madVR - high quality video renderer (GPU assisted)

[leeperry]

oh, and I thought it was PotP that was hiding the mouse cursor in FSE playback. It's of course extremely convenient IME.

[ryrynz]

Madshi, when using overlay and playing next file in folder I get a screen color change I've seen a dull green or dull pink just before playing the next file.

[Matching_Mole]

I have some Telecine Blu-ray (I'm still wandering why some companies keep using this aberration for their Blu-ray release...) and I experienced some stuttering (clearly visible visible on traveling sequences ) with Madvr IVTC "Film Mode" and I try to understand why. I will be very grateful for any help Madshi!

Currently I use the last stable version of MPC-HC (non BE version), the last version of LAV suite (splitter,Video and Audio decoder), the last version of Reclock and the last version of Madvr, so nothing very exotic.
The only specific thing is Reclock is used only as audio renderer, all video speed adaptation has been deactivated. I use Powerstrip to get a frequency of almost 47.952hz (at 0.0005hz) from my HD5750 (1Go DDR5) and my screen resolution is 1280x768. The option I used for Madvr are Jinc3 (with AR) for Chroma and Catmull-Rom (with AR) for image downscaling, all others are the default ones.
This set up give a smooth playback without any noticeable stuttering for all the progressive video. But not using the IVTC "Film Mode" for telecined Blu-ray or DVD.

Currently, the information given by Madvr is it is able to detect the cadence (3:2 for 29.97 video), the IVTC take between 12 and 14 ms and the rendering between 5 and 10ms. I have no delayed or dropped frames and no presentation glitch. so all seems quite ok but the playback is not smooth. I try different resizing algorithm (even nearest Neighbor) in order to check if this comes from my video card which could be not enough powerful but I'm not sure that have changed anything.

Do you think this comes from my video card which cannot manage IVTC and Image Resizing together? And if it is why the Madvr stats seems ok?

[e-t172]

Originally Posted by Matching_Mole

I have some Telecine Blu-ray

http://www.youtube.com/watch?v=umDr0mPuyQc

[Thunderbolt8]

for a lot of documentaries (e.g. die alpen, die südsee, wildes china, die südsee), the automatic detection whether deinterlacing is needed is wrong. in all cases madvr detects deinterlacing to be needed, but then the playback gets really choppy. it only becomes smooth when manually disabling deinterlacing.

there have been wrong guesses by madvr before, but I am not sure if it was that much before. maybe the lastest update(s) broke anything?

[ralle_h]

Hey again,

I'm currently using these settings on my desktop (having a GTX 670 and a 3570k there):



They are just mindblowing, I can't say this often enough... tons of props to madshi and nevcairiel for their genious work!

Anyway, I'm wondering about which settings could be optimal for a HTPC with an i3-2120 (SandyBridge with the small HD 2000 only)? An acquaintance gave me some good settings already (thank god, or acctually him ), but I'm wondering which algorithms I could try to touch to see if I'm getting lags/problems (there a quite a few combinations to test, so it would take a while trying them one after the other I guess) and which are simply unrealistic for my i3-2120 (so I can leaven them aside).

Thanks in advance

[pei]

Originally Posted by pei

Using madVR in mpc-hc on a Toshiba Satellite C655D (AMD) Using DXVA2, the mouse cursor disappears in full screen but if I move the mouse down, the seek bar still appears. Anyone know what I can do to fix this?

Yes, it disappears but stays that way even when I move the mouse around. Shouldn't it come back so I can see where I want to seek to?

[madshi]

Originally Posted by aufkrawall

Ok, thanks. Regarding GPU usage it makes a huge difference with 4k, but not with 1080p. CPU usage isn't really higher either, my 2500k remains in low power state.

Ok, good to know.

