madVR - high quality video renderer (GPU assisted)

[madshi]

feedback

-------

Now that we've summed up debanding, I'd love to hear your feedback about "image enhancements" and "upscaling refinement". That's a rather broad topic, covering FineSharp, LumaSharpen und SuperRes, both before and after scaling. Tons of options, so a lot to test, a lot to play with. Let me know what you think, what you like and what you don't like.

-------

Let me describe again the concept I had in mind for these algorithms: In my experience there are 2 different causes to get soft/blurry video playback:

A) Either the source is already blurry.

If this is the case, a good idea might be to sharpen the source before upscaling it any further. The reason for this is that sharpening algorithms work by analyzing the direct neighborhood of each pixel in the image. Upscaling causes the image to get bigger, which means the neighbor pixels in the original image are much further apart from each other in the upscaled image. But the sharpening algorithm only looks at the near neighborhood. Which means that if we sharpen after upscaling, although the sharpening still somewhat works, it loses much of its effectiveness. If the source is already soft to begin with, sharpening after upscaling won't do very much, so for soft sources my idea was to sharpen them before upscaling.

This is what the "madVR\processing\image enhancement" settings page is for. You can think of "softness" as being just another artifact of some sources, and we fix it by sharpening the source before we scale it. This only applies to soft sources, though, so my suggestion would be to not enable these settings by default, but only by demand, if you have a really soft source.

B) Even a sharp source becomes blurry if you upscale it a lot.

Doesn't matter which scaling algorithm you use. Even Lanczos8 and NNEDI3 become soft if you upscale an image a lot. The amount of softness introduced by upscaling is probably linear to the upscaling factor, I think. My hope here is that we can find a way to setup the "upscaling refinement" in such a way that every source maintains its natural sharpness after upscaling, regardless of which upscaling factor is used. I would like to find upscaling refinement settings which we can setup once and then forget, settings which work for any source and any upscaling factor. And I think the "secret" to make this work is by splitting the upscaling into multiple 200% upscaling steps. For each such step we apply a specific amount of sharpening, just so that the image afterwards has roughly the same sharpness as the source. If we do that, we can repeat the 200% upscaling steps multipe times, and still end up with the same sharpness the source had.

At least that's the idea. I think the option "refine the image after every ~2x upscaling step" should be the right one to use to realize the above approach. But for testing purposes I want to let you choose to only apply upscaling refinement once after all upscaling is done, so you can compare the effect this option has.

-------

Some specific questions I'd like to get an answer for:

1) In my tests medium "error upscaling quality" was quite a bit faster than high, while producing almost the same quality. Do you agree? Can I drop the "high" setting and simply always use "medium"?

2) It seems that running FineSharp and LumaSharpen *before* SuperRes has a much smaller effect than running them afterwards. So my question would be: Which in your opinion produces the better results: Running FineSharp / LumaSharpen before or after SuperRes?

3) Which combination of settings do you like most? SuperRes alone? Or in combination with FineSharp and/or LumaSharpen? Or maybe just FineSharp?

4) Does using the upscaling refinement options change your preferred upscaling/doubling algorithm? Which did you prefer without upscaling refinement? Which do you prefer with upscaling refinement?

5) Which SuperRes parameters do you like?

6) For experts: When upscaling with a large factor, which results in more than just one refinement pass (see debug OSD), do you prefer to refine the image after every ~2x upscaling step? Or do you prefer to refine the image only once after upscaling is complete? Of course when refining the image after every ~2x upscaling step, you should lower the values/passes to make the overall refinement strength more similar to refining only once.

7) For experts: Considering my goal (see above) to find a setting which works for every upscaling factor to maintain the sharpness of the original goal, which exact upscaling algorithm and which upscaling refinement options do you like? I'm hoping to end up with just a low/medium/high setting, where high would maintain the exact source sharpness, and where medium and low would be toned down settings do apply less strong sharpening. Low should provide a small but noticeable sharpness boost (compared to no refinement at all). Medium should be in the middle between Low and High.

P.S: Forgot:

8) Do you prefer "linear light" for FineSharp turned on or off?
9) Do you like the LumaSharpen "experimental limiter" turned on or off?

Thanks!! :)

[madshi]

Yep. The only way I can get the queues to fill is to reduce how many video frames shall be presented in advance to 6.

Downgrading to madVR 0.88.8, all queues fill perfectly fine. Any version newer than that, it doesn't. Isn't the workaround in 0.88.11 suppose to fix the issue and bring functionality back towards the latest build it worked? For what it's worth, using x64 version.
Oh well. If you're willing to run some tests, I could make a couple of test builds to get to the bottom of this problem on your PC. You'd need to be available for tests, though, with a reasonable reply speed, for that to make sense. What do you think?

