SuperRes can be applied either in Upscaling Refinement or in Chroma Upscaling. What's the difference?

FineSharp and LumaSharpen can either be applied in Image Enhancements or Upscaling Refinement. What's the difference?
I've explained this in detail in the first couple of pages following the initial v0.88.0 release.

Quote:
Originally Posted by MS-DOS View Post
FSE mode works fine here in both 8 and 10 bit modes. 2 hours of playback without a single drop\glitch, "present a frame for every Vsync" unchecked.
All queues are full in windowed mode, while present queue is at 6-7\8 in FSE.
That's what I want to hear...
madshi is offline   Reply With Quote
Old 14th May 2015, 23:29   #29984  |  Link
aufkrawall
Registered User
 
Join Date: Dec 2011
Posts: 1,723
Quote:
Originally Posted by huhn View Post
can you please test d3d11 10 bit FSE? you are on AMD right?
would be nice to know if my system is just on a rampage.
I'm on NV (GTX 980, Win 10074, driver 350.12, x64).
Seems like 10 bit D3D11 FSE is working fine (tested in 10 bit mode with 8 and 10 bit video), but I have to enable frame for every sync to avoid massive presentation glitches. No tearing etc.
aufkrawall is offline   Reply With Quote
Old 14th May 2015, 23:33   #29985  |  Link
XMonarchY
Registered User
 
Join Date: Jan 2014
Posts: 489
Quote:
Originally Posted by madshi View Post
How do you know it uses the BT.709 3dlut? If that is really the case, it's a bug and then please enter it to the madVR bug tracker.
Its difficult to explain, but I do know it for a fact. When I made Rec.709 3DLUT I had my TV's setting for Flesh Tone set to 0, but I accidentally left Flash Tone set to -15 when I created Rec.601 3DLUT. That creates an obvious difference when I switch between Rec.709 and Rec.601 3DLUT's using identical TV settings. Rec.709 line is used for SMPTE C content for sure, at least in 88.5.
__________________
8700K @ 5Ghz | ASUS Z370 Hero X | Corsair 16GB @ 3200Mhz | RTX 2080 Ti @ 2100Mhz | Samsung 970 NVMe 250GB | WD Black 2TB | Corsair AX 850W | LG 32GK850G-B @ 165Hz | Xonar DGX | Windows 10 LTSC 1809
XMonarchY is offline   Reply With Quote
Old 14th May 2015, 23:37   #29986  |  Link
MysteryX
Soul Architect
 
MysteryX's Avatar
 
Join Date: Apr 2014
Posts: 2,173
Shouldn't "Upscaling Refinement" be within profile groups instead of being on its own for all profiles?

Last edited by MysteryX; 24th June 2015 at 07:12.
MysteryX is offline   Reply With Quote
Old 14th May 2015, 23:39   #29987  |  Link
aufkrawall
Registered User
 
Join Date: Dec 2011
Posts: 1,723
Quote:
Originally Posted by madshi View Post
Thanks for the feedback! Seems there'll be another Shiandow deband iteration coming. <sigh> Of course it's all good, the algorithm getting better all the time. But I hope you users/testers won't be bored at some point, retesting revised algos all the time?
It's not very time consuming, and if one wants better quality it's not really a sacrifice.

However, have we seen yet example images that show a real advantage for Shiandow's current implementation?
I for myself haven't found such an example yet. Higher threshould values with bad sources can be nice because of that deblocking effect, but it vanishes too much detail with good sources.
aufkrawall is offline   Reply With Quote
Old 14th May 2015, 23:40   #29988  |  Link
XMonarchY
Registered User
 
Join Date: Jan 2014
Posts: 489
88.7 works wonderful for me! I get no frame drops, even when I use Chroma Upscaling SuperRes Filter and also enable SuperRes in Image Refinement (High setting). Hell, I even enabled Octuple settings and everything is peachy! Shiandow's debanding and new setting (0.2 for threshold and 1 for detail level) are also better. I used to get obvious big shimmering squares (NOT dithering) on darker grays in Mid/Low-HQ content that were not present when only the original debanding was used. With 88.7 and new Shaindow's settings, those shimmering squares are now a lot smaller. Sometimes I think the original de-banding is still better, not as noisy.
__________________
8700K @ 5Ghz | ASUS Z370 Hero X | Corsair 16GB @ 3200Mhz | RTX 2080 Ti @ 2100Mhz | Samsung 970 NVMe 250GB | WD Black 2TB | Corsair AX 850W | LG 32GK850G-B @ 165Hz | Xonar DGX | Windows 10 LTSC 1809