Originally Posted by michkrol

I've encountered a rather strange bug, happens randomly and with quite specific settings. Here's how to reproduce:

1. Enable both "use separate device for presentation" and "use separate device for DXVA processing".
2. Set "DXVA2" for image up-/downscaling.
3. Play a h264 10bit file in windowed mode, while using downscaling (or upscaling).
4. The video is "jumping", as if instead of next frame, the previous frame gets rendered.
   What I mean is instead of frames 1 2 3 4, you get frames 1 2 1 4 or something.

I've downloaded your files, I'll look at them later, but it will take a while.

Originally Posted by hannes69

Remember this problem madshi?
I have redone the tests with newest madvr build. I also noticed that I didn´t have newest VGA bios so I flashed the newest one. Don´t know if the change that now happened comes from VGA bios or madvr (don´t even know if decoding matrix has to do with settings in VGA bios...)

DXVA decoding still delivers right colors. And DXVA scaling for BT709 now seems to be the same like in your setup (very similar numbers in decoding matrix:

AnalyzeGpuNv12ToRgb(VideoProcessBlt): +, bitdepth: 17, yFactor: 0.99998474, yOffset: 0.00000000, range: 15.99975586..234.99641418
YCbCr -> RGB decoding matrix used by GPU:
0.99998474 0.00000000 1.53823853
0.99998474 -0.18417358 -0.45913696
0.99998474 1.81320190 0.00000000

Difference now: bitdepth 17 instead of 32 ???

Using DXVA scaling with BT601 content now leads to following numbers:

AnalyzeGpuNv12ToRgb(VideoProcessBlt): +, bitdepth: 17, yFactor: 0.99998474, yOffset: 0.00000000, range: 15.99975586..234.99641418
YCbCr -> RGB decoding matrix used by GPU:
0.99998474 0.00000000 1.73019409
0.99998474 -0.21400452 -0.51620483
0.99998474 2.11410522 0.00000000

The numbers have slightly changed in comparison to old test, but it´s more or less the same decoding matix. I don´t know the numbers for a correct BT601 decoding matrix, but the numbers given are the ones used.

So now DXVA decoding and BT709 DXVA scaling are right, BT601 I don´t know....

So the problem is fully fixed now?

Originally Posted by bugmen0t

I tried the avisnyth script below to do a 24p to 60p conversion using blending (my first script, so no idea if it's correct).
It causes the madVR render times to decrease massively. How can that be? There must be something wrong, right?

That doesn't really make much sense, but then, rendering times aren't always the most reliable measurements. Just check whether playback is smooth with no dropped frames.

Originally Posted by Peuj

I've tried to reset the settings using "restore default settings.bat" but nothing has changed.
I've looked in the registry to "HKEY_CURRENT_USER\Software\madshi\madVR" and there is only the key "LastSettingsKey".

How can I do the reset?

The batch file should do the trick. Make sure the "settings.bin" file in the madVR folder is deleted, too.

Originally Posted by aufkrawall

I found windowed overlay to be very annyong when moving player's window while in a DVD menu, it was unusable for some seconds after.

Which GPU, OS, decoder, splitter? Aero on/off?

Originally Posted by cafevn

madVR - crash report

Thanks, will look at that later...

Originally Posted by ajp_anton

From another forum:

Oh well. I've cropped to mod 4 to make DXVA deinterlacing and scaling happy when the media player uses a cropped source rectangle. I guess I could disable the mod 4 cropping if the default source rectangle is used. That should fix the problem.

Originally Posted by secvensor

I suggest to give the chance to adjust color and a transparency of panels Exclusive mode.

Maybe in some far away future version. This is minor cosmetical stuff, which is simply not important enough to bother about at the moment.

Originally Posted by turbojet

I spoke too soon about auto play working again with latest madvr+mpc-be, it only works when player is opened on primary display then moved to extended display when it used to pause. If player is open open extended display it starts paused every time still. Madshi can you let me know if it's madvr or mpc-be's issue to fix?

I've already explained it in detail in one of my earlier posts.

Originally Posted by ryrynz

I honestly don't think my eyes are deceiving me, overlay looks to be smoothest option.
I don't have dropped frames without overlay, possible or placebo? Each time I look at overlay I can see it, like it's rendered to the screen ever so slightly faster.