Oh damn, I feel so horrible that Shiandow's de-banding algorithm was pulled. Maybe in the future that algorithm will get much better? I shouldn't have criticized it...:(.
Don't feel bad. I asked for feedback, and Shiandow wanted honst feedback, too. It would have helped nobody if you simply praised both algorithms. The feedback in this thread helped Shiandow to noticeably improve the quality of his algorithm. So I'm sure he doesn't regret the whole feedback and algorithm refinement cycle we did in this thread. His algorithm isn't "lost", it will be used in other renderers, just not in madVR (at least for now). FWIW, my decision was based on the feedback of several users, not just you alone.

madshi - thanks for the new madVR version! Should previous versions be uninstalled before replacing files with new version of madVR? Or maybe settings should be reset before replacing old madVR files with new ones? Are those proper instructions?
You can do that, but you don't have to. You can simply replace the files.

One more issue I have had with madVR for a VERY long time. Its about madTPG. Whenever I use dispcalGUI and madTPG, I set madTPG to "Use FullScreen" and "Stay on Top". Yet, at times when calibration starts, madTPG expands to FullScreen, but no patterns are shown. Then I press Esc or Alt+Tab and notice there is another madTPG window in the background that does display patterns. Now I always make sure there is only ONE madTPG process and only ONE mad* home cinema control process. I tried to toggle between "Disable OSD" enabling and disabling, but that made no different. Pretty much, I have to use madTPG in window mode, which is no big deal, but I don't understand why FullScreen mode creates this problem...
I think ArgyllCMS/dispcalGUI try to connect to the running madTPG instance, and if that fails they start a new instance themselves. So in your case it seems that sometimes ArgyllCMS/dispcalGUI seem to be unable to connect to the already running madTPG instance for some reason. I don't know why. It probably has nothing to do with FSE mode. I can't reproduce it on my PC. And seemingly the ArgyllCMS/dispcalGUI devs can't reproduce it, either, or they would have complained to me. So there probably isn't much we can do. You could try to narrow down the exact circumstances when this happens. Maybe that would help, or maybe not.

Is there any way to install or use madVR without administrator rights?
I think you do need at least to register the ax file using regsvr32.exe, to make most media players be able to use madVR. I'm not 100% sure if that's possible without admin rights, but I rather don't think so, because registering madVR means writing to registry keys which probably can't be written to without admin rights. There are some media players which can run madVR without installing it, though. I think JRiver MC does that. But I don't know if JRiver MC itself requires admin rights during installation or not...

[Vyral]

My feedback about "upscaling refinement" :
When upscaling from 720 to 1080 with Jinc 3AR, the image was a little bit blurry.
I've tested LumaSharpen and SuperRes but they were too strong for me.
I'm currently using FineSharp with a strength of 1.0 and linear light, everything else is default (mode=1 ; thinning=0.019 ; repair=0.25), and I'm very satisfied with it.

[madshi]

My feedback about "upscaling refinement" :
When upscaling from 720 to 1080 with Jinc 3AR, the image was a little bit blurry.
I've tested LumaSharpen and SuperRes but they were too strong for me.
I'm currently using FineSharp with a strength of 1.0 and linear light, everything else is default (mode=1 ; thinning=0.019 ; repair=0.25), and I'm very satisfied with it.
I like FineSharp a lot, too. But FWIW, LumaSharpen und SuperRes can be toned down, too, of course, e.g. by reducing the SuperRes passes, or all of the parameter numbers. It's a lot of values you can adjust, so it's kinda hard to "test it all". I know...

Of course 720 to 1080p is a rather small scaling factor (1.5x), so there the refinement options have a quite strong effect. If you choose a bigger upscaling factor (e.g. "Double Size" in MPC-HC), the sharpening by all the algorithms would be less strong. That's something I'll need to take into account, too, when defining the final Low/Medium/High settings.

[SecurityBunny]

Oh well. If you're willing to run some tests, I could make a couple of test builds to get to the bottom of this problem on your PC. You'd need to be available for tests, though, with a reasonable reply speed, for that to make sense. What do you think?


Sure thing. I'm free. :)

[baii]

Since upscaling is gonna get mentioned, I need some help on setting profile with scale factor. Not sure if I did it wrong

if (scalingFactor.x > 1 ) "upscale"

else "normal"

This is my profile rule, but don't seem to always work? Am I missing something?

My initial impression of the refinements are that

the default strength (2.0) for fine sharp is lil bit high~
lumasharpen is cheap to use
superres for 720p+ source is pretty demanding.

[Anima123]

madshi,

Does upscaling refinement works for other upscaling algorithms, like Jinc or Bicubic? And how does it differs to image doubling algorithms, as NNEDI3 or NEDI?