Last edited by XMonarchY; 14th May 2015 at 23:45.
XMonarchY is offline   Reply With Quote
Old 14th May 2015, 23:46   #29989  |  Link
huhn
Registered User
 
Join Date: Oct 2012
Posts: 6,149
Quote:
Originally Posted by aufkrawall View Post
I'm on NV (GTX 980, Win 10074, driver 350.12, x64).
Seems like 10 bit D3D11 FSE is working fine (tested in 10 bit mode with 8 and 10 bit video), but I have to enable frame for every sync to avoid massive presentation glitches. No tearing etc.
could be limited to the AMD WDDM 2.0 driver.
thanks.
huhn is offline   Reply With Quote
Old 15th May 2015, 00:09   #29990  |  Link
tickled_pink
Registered User
 
Join Date: Dec 2011
Posts: 12
Stupid(?) question - When testing deband algos, we should check one or the other, correct? Leaving both checked is effectively using both algos simultaneously, or am I missing something?
tickled_pink is offline   Reply With Quote
Old 15th May 2015, 00:24   #29991  |  Link
cyberbeing
Broadband Junkie
 
Join Date: Oct 2005
Posts: 1,859
Quote:
Originally Posted by madshi View Post
madVR v0.88.7 released

* fixed: D3D11 render queue didn't fill in Windows 7
This version appears to have broken D3D11 entirely on my Win7 GTX770 system, or rather it never activates and madVR always uses D3D9. It works when I reset madVR to defaults though, so it must be something related to my settings/profiles, but so far I've not been able to figure out which causes it.

https://www.mediafire.com/?x1fgya1148ecsm4
cyberbeing is offline   Reply With Quote
Old 15th May 2015, 00:28   #29992  |  Link
x7007
Registered User
 
Join Date: Apr 2013
Posts: 253
I have couple of questions.

Currently by your eyes if I have 970 GTX I have Philips 7007 3D it's 16-235. Should I use native display bitrate 8bit or 10bit ? I'm using the D3D11.

What should I choose for debanding Debanding Strength - medium / fade in/out high ? or to enable shaindow's deband threshold 0.5 detail level 2 ? I don't remember if that's the default

Should I use quadruple luma resolution ?

How come without the Settings.bin file in the folder Madvr keep my settings ? and after that it makes the file, how it remembers all my settings without the settings.bin ?

Those settings bothering me cause I can't be sure what's the best.

Thanks

Last edited by x7007; 15th May 2015 at 01:49.
x7007 is offline   Reply With Quote
Old 15th May 2015, 00:37   #29993  |  Link
aufkrawall
Registered User
 
Join Date: Dec 2011
Posts: 1,723
Quote:
Originally Posted by huhn View Post
could be limited to the AMD WDDM 2.0 driver.
thanks.
Funny, I had the exactly same problem with the Nvidia WDDM 2.0 driver.
Edit: But this was not related to 10 bit, but to the D3D11 renderer in general.

Last edited by aufkrawall; 15th May 2015 at 00:39.
aufkrawall is offline   Reply With Quote
Old 15th May 2015, 00:38   #29994  |  Link
daert
Registered User
 