It depends on the GPU. On my PC (with Aero off) the difference between Overlay and normal Windowed mode is like night and day. But that's because my GPU is very slow.

Originally Posted by annovif

Are you sure there is the same quality in this Jinc3 than before? I have a bad feeling about this and jinc3 works too good to be ruined by compromise.

Actually the quality might be ever so slightly better because I'm using exact weights instead of using a lookup table. But I doubt anybody could see the difference with naked eyes.

Originally Posted by 06_taro

I wonder if upscaling to 2x by Jinc3 and then downscaling to 1.9x by Bicubic (or any other 2-tap stuffs, or even Bilinear) would be faster than directly upscaling to 1.9x by Jinc3 or not since 0.85.5. If it's significantly faster, it might probably worth a try even if there is a tiny quality loss.

2x Jinc + Bilinear would probably be quite a bit faster than 1.9x Jinc, but then, Bilinear might damage quality a bit. Maybe not too much with 1.9, but the bigger the downscaling factor, the bigger the damage.

Originally Posted by Joniii

Using 1080p23 in display modes, changes to 24p even with PAL DVD's. Also using 720p50 or 1080p50 causes some occasional crashes when entering full screen or windowed.

There's a well known bug in win8 where the GPU driver (or the OS) switches to a wrong refresh rate. It doesn't seem to be madVR's fault. Can't do much about the crashes without a proper crash report.

Originally Posted by aufkrawall

Edit: I also think presented frames in advance should be set to 8 (at least) by default, 4 is definitely too few for 4k or other heavy stuff (10 bit I444 etc).

I even need to set it to 12 on my slow NVidia 9400 HTPC to get totally smooth playback. But then, there are negative side effects. E.g. switching between exclusive and windowed mode gets slower. And some people have better results with a smaller presentation queue. So it's really hard to choose the "right" default value.

Originally Posted by kerman

It says to support with external dxva decoders, but if I enable coreavc dxva acceleration I get a full green screen... is it a bug??

Which GPU and which OS? Are you using madVR default settings? Does the same problem occur when using LAV Video Decoder?

Originally Posted by turbojet

Is there any way to play 2 videos simultaneously with dxva2 resizing without a bunch of dropped frames?
It's not a load issue, 2 videos with Jinc work fine on same hardware.

Must be some bottleneck in DXVA, I would guess. But I can't really say for sure. I don't think it's madVR's fault...

Originally Posted by pie1394

With P10 input contents, DXVA scaling causes higher CPU loading than Bilinear scaling on 1920x1080p24 --> 960x540p60 at window mode.

Yes, and it makes sense. For DXVA scaling, madVR has to convert the 10bit video to 8bit first. This is done in hand optimized SSE2 code, but it's still taking CPU time. When using Bilinear scaling, the CPU doesn't have to do this conversion.

Originally Posted by shawkie

I'm running MPC-HC + madVR on an HD4000. At the moment I have disabled both the internal MPEG2 decoders in MPC-HC and enabled the Intel decoders in madVR instead. I'm having a couple of problems with DVD menus. Firstly, I'm not getting any kind of highlight on the selected menu item - this is going to make it impossible to navigate the menu using a remote control. Secondly, on "The Business" (UK) the menu doesn't respond to the mouse either.

The internal decoders don't really support DVD playback at the moment. If you want to do DVD playback, you should probably disable the internal decoders, or set LAV Video Decoder to "preferred" in the MPC-HC configuration.

Originally Posted by Luv

Madshi,there's a Intel Media SDK 2012 available.Is there something that needs to be updated (I have the Intel MPEG2 decoder enabled)?

Maybe they've updated the MPEG2 decoder, but I doubt they might it any faster. Still, I guess I should update the Intel decoder download.

[madshi]

Originally Posted by Niyawa

I still want to know about the overlay bug thing, you said you would look up into it. Just to add more, it seems that it also changes my video brightness when it's enabled (making it somewhat more bright), and it seems it's not something normal.