[madshi]

Sure thing. I'm free. :)
Great! Then please start with creating a debug log with 30 seconds of playback, using v0.88.11. Those 30 seconds should show how the queues are not filing. Please try to avoid switching back and forth between windowed <-> FSE all the time, to make the log easier to understand. Upload to e.g. tinyupload.com or similar site. Thanks!

This is my profile rule, but don't seem to always work? Am I missing something?
I don't know. Some more details would be helpful. Which video and target resolutions does it work with and which not? For each such combination please post the "movie resolution" and "target rectangle" information from the debug OSD (Ctrl+J). Just the text, no screenshots, please.

Does upscaling refinement works for other upscaling algorithms, like Jinc or Bicubic? And how does it differs to image doubling algorithms, as NNEDI3 or NEDI?
Yes, it even works for Bilinear upscaling.

Jinc, Bicubic, NNEDI3 and NEDI are all upscaling algorithms.

The upscaling refinement stuff instead is some sort of sharpening/enhancement, which is applied to the video frame after upscaling.

[foozoor]

This is what I use to improve/enhance my videos with vapoursynth.
After a lot of experimentations, this is the best for me for almost any videos.
I think it will be possible with madvr.

This is my post processing chain:

1) Uscale with needi3 (1,5x or 2x).
2) Sharpen with FineSharp with low strength (0,8 to 1,5).
3) Rescale the video to its original resolution, I use spline64.
4) Debanding.

[Anima123]

Yes, it even works for Bilinear upscaling.

Jinc, Bicubic, NNEDI3 and NEDI are all upscaling algorithms.

The upscaling refinement stuff instead is some sort of sharpening/enhancement, which is applied to the video frame after upscaling.

When using SuperRes with Bicubic, for example, the SuperRes works directly on the targeted image size, right? While the image doubling algorithms might need to be downscaled or upscaled if the target size is not exactly 2x or 4x?

There's a lot of combinations to be tried and tested.

[SecurityBunny]

Great! Then please start with creating a debug log with 30 seconds of playback, using v0.88.11. Those 30 seconds should show how the queues are not filing. Please try to avoid switching back and forth between windowed <-> FSE all the time, to make the log easier to understand. Upload to e.g. tinyupload.com or similar site. Thanks!


I'm assuming I just run the 'activate debug mode' bat file to toggle debugging mode? Running it outputs;

Access is denied.
Access is denied.
Access is denied.
Access is denied.
activating debug mode failed

Running it as admin outputs;

file "madVR.ax" not found (Though it does exist in the folder!)

Anyways, manually renamed the debug .ax file to madVR64.ax to get the log. Ran a video for 30 seconds. It created a log file 110MB in size.

Debug log: https://www.dropbox.com/s/hi8yiom7xcwayc3/madVR%20-%20log.txt?dl=0

D3D11 FSE, 10-bit, x64 player
Queues are set to the following.

CPU queue size: 24
GPU queue size: 20
frames in advance: 16

Queue fill rate was displaying as the following.

Decoder queue - 22-25/24
Upload queue - 17-20/20
Render queue - 1-5/20
Present queue - 1-8(9, bouncing around)/15

Let me know if you need me to export my madVR registry settings or test with more specific video conditions. (10 bit, 23.976 fps, etc.)

[Asmodian]

When using SuperRes with Bicubic, for example, the SuperRes works directly on the targeted image size, right? While the image doubling algorithms might need to be downscaled or upscaled if the target size is not exactly 2x or 4x?

There's a lot of combinations to be tried and tested.

Even when using doubling SuperRes (all enhancements) only run once per ~2x or once (depending on the setting). e.g. SuperRes doesn't run immediately after doubling, it runs after scaling to the target resolution.

Though there is still a lot of testing to do, of course. :D

I have noticed I prefer SuperRes only on high quality and high resolution sources, FineSharp on high quality sources, and LumaSharpen otherwise.

Both SuperRes and FineSharp seem to enhance detail (artifacts) very well.

I need to do more testing but all my fine tuning (so far) needs to be thrown out when I switch sources, even when they the same resolution.

[madshi]

This is what I use to improve/enhance my videos with vapoursynth.
After a lot of experimentations, this is the best for me for almost any videos.
I think it will be possible with madvr.

This is my post processing chain:

1) Uscale with needi3 (1,5x or 2x).
2) Sharpen with FineSharp with low strength (0,8 to 1,5).
3) Rescale the video to its original resolution, I use spline64.
4) Debanding.
So basically you're "oversampling"? Meaning you upscale to a higher resolution than the source, then sharpen in the higher resolution, then downscale again? That's a lot of processing. Does the upscaling + downscaling really bring that much benefit?