Join Date: May 2015
Posts: 16
Hi madshi. First I want to thank you for your hard work. I've been using madvr for a long time without any problem but the last few builds doesn't work well for me.
In the latest build (0.88.7) d3d11 is not enabled at all even if I check the option. The OSD shows always "(new path)" both in windowed mode and FSE.
In 0.88.6 the refresh rate fix doesn't work. With D3d9 FSE both 24p and 60p work but with d3d11 FSE only 60p works, 24p doesn't work yet.
Let me explain what happens: my monitor has a 23p with a blurred image and a timing of 23.970 so I created a 24p custom resolution with the exact timing of 23.976 and a clear image. When MPC enters in FSE the resolution is switched back to the blurred 23p even if I set my 24p in the nvidia control panel.
I'm using MPC-HC 1.7.8.162 x64, Lav 0.65-2 and Windows 8.1 x64. I have a gtx 660 with 350.12 and an intel q9550.
daert is offline   Reply With Quote
Old 15th May 2015, 01:07   #29995  |  Link
JarrettH
Registered User
 
Join Date: Aug 2004
Posts: 842
Quote:
Originally Posted by MysteryX View Post
And because SuperRes includes anti-ringing, can I safely remove Anti-Ringing with Jinc if combined with SuperRes?
I'd like to know too, and what is "error upscaling quality"? It's either medium or high. Sorry if I'm asking about SuperRes too soon.
JarrettH is offline   Reply With Quote
Old 15th May 2015, 01:11   #29996  |  Link
MysteryX
Soul Architect
 
MysteryX's Avatar
 
Join Date: Apr 2014
Posts: 2,173
Quote:
Originally Posted by JarrettH View Post
I'd like to know too, and what is "error upscaling quality"? It's either medium or high. Sorry if I'm asking about SuperRes too soon.
If you set it to high, rendering times go up considerably, so it is simply the quality setting.

Last edited by MysteryX; 24th June 2015 at 07:12.
MysteryX is offline   Reply With Quote
Old 15th May 2015, 01:45   #29997  |  Link
Asmodian
Registered User
 
Join Date: Feb 2002
Location: San Jose, California
Posts: 3,782
Quote:
Originally Posted by MysteryX View Post
If you set it to high, rendering times go up considerably, so it is simply the quality setting.
It probably changes the error upscale step from bilinear to something else, maybe Jinc3 based on the rendering times?
__________________
madVR options explained
Asmodian is offline   Reply With Quote
Old 15th May 2015, 02:37   #29998  |  Link
huhn
Registered User
 
Join Date: Oct 2012
Posts: 6,149
Quote:
Originally Posted by aufkrawall View Post
Funny, I had the exactly same problem with the Nvidia WDDM 2.0 driver.
Edit: But this was not related to 10 bit, but to the D3D11 renderer in general.
i could get problem with windowed D3D11 too. but 10 bit was on a whole different level. and 88.6 worked pretty good after some tweaks.

but i come back to this when windows 10 is "final".
huhn is offline   Reply With Quote
Old 15th May 2015, 02:41   #29999  |  Link
BetA13
cosmic entity
 
BetA13's Avatar
 
Join Date: May 2011
Location: outside the Box
Posts: 219
hmm,

i just tested teh new version. so far so good, enjoying teh new settings.
i only have 2 problems..

1. the artefacts are back when using nnedi3, both, in chroma upscaling and image doubling.
It was fine with v0.87.21. (was about kepler gtx670 , win 764bit and 32bit..)

2. when using d3d11 mode in FSE/windowed my monitor resolution resets and stuff is cut off..it changes my desktop resolution also and changes teh hz form 60 to 25...
with the normal d3d9 (new path) mode it works fine doh, all 60 hz...

greetz
BetA13 is offline   Reply With Quote
Old 15th May 2015, 03:08   #30000  |  Link
bozokaydin
Registered User
 
Join Date: Apr 2011
Posts: 16
i amusign windows 8 with gtx 690 nvidia driver 347.88 when i tried to install madvr 88.7 as usually right click and run as administrator to install.bat nothing happens. Just blank command screen. when i hit enter it says you have to right click to install bat and run as administrator. but i already did that. version 87.21 run with no problem no problems. how can i install 88.7 thank you. sorry for my bad english
__________________
HTPC:

Windows 8 64bit
Nvdia GTX 690
i7 3570k
Corsair H100 Water Cooling
Corsair 128 Gb Force SSD x 2 Raid 0+ Samsung 2 TBx 2 Hard Disc Raid 0
Creative Titanium HD
bozokaydin 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 08:27.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, vBulletin Solutions Inc.