Not sure what I said I would look into? I don't remember. Video overlay is managed by the GPU, so there is some "danger" that it might use different output levels. If your output is more bright when using Overlay then either your GPU outputs video levels instead of PC levels with Overlay, or maybe some color controls in your GPU driver's control panel are not activated for some reason. It's hard for me to say.

Originally Posted by ralle_h

first of all I wanted to thank you for your great and continuous work on the best renderer out there. I really appreciate your efforts!

However, I'm having issues with madVR on one of my setups (on a lot of other desktop and HTPC's I did build there are no problems, so I doubt it is the configuration), which is an i3 2120 with the HD2000 IGPu, 8GB ram on a H77 PantherPoint Mainboard, running Windows Server 2008 R2. Furthermore I use the latest version of MPC-BE and LAV Splitter + Decoders for playback.

I think the OS could be the problem here, but I'm not sure and would appreciate any hint or tip coming from your side (or any other user who had similar issues or ideas about this).

1. Even though it's not true, the madVR menu shows that 2 instances of the player (in my case MPC-BE)

Do you have "madVR" set as "preferred" in the list of filters? If so, please remove that. It seems that MPC-BE loads one madVR instance for decoding, and another one for displaying. That might be reponsible for the issues you're seeing.

Originally Posted by ryrynz

Madshi, can overlay's transition to windowed mode fullscreen be improved? The windowed image tends to stick on screen for a bit, with the enlarged image sometimes shifting back and forward a few frames before playing normally whilst the windowed image is still visable in the top left corner. FSE mode when enabled provides a much smoother transition without any issues. On HD3000/4000 7 x64.

I don't think I can do much about this. But I'm not going to try now, anyway. I consider this a cosmetical issue, and as such it's WAY down my priority list.

Originally Posted by xabregas

im going to buy a new gpu to replace my old 8800GT because 8800 GT doesnt have directx11 support and can`t handle well madvr in windows 7 with hardware acceleration...

OFC im going to make a bit of light gaming, so i want to know wich of this 2 graphic cards is the better for madvr:

GTX 650 Ti 1GB/2GB SSC from EVGA
HD 7770 1GB OC from MSI

That's hard to say. Personally, I'd take the NVidia for various reasons. But maybe next week I'll have a different opinion, I don't know...

Originally Posted by Jong

If I load MC18 straight into HTPC friendly "Theaterview" MadVR does not switch into exclusive mode - stays in full screen windowed mode, but if I open first in "standard mode", or switch to standard mode and then back to Theaterview, MadVR will immediately enter exclusive mode. madVR settings related to this are all default.

What could make MadVR not switch to exclusive mode although its settings say it should? Any ideas very welcome!

I don't know. A log might help. Make sure you turn the OSD on, that makes the log more helpful for these kind of issues. Then reproduce the situation where madVR stays in fullscreen windowed mode. Do *not* do anything else. I don't need to see the other situation where it works in the log.

Originally Posted by nub234

I noticed my media player was freezing recently when frame stepping and bookmarking videos, so I tried an alternate player which to my surprise gave me the same issue. I narrowed it down to being due to madVR after messing around with all my filters and so on, and found that changing the renderer solved the freezes. I've reproduced the issue on both PotPlayer and MPC-HC and tried restoring to default settings with no luck.

Can't reproduce that here. A freeze report might help. Reproduce the freeze, then press Ctrl+Alt+Shift+Break maybe 3 times, with 5 seconds in between, then on your desktop should be 3 freeze report text files. Zip them and upload them somewhere for me to look at.

Originally Posted by G_M_C

About linear light scaling in madVR. I would like some more information about that feature. I understand that it does scaling without correcting for shift in gamma, and i got the impression that -maybe only in theory- it can lead to better/more realistic results (see for instance: http://www.glennchan.info/articles/t...inearLight.htm ).