When using SuperRes with Bicubic, for example, the SuperRes works directly on the targeted image size, right? While the image doubling algorithms might need to be downscaled or upscaled if the target size is not exactly 2x or 4x?
When using SuperRes with Bicubic, still the upscaling is split into several doubling steps, before reaching the final resolution. It depends on the scaling factor, though. If the overall scaling factor is e.g. 4.1x, then with Bicubic the image is doubled to 2.0x, then scaled to 4.1x, and upscaling refinement is performend twice. With an overall scaling factor of e.g. 2.1x, scaling is only done once directly to the target resolution and upscaling refinement is only performed once.

The above applies only when "refine the image after every ~2x upscaling step" is activated, of course.

I'm assuming I just run the 'activate debug mode' bat file to toggle debugging mode? Running it outputs;

Access is denied.
Access is denied.
Access is denied.
Access is denied.
activating debug mode failed
Probably the batch needs write access to the madVR folder for your normal non-admin user to work correctly, I suppose. Not sure why it doesn't work as admin, though. Anyway, doing the renaming manually works, too, of course.

Debug log: https://www.dropbox.com/s/hi8yiom7xcwayc3/madVR%20-%20log.txt?dl=0
Argh, stupid me, my workaround didn't work properly. Does the following release mode test build work better? If not, try lowering the number of prepresented frames once again:

http://madshi.net/madVR8811test.zip

SuperRes doesn't run immediately after doubling, it runs after scaling to the target resolution.
It always does run immediately after doubling, if I remember my own code correctly. Whether that's a good idea is another question. Probably in the last step I should only run it after scaling is complete, instead.

I have noticed I prefer SuperRes only on high quality and high resolution sources, FineSharp on high quality sources, and LumaSharpen otherwise.

Both SuperRes and FineSharp seem to enhance detail (artifacts) very well.

I need to do more testing but all my fine tuning (so far) needs to be thrown out when I switch sources, even when they the same resolution.
Don't say that, it lowers my hope for finding a simple setting that works for every source... :(

No, of course do say that, I want to hear the truth, even if I don't like it...

[6233638]

Was this a one-time-only problem, or does it happen every time? Does this only happen with madTPG, or also with your media player?Every time with madTPG. Doesn't seem to have been an issue with video playback that I've seen.
It even happens without CalMAN being open. Double-clicking the madTPG window to exit full-screen mode results in a black screen and I have to hit Esc to close madTPG to get rid of it.
EDIT: Seems to intermittently work with this test build (http://madshi.net/madVR8811test.zip). Prior to that it would black out 100% of the time. Now it's more like 75% of the time.

I think the ideal approach would be to run madTPG on your HTPC, and then to use a laptop to run CalMAN. If you have to run both on the same PC, then yes, maybe that's currently not as smooth an experience as it could be. It's not something I can fix on my own, though. Would have to work with Calman together on that.That does seem like it would be best, though the only laptops I have access to right now are Macs which can't run CalMAN or madTPG.
I understand that this may not be something you can change on your own, though it would be good if some sort of "single display mode" where madTPG only shows during measurements could be supported - ideally with CalMAN pausing at the beginning of each set to allow the switch into FSE.

[James Freeman]

@madshi

I noticed an unexpected behavior in the bitdepth conversion in madVR with the OS profile.
I bought an aliexpress chinese LED projector for quick testing, and I noticed it only accepts 6bit without banding.
With 8bit+D smooth pattern has huge bands, while 6bit+D is smooth; From this I deduct that these chinese projectors only accept 6bit and lower without banding.
Bless you for having the option to select output bitdepth in madVR!

Now for the problem I see,
When I output 6bit+D smooth pattern from madVR without OS .icm color management profile, i see no banding at all...Smooth.
When I turn ON the OS color management I see (rainbow) banding with 6bit+D even though it was smooth, to get the pattern smooth again (without rainbow banding) I have to lower the bitdepth to 4bit+D in madVR...??
With 3DLUT the pattern is smooth with 6bit+D with no rainbow banding.

I can only guess that there is something wrong with bitdepth conversion in madVR of the OS profile.
Judging by eye, 4bit is actually higher when using OS profile because I see no dithering or banding, but it is correct(lower) when using 3DLUT and the dithering pattern is a lot more visible.

Let me rephrase,
8bit = banding -> crap projector.
NO system profile 6bit+D = Smooth.
System profile 6bid+D = rainbow banding.
3DLUT 6bit+D = Smooth.
System profile 4bid+D = Smooth with no dithering visible or rainbow banding.
3DLUT 4bit+D = Smooth but heavy dithering visible.

I think there is a miss match between 3DLUT and System Profile bit depths in madVR.
Please look into it.

Thanks

------

Thank you for 88.11, I'll be testing the image enhancement and report.
And congratulations for the new website... may I say, FINALLY!

[SecurityBunny]

Probably the batch needs write access to the madVR folder for your normal non-admin user to work correctly, I suppose. Not sure why it doesn't work as admin, though. Anyway, doing the renaming manually works, too, of course.