Some time go i read some advise not to use it in madVR because it because it adds ringing (see madshi's post here: http://forum.doom9.org/showthread.ph...85#post1532185 )?
But is that still true now madshi implemented his anti-ringing filter?

Maybe someone can explain this feature a bit better? And also if/when/why it gets better or worse quality.
For me, some explanation would make it easier to understand the tickbox, and to make a better choice in using it or not.

It's not recommended for upscaling because it introduces more problems than it solves for most video content. For downscaling, especially when using rather large downscaling factors, it can help to produce more faithful results, but it also introduces quite a lot of additional ringing, so only use it in combination with the anti-ringing filter, and also use a 2-tap algorithm, only. E.g. Catmull-Rom is recommended for that specific purpose.

Originally Posted by Joniii

How does madVR launch the seekbar. Does it detect when mouse moves on the bottom

Yes.

Originally Posted by pei

Using madVR in mpc-hc on a Toshiba Satellite C655D (AMD) Using DXVA2, the mouse cursor disappears in full screen but if I move the mouse down, the seek bar still appears. Anyone know what I can do to fix this?

That's weird. On my PC the mouse cursor disappears if you don't move it for a while. Then as soon as you move the mouse, the cursor comes back. Not sure why that doesn't work for you. Might have to do with the GPU/drivers. Does that not happen if you disable DXVA2?

Originally Posted by ryrynz

Madshi, when using overlay and playing next file in folder I get a screen color change I've seen a dull green or dull pink just before playing the next file.

That's probably the color Windows has chosen for Overlay.

Originally Posted by Matching_Mole

I have some Telecine Blu-ray (I'm still wandering why some companies keep using this aberration for their Blu-ray release...) and I experienced some stuttering (clearly visible visible on traveling sequences ) with Madvr IVTC "Film Mode" and I try to understand why. I will be very grateful for any help Madshi!

Currently I use the last stable version of MPC-HC (non BE version), the last version of LAV suite (splitter,Video and Audio decoder), the last version of Reclock and the last version of Madvr, so nothing very exotic.
The only specific thing is Reclock is used only as audio renderer, all video speed adaptation has been deactivated. I use Powerstrip to get a frequency of almost 47.952hz (at 0.0005hz) from my HD5750 (1Go DDR5) and my screen resolution is 1280x768. The option I used for Madvr are Jinc3 (with AR) for Chroma and Catmull-Rom (with AR) for image downscaling, all others are the default ones.
This set up give a smooth playback without any noticeable stuttering for all the progressive video. But not using the IVTC "Film Mode" for telecined Blu-ray or DVD.

Currently, the information given by Madvr is it is able to detect the cadence (3:2 for 29.97 video), the IVTC take between 12 and 14 ms and the rendering between 5 and 10ms. I have no delayed or dropped frames and no presentation glitch. so all seems quite ok but the playback is not smooth. I try different resizing algorithm (even nearest Neighbor) in order to check if this comes from my video card which could be not enough powerful but I'm not sure that have changed anything.

Do you think this comes from my video card which cannot manage IVTC and Image Resizing together? And if it is why the Madvr stats seems ok?

IVTC is done on the CPU at the moment. If madVR doesn't report any dropped frames and no presentation glitches then my best guess is that for some reason the timings are screwed up so that instead of having every frame being shown for 2 vsyncs, some are shown for 3 vsyncs and some only for 1. Maybe if you create a debug log, I could confirm that guess.

Is there a reason you can't use simple 1080p24 output? That way the problem would probably go away. You could also try different video decoders and/or splitters, just to make sure it's not related to that.

Originally Posted by Thunderbolt8

for a lot of documentaries (e.g. die alpen, die südsee, wildes china, die südsee), the automatic detection whether deinterlacing is needed is wrong. in all cases madvr detects deinterlacing to be needed, but then the playback gets really choppy. it only becomes smooth when manually disabling deinterlacing.

There isn't really much information madVR can use for guessing at the moment, so the guess will be often wrong. It shouldn't matter too much, though, if your PC is fast enough. I guess you're GPU can't handle this content with the "double" framerate deinterlacing produced by DXVA2? Or do you have madVR set to forced film mode? There's also a switch in the madVR settings which allows you to fine tune whether madVR enabled/disables deinterlacing if it isn't sure what to do. Maybe you have that set the "wrong" way? Although there isn't really a wrong way there. But it might work better for you the other way, I don't know...

Originally Posted by ralle_h

Anyway, I'm wondering about which settings could be optimal for a HTPC with an i3-2120 (SandyBridge with the small HD 2000 only)? An acquaintance gave me some good settings already (thank god, or acctually him ), but I'm wondering which algorithms I could try to touch to see if I'm getting lags/problems (there a quite a few combinations to test, so it would take a while trying them one after the other I guess) and which are simply unrealistic for my i3-2120 (so I can leaven them aside).

If playback is smooth with those scaling settings you're using then there's not much you can improve, in terms of scaling. Only your downscaling settings look a bit unusual to me. But if you like it that way, why not. High quality default settings for downscaling would be either Lanczos3 AR without linear light, or alternatively Catmull-Rom AR with linear light. But in the end, trust your eyes. If you like your current settings better, then stick to them.

[Niyawa]

Originally Posted by madshi

Not sure what I said I would look into? I don't remember. Video overlay is managed by the GPU, so there is some "danger" that it might use different output levels. If your output is more bright when using Overlay then either your GPU outputs video levels instead of PC levels with Overlay, or maybe some color controls in your GPU driver's control panel are not activated for some reason. It's hard for me to say.

I can't get screenshots with Overlay enabled but it seems the brightness is normal now. It only gets like this when DXVA2 is enabled, strange thing. Thanks either way.

Oh yeah, someone asked me a question in my MPC guide thread:

"It it possible to use Floating Point Processing and 10-bit output together, or is it only one or the other? Again, it sounds stupid, and I'd like to go into more detail but I'm not good at wording tech stuff even though my brain can process it pretty well. And which one would be the best in terms of quality? 10-bit would give better colours, I know that, and that's something I'd like to check out, but would the 32-bit processing give an overall better-looking result, even when dithered to 8-bit? I know that's how MadVR works.

I am aware that you need a 10-bit monitor and compatible card for 10-bit output."

He already made a search on the matter, but both of us never found anything solid to go on in a conclusion. I'm not familiar with FPP and 10-bit output myself and neither how exactly they work, so if anyone can help with an explanation I would appreciate.

[jkauff]

Originally Posted by pei

Yes, it disappears but stays that way even when I move the mouse around. Shouldn't it come back so I can see where I want to seek to?

Yes, it should, and it does for me in MPC and Zoom Player using DXVA2 and madVR, so I don't think that's where the issue is. I'm using Win7 64 Aero, but PowerDVD turns off Aero and the pointer still reappears when the mouse is moved, so I don't think Aero is the issue. Maybe your video driver?

[Matching_Mole]

Originally Posted by madshi

IVTC is done on the CPU at the moment. If madVR doesn't report any dropped frames and no presentation glitches then my best guess is that for some reason the timings are screwed up so that instead of having every frame being shown for 2 vsyncs, some are shown for 3 vsyncs and some only for 1. Maybe if you create a debug log, I could confirm that guess.

Is there a reason you can't use simple 1080p24 output? That way the problem would probably go away. You could also try different video decoders and/or splitters, just to make sure it's not related to that.

I made 2 different tests where you will find the attached log:
- the first one at 48hz:
http://www.mediafire.com/?twb7bhva342auqd
- the second one at 47.952hz:
http://www.mediafire.com/?41dndx6bjszhb4o

I have stuttering for this two configurations. Unfortunately, I can't use 1080p24 with my projector (Mitsubishi HC3100). its 1080 mode is interlaced and I try to have 24hz in its 768p mode and it reacted not very well.

I will take time to try with different decoders, splitters and players as you suggest.