Only have one user and it is an administrator account. Should have full permission access to the madVR folder. If I had to guess why running as admin doesn't work, perhaps it is related to how running command prompt as admin changes the default directory to system32 instead of the user folder?


Argh, stupid me, my workaround didn't work properly. Does the following release mode test build work better? If not, try lowering the number of prepresented frames once again:

http://madshi.net/madVR8811test.zip

Unfortunately not. Same queue numbers as before. And similar to before, have to reduce prepresented frames to 6 to get it to fill completely. (Am able to use 8-16 perfectly fine with 0.88.8.)

[v0lt]

@madshi
Frame rate = 59.94, refresh rate = 75 Hz, madVR writes: "1 frame repeat every 3.89 minutes". I think it is wrong.

[Budtz]

Im very happy with just using finesharp. I do hope the option to ajust strength of sharpnes does not get reduced to low/medium/high. Also i am hoping for limitedsharpenfaster as i can not get it stable with avisynth. I know expanded shadersupport is needed for this one.

[madshi]

Every time with madTPG. Doesn't seem to have been an issue with video playback that I've seen.
It even happens without CalMAN being open. Double-clicking the madTPG window to exit full-screen mode results in a black screen and I have to hit Esc to close madTPG to get rid of it.
EDIT: Seems to intermittently work with this test build (http://madshi.net/madVR8811test.zip). Prior to that it would black out 100% of the time. Now it's more like 75% of the time.
Hmmmm... I seem to be able to reproduce the issue - but only when doubling clicking. Try Alt+Enter, that works for me. For you, too?

I noticed an unexpected behavior in the bitdepth conversion in madVR with the OS profile.
I bought an aliexpress chinese LED projector for quick testing, and I noticed it only accepts 6bit without banding.
With 8bit+D smooth pattern has huge bands, while 6bit+D is smooth; From this I deduct that these chinese projectors only accept 6bit and lower without banding.

Now for the problem I see,
When I output 6bit+D smooth pattern from madVR without OS .icm color management profile, i see no banding at all...Smooth.
When I turn ON the OS color management I see (rainbow) banding with 6bit+D even though it was smooth, to get the pattern smooth again (without rainbow banding) I have to lower the bitdepth to 4bit+D in madVR...??
It is known that the OS color management and DXVA color management sucks.

It is known, khaleesi.

Unfortunately not. Same queue numbers as before. And similar to before, have to reduce prepresented frames to 6 to get it to fill completely. (Am able to use 8-16 perfectly fine with 0.88.8.)
Ok, could you please create another debug log with this build:

http://madshi.net/madVR8811test2.zip

Frame rate = 59.94, refresh rate = 75 Hz, madVR writes: "1 frame repeat every 3.89 minutes". I think it is wrong.
Yes, that looks weird.

[SecurityBunny]

Ok, could you please create another debug log with this build:

http://madshi.net/madVR8811test2.zip


Debug log: https://www.dropbox.com/s/hi8yiom7xcwayc3/madVR%20-%20log.txt?dl=0

Same results as before - unfilled queues with high prepresented frames.

[RainyDog]

Hi madshi,

Thanks for including LumaSharpen. I did some testing and here is some feedback:

Looks like currently you are using the old version of LumaSharpen, which is 1.4.1. Current version is 1.5.0, which is included with Reshade + SweetFX bundle. I have already modified it for MPC-HC use, you can find it here (http://pastebin.com/rxDjqmGg).
Current madVR implementation uses Pattern 2 (9 tap gaussian, 4+1 texture fetches), although it looks nice on video games it doesn't look that good when it comes to videos (http://forum.doom9.org/showthread.php?p=1673897#post1673897). In my opinion, it would be better if you'd include the other patterns as options.
This is more of a personal request: I am normally used to MPC-HC's shader management. For instance I do always apply LumaSharpen, no matter if there is resizing or not. With madVR even though I can use profiles it gets complicated in this case. Because sometimes I watch videos downsized in windowed mode, which LumaSharpen does not apply unless it is pre-resize. And pre-resize and post-resize(upscale) applications of LumaSharpen is quite different from each other and requires additional fiddling. It would be nice if there would be an option to apply LumaSharpen always as post-resize.


If you're looking for feedback for LumaSharpen madshi, I remember seeing this post a few weeks ago and it seems to have been lost amongst the influx of new stuff happening at the time.

The first two points are certainly important things for you to look into. In the original SweetFX thread (http://forum.doom9.org/showthread.php?t=170357&highlight=sweetfx) opinion was unanimous that Pattern 3 was better for video than Pattern 2.

[6233638]

Hmmmm... I seem to be able to reproduce the issue - but only when doubling clicking. Try Alt+Enter, that works for me. For you, too?When using that test build, I tried maybe 50 times and did not see a black screen when hitting alt+enter.
The first time that I tried double-clicking after that, it turned black again.
After I got the first black screen, even using alt+enter results in a black screen every time.

I'll also point out that prior to 0.80.11 and that test build, I did try using alt+enter and was still getting black screens occasionally.

[krille]

I get a lot if dropped frames if I enable madvr;s debanding. Also the queues does not stay stable.

[cyberscott]

Hope v0.88.11 is working fine for you?


It works as well as test build 5 in DSD11 10 bit exclusive mode. As long as I set the "presented in advance" frames to 4 it works ok, though I do tend to take more repeated and dropped frames with higher scaling settings and when things get "busy" on video with action and/or a lot of subtitles displaying at the same time. I've attributed this to the lower pre-presented frames and not enough frames in advance to compensate for all the action /subtitles on screen.
It was more stable on v0. 88.8 in DSD11 10 bit exclusive on several of my "test" videos due to being able to set the pre-presented frames higher. When not in 10 bit mode, all is well with the queues.
It looks like other have started to chime in about the pre-presented queues not totally filling in DSD11 10 bit mode after build v.088.8. I was beginning to wonder if I was the only one to be experiencing this.
To end on a good note, the revised de-banding is working good for me in this build.

[vivan]

I get a lot if dropped frames if I enable madvr;s debanding. Also the queues does not stay stable.This mean that your hw is not fast enough / your settings are too demanding.
What about rendering time? Video fps, display and video resolution? What is your hardware?

[tobindac]

yeyy, still no "ding" from the installer.

[GCRaistlin]

FWIW, madVR is reporting the detected refresh rate via the madVR interface. If Reclock wanted, it could get the refresh rate from there.
Does it mean that ReClock icon could be yellow because it detects the refresh rate incorrectly? My projector supports 24 Hz, and madVR is customized to switch the projector to this mode while playing 24/23.976 fps movies, but ReClock goes yellow almost always. I have to manually change the video playback rate to get the green icon.

There is an issue with 0.88.10 and 0.88.11: the playback freezes sometimes. It happens to me a couple of times on two different BDRemuxes. The issue is unstable; I've tried to reproduce it later (same video, same time, same player, etc.) but had no luck. Maybe the issue was introduced in one of the earlier 0.88.x builds - I can't use them on the regular basis because of http://bugs.madshi.net/view.php?id=284. But I've never experienced it on 0.87.14. WinXP, GF 8800 GT, driver v340.52.

[nlnl]

madshi
Tested new image ehancement option using this image https://cloud.mail.ru/public/AbGY/UwuASurxH
The results:
Fine Sharp ON
http://i71.fastpic.ru/thumb/2015/0606/71/13ccff64eada2845eb182c947ee5e371.jpeg (http://fastpic.ru/view/71/2015/0606/13ccff64eada2845eb182c947ee5e371.png.html)
FS OFF
http://i71.fastpic.ru/thumb/2015/0606/48/63c8ef050be3d12bf0a61c92cbe5da48.jpeg (http://fastpic.ru/view/71/2015/0606/63c8ef050be3d12bf0a61c92cbe5da48.png.html)

Can not see horisontal lines when FS OFF and the same with SuperRes filter OFF for chroma upsampling.

[krille]

This mean that your hw is not fast enough / your settings are too demanding.
What about rendering time? Video fps, display and video resolution? What is your hardware?

Intel Corei5 2400, Nvidia gtx 960. Rendering g times 5,5 ms average, 7,4 ms max. 1080p mkv, gpu usage about 14%.

Hade the same problem with 0,88.10, but then I could use Shiandows debanding witch caused no problems.

[MS-DOS]

Intel Corei5 2400, Nvidia gtx 960. Rendering g times 5,5 ms average, 7,4 ms max. 1080p mkv, gpu usage about 14%.

Hade the same problem with 0,88.10, but then I could use Shiandows debanding witch caused no problems.

Do you have a trade option "don't render frames when fade in\out is detected" checked ?

[krille]

Do you have a trade option "don't render frames when fade in\out is detected" checked ?

Oh, great, that solved the problem! Thanks a lot MS-DOS! :)

[XMonarchY]

Now that we moved on to Image Refinement, may I ask - what exactly is the different between LumaSharpen, FineSharp, and SuperRes? I assume they deal with increasing sharpness, which is a bit odd because AFAIK madVR was all about making the image as soft as possible. I guess overwhelming popular requests for these options made them possible, especially the LumaSharpen option.

I tested default settings LumaSharpen (Image Enhancement) and default settings FineSharp (Image Enhancement) with both HQ and MQ content, but I did not like either at all. Both features made dithering, noise, blockiness, banding, or artifacts more obvious/visible/emphasized in MQ, LQ, and even moderately HQ content. Its like these settings canceled all the madVR's image improvement through softness. LumaSharpen and FineSharp were only helpful in absolute best HQ content (full BD's, not rips). I have not tested LumaSharpen and FineSharp Upscaling Refinement.