[oddball]

Has anyone been playing with Complex Sharpen 2? I find it a bit agressive and causing ringing/aliasing on detailed images. Has anyone edited the shading code in MPC to see if they can get a nicer and less aggressive shader algorithm? Something that works nice with Jinc 3 tap perhaps?

[madshi]

Originally Posted by Niyawa

I can't get screenshots with Overlay enabled

Screenshots with Print-Screen? Those never work with Overlay. However, using the media player screenshot functionality should work.

Originally Posted by Niyawa

Oh yeah, someone asked me a question in my MPC guide thread:

"It it possible to use Floating Point Processing and 10-bit output together, or is it only one or the other? Again, it sounds stupid, and I'd like to go into more detail but I'm not good at wording tech stuff even though my brain can process it pretty well. And which one would be the best in terms of quality? 10-bit would give better colours, I know that, and that's something I'd like to check out, but would the 32-bit processing give an overall better-looking result, even when dithered to 8-bit? I know that's how MadVR works.

I am aware that you need a 10-bit monitor and compatible card for 10-bit output."

madVR always use floating point processing. 10bit output has nothing whatsoever to do with the way the processing is done. So if/when madVR gets 10bit output capability, it will be together with (and not instead of) floating point processing.

10bit output will *not* give better colors. The only thing it will do is lower the dithering noise. You can play with the native bitdepth of the display in the "device" part of the madVR settings dialog. If you set the display to 7bit the dithering noise will be stronger. If you set it to 6bit, it will be stronger again. You should easily see the noise at 6bit. Probably even still at 7bit. The difference between 6bit and 7bit should be more obvious than the difference between 7bit and 8bit. Now just imagine there were 9bit and 10bit options available. If you go from 6bit to 7bit, then to 8bit. The same difference you see there you will also see when going to 9bit and then 10bit. Just the noise level will be lower, but the colors will stay identical.

Originally Posted by Matching_Mole

I made 2 different tests where you will find the attached log:
- the first one at 48hz:
http://www.mediafire.com/?twb7bhva342auqd
- the second one at 47.952hz:
http://www.mediafire.com/?41dndx6bjszhb4o

I have stuttering for this two configurations. Unfortunately, I can't use 1080p24 with my projector (Mitsubishi HC3100). its 1080 mode is interlaced and I try to have 24hz in its 768p mode and it reacted not very well.

I will take time to try with different decoders, splitters and players as you suggest.

Yeah. Here's an extract from the log:

337: (567.95-567.95 568 112529078), 339: (571.16-571.16 571 113196412), 340: (571.96-571.96 572 113363246), 341: (573.96-573.96 574 113780329),
342: (575.97-575.97 576 114197412), 344: (579.18-579.18 579 114864745), 345: (579.98-579.98 580 115031580), 346: (581.99-581.99 582 115448662),
347: (583.99-583.99 584 115865745), 349: (587.19-587.19 587 116533078), 350: (588.39-588.39 588 116783329), 351: (589.99-589.99 590 117116996)

You can see, that the planned vsync numbers for the frames 337 to 351 are:

568, 571, 572, 574, 576, 579, 580, 582, 584, 587, 588, 590

The ideal vsync numbers for smooth playback should be:

568, 570, 572, 574, 576, 578, 580, 582, 584, 586, 588, 590

The cause of the "uneven" vsync numbers are the timestamps which are a result of the timestamps coming from the decoder combined with the IVTC processing. Currently madVR takes these timestamps and calculates the vsync number at which each frame should be shown. In your case that results in non-smooth playback, unfortunately. I guess I could add some sort of "smoothing" algorithm which, after looking at those vsync results, tries to achieve more regular vsync intervals. But then what happens with true VFR (variable framerate) content? I guess I'd first have to write an algorithm to detect whether the decoded (and IVTC'ed) content is VFR or CFR. But this can all become relatively complicated fast.

I do have on my to do list to revisit the whole presentation logic and find more optimal solutions for when the display refresh rate is significantly higher than the movie framerate (as in your case). But it will take some time before I get to that. So for now I'd suggest to try different splitter and decoders to see if any of them make a difference. If not then I don't have a quick solution available right now...