SuperRes as a Chroma Upscaling filter and as Image Refinement option seem to improve overall image quality, making it slightly sharper, but without emphasizing the blockiness, noise, dithering, artifacts, or banding in LQ, MQ, and HQ content. I wish I knew exactly what SuperRes was doing (without getting too technical) so I could look for those specific effects.

[Shiandow]

SuperRes as a Chroma Upscaling filter and as Image Refinement option seem to improve overall image quality, making it slightly sharper, but without emphasizing the blockiness, noise, dithering, artifacts, or banding in LQ, MQ, and HQ content. I wish I knew exactly what SuperRes was doing (without getting too technical) so I could look for those specific effects.

I commented on that a while back, but it might be a good idea to repeat it here:

There have been some changes to the algorithm, although the general idea has remained more or less the same. The steps now look more like the following:


Calculate an initial guess (using a different upscaler)
Downscale and calculate differences with original image.
Scale those differences to the final size
Improve guess by:

Softening the image
Subtracting differences with the original image
Sharpening (optional)
Removing aliasing (optional)
Removing ringing (optional)

Repeat steps 2-4 several times.


It basically tries to bring the upscaled image closer to some "ideal" image with no aliasing / ringing / blurriness, while remaining faithful to the original image.

[huhn]

It basically tries to bring the upscaled image closer to some "ideal" image with no aliasing / ringing / blurriness, while remaining faithful to the original image.

and that works pretty awesome.

[MysteryX]

SuperRes is very expensive on the GPU so I definitely can't run it more than once. For me, it's only working on 288p videos with 1 pass, and I have to say I love the result. it greatly enhances the image of low-quality videos. If quadrupling resolution, would it be better to run SuperRes twice with 1 pass or once with 2 passes?

The TV already has image enhancing features so I'm disabling those when the output is the TV, except SuperRes for 288p videos.

When displaying on the laptop screen, I like LumaSharpen but find FineSharp too strong. This is useful for low quality videos and low quality videos often have bad artifacts. Sharpening amplifies the artifacts.

Again, I love SuperRes but it is expensive to run, and somehow it is much more expensive than it should on 480p videos.

[James Freeman]

With 88.11 mpc-hc the queues emptying and will NOT filling back up after I seek, thus the videos stuttering after seeking.
I need to pause or exit full screen (not FSE) to "refresh" the queues.

EDIT:
Sometimes it happen all by itself without seeking.

EDIT2:
*On secondary screen only!

[mogli]

[...] However since v0.88.x with D3D11 activated (and frame for every VSync off) very rarely randomly in a movie drops and glitches start to occur at about 20 per second [...]
Some more with 'present a frame for every VSync' turned off:
Found a NTSC file which reliably causes constant glitches only in FSE mode [...]
As of v0.88.11 this doesn't happen anymore both with 'present a frame for every VSync' on or off. This applies both to the mentioned NTSC file and also to it happening seemingly randomly on other material.
So whatever you did mashi, thanks for that!

[huhn]

here my founds about superres so far.

superres can make bilinear into a ok resizer.

i never saw a picture where anti ring even at 1.0 would harm the picture. i totally love it!

anti aliasing can create out of place pixels when set over 0.30. i'm not a huge fan of it.

finesharp:

i don't like the results of it. it is simply not neutral on my test even at weak strength.

lumasharpen:

lumasharpen can create ringing/holoing when the radius is over 1.0. but i think it is very useful at radius 1.0.

i currently use these settings upscale refinement:

LumaSharpen:

S=0.70 C=0.035 R=1.0 no limiter

SuperRes
passes 1 quality medium (i never saw any benefits from high)
strenght 0.40 sharpness 0.25 softness 0.25 anti-aliasing 0.20 anti-ringing 1.00

upscaler is currently jinc 3 LL AR.

[cyberscott]

My thoughts about SuperRes and LumaSharpen under upscaling refinement.
LumaSharpen at default setting works real well , gives video a bit more pop.
SuperRes works well, especially with upscaled standard definition vidoes and I currently have it set to medium with the following...:
passes 2, strength .40, sharpenss, .25, anti-aliasing .25 and anti-ringing at 1.0
Setting SuperRes to high shows a slight improvement in picture quality but at a high cost of around 10-12 milliseconds in rendering time so I just keep it at medium.

[XMonarchY]

here my founds about superres so far.

superres can make bilinear into a ok resizer.

i never saw a picture where anti ring even at 1.0 would harm the picture. i totally love it!

anti aliasing can create out of place pixels when set over 0.30. i'm not a huge fan of it.

finesharp:

i don't like the results of it. it is simply not neutral on my test even at weak strength.

lumasharpen:

lumasharpen can create ringing/holoing when the radius is over 1.0. but i think it is very useful at radius 1.0.

i currently use these settings upscale refinement:

LumaSharpen:

S=0.70 C=0.035 R=1.0 no limiter

SuperRes
passes 1 quality medium (i never saw any benefits from high)
strenght 0.40 sharpness 0.25 softness 0.25 anti-aliasing 0.20 anti-ringing 1.00

upscaler is currently jinc 3 LL AR.

Is that for LQ, MQ, HQ, or full BD-quality content? LumaSharpen with LQ and MQ content makes blockiness more obvious to me, so I thought it would be useful if we provided feedback for different quality content.

[MS-DOS]

For AMD users who have slower NNEDI3 with the drivers newer than 13.12, put these two dlls (http://www11.zippyshare.com/v/291t9BAD/file.html) in your MPC-HC x64 folder and see. They're from 13.12 package I stored before moving from it to 15.5 beta.

[huhn]

Is that for LQ, MQ, HQ, or full BD-quality content? LumaSharpen with LQ and MQ content makes blockiness more obvious to me, so I thought it would be useful if we provided feedback for different quality content.

i'm currently looking for general issues like anti aliasing at 1.5 or things like that.

[QBhd]

For AMD users who have slower NNEDI3 with the drivers newer than 13.12, put these two dlls (http://www11.zippyshare.com/v/291t9BAD/file.html) in your MPC-HC x64 folder and see. They're from 13.12 package I stored before moving from it to 15.5 beta.

Could you just list the dll's so I can test on my own... I don't download such files

QB

[MysteryX]

I honestly am not noticing much debanding in the videos.

Shiandow's debanding algorithm seemed more to be an image enhancement than a debanding algorithm. It was doing a pretty good job without being expensive, but I'm not sure what it was actually doing.

Shiandow produces a slightly sharper image in a way that looks more natural and neutral than FineSharp.

[noee]

Could you just list the dll's so I can test on my own... I don't download such files

QB

amdocl64.dll
opencl.dll

[huhn]

I honestly am not noticing much debanding in the videos.

Shiandow's debanding algorithm seemed more to be an image enhancement than a debanding algorithm. It was doing a pretty good job without being expensive, but I'm not sure what it was actually doing.

Shiandow produces a slightly sharper image in a way that looks more natural and neutral than FineSharp.

debanding algo doesn't sharp anything. the image doesn't get any sharper.

[krille]

If I choose to double luma resolution with NNEDI3 the render queue doesnt fill up. It says 6-7/8. Does that matter at all?

[huhn]

If I choose to double luma resolution with NNEDI3 the render queue doesnt fill up. It says 6-7/8. Does that matter at all?

if no frames drop you should be fine it is still strange.

[AngelGraves13]

We need a deblocking option sometime in the future. The debanding works well enough.

[XRyche]

SuperRes is very expensive on the GPU so I definitely can't run it more than once. For me, it's only working on 288p videos with 1 pass, and I have to say I love the result. it greatly enhances the image of low-quality videos. If quadrupling resolution, would it be better to run SuperRes twice with 1 pass or once with 2 passes?


Again, I love SuperRes but it is expensive to run, and somehow it is much more expensive than it should on 480p videos.

That's weird. I found SuperRes to make quality considerably worse with super lo-res material (240-360p) . The more passes I have it do the worse the effect. Most of my super lo-res material is DVR/TIVO/VHS/Capture card rips from late 90's to mid 2000's. Needless to say they aren't pristine, tbh though, I don't think you will get pristine image quality at those resolution anyways. If there is any type of blocking or aliasing it makes matters worse. When I try to compensate with increasing softness or anti-alaising it creates very weird artifacts. I lack the knowledge to properly describe it but, it looks like I put plastic wrap over my screen. Strange distortions on edges and such. This is leaving the settings at default. It gets much worse when increasing softness or anti-alaising to compensate.

That being said SuperRes works quite well with good 480p and up material. With a good DVD (Friends Box Set) it can improve the image quality rather nicely even at default settings.

(EDIT)I forgot to add that SuperRes works quite well for Chroma with Jinc. It makes the colors crisp and distinctive without adding any artifacts I can see regardless of the resolution. Also when upscaling I actually prefer using SuperRes each time the image is doubled. It seems to improve perceived sharpness better that way.(EDIT)


I've pretty much stay away from LumaSharp because for me it tends to introduce too much aliasing at strengths it looks like it is doing any good to me. MadVR's version is no exception. Not to say LumaSharpen is bad by any means. I use it a lot for video games :D . It just doesn't do it for me with video.

Finesharpen is actually pretty nice for me with scaling 720p to 1080p. It adds just enough subtle sharpening to make a difference.

This is all for upscaling. I prefer not to use any sharpening for downscaling on the rare occasions I have to downscale.