[nub234]

Originally Posted by madshi

Can't reproduce that here. A freeze report might help. Reproduce the freeze, then press Ctrl+Alt+Shift+Break maybe 3 times, with 5 seconds in between, then on your desktop should be 3 freeze report text files. Zip them and upload them somewhere for me to look at.

I've uploaded both a debug log as I finally managed to produce one, and I've also produced freeze reports. I noticed that the debug log seemed to log everything, so just incase it helps, the freeze occured once at 1second into playback, and once at end of playback which was also when I closed the player, so should be towards the end of the log. Both times I was frame stepping in mpc-hc.

http://www73.zippyshare.com/v/92141861/file.html

[madshi]

Originally Posted by nub234

I've uploaded both a debug log as I finally managed to produce one, and I've also produced freeze reports. I noticed that the debug log seemed to log everything, so just incase it helps, the freeze occured once at 1second into playback, and once at end of playback which was also when I closed the player, so should be towards the end of the log. Both times I was frame stepping in mpc-hc.

http://www73.zippyshare.com/v/92141861/file.html

I'm not sure, do we have a different definition of what "freeze" means? Please describe exactly what you mean. My understanding of a freeze is if the media player stops responding. The menu doesn't work, anymore, you can't press buttons anymore etc. Is that what happened? But usually such a freeze is "forever". Now you're saying there were 2 freezes. That doesn't really match my definition of a freeze. I guess I first need a detailed description of what happened exactly and what your definition of a "freeze" is.

In the debug log I can see that playback is paused about 1.5 seconds after madVR was created. Then about 7.5 seconds later you've pressed Ctrl+RightArrow to execute a frame step.

[leeperry]

Originally Posted by madshi

The cause of the "uneven" vsync numbers are the timestamps

Indeed, and that's what's Reclock's for as it makes 24p@48.000Hz dead smooth on the HC3100.

[Matching_Mole]

Originally Posted by madshi

Yeah. Here's an extract from the log:

337: (567.95-567.95 568 112529078), 339: (571.16-571.16 571 113196412), 340: (571.96-571.96 572 113363246), 341: (573.96-573.96 574 113780329),
342: (575.97-575.97 576 114197412), 344: (579.18-579.18 579 114864745), 345: (579.98-579.98 580 115031580), 346: (581.99-581.99 582 115448662),
347: (583.99-583.99 584 115865745), 349: (587.19-587.19 587 116533078), 350: (588.39-588.39 588 116783329), 351: (589.99-589.99 590 117116996)

You can see, that the planned vsync numbers for the frames 337 to 351 are:

568, 571, 572, 574, 576, 579, 580, 582, 584, 587, 588, 590

The ideal vsync numbers for smooth playback should be:

568, 570, 572, 574, 576, 578, 580, 582, 584, 586, 588, 590

The cause of the "uneven" vsync numbers are the timestamps which are a result of the timestamps coming from the decoder combined with the IVTC processing. Currently madVR takes these timestamps and calculates the vsync number at which each frame should be shown. In your case that results in non-smooth playback, unfortunately. I guess I could add some sort of "smoothing" algorithm which, after looking at those vsync results, tries to achieve more regular vsync intervals. But then what happens with true VFR (variable framerate) content? I guess I'd first have to write an algorithm to detect whether the decoded (and IVTC'ed) content is VFR or CFR. But this can all become relatively complicated fast.

I do have on my to do list to revisit the whole presentation logic and find more optimal solutions for when the display refresh rate is significantly higher than the movie framerate (as in your case). But it will take some time before I get to that. So for now I'd suggest to try different splitter and decoders to see if any of them make a difference. If not then I don't have a quick solution available right now...

Thanks for your answer, many things are more clear now!

But just for my understanding, the IVTC corrected frame rate is 23.976fps, right? I try to use the exact double framerate to achieve smooth playback and it work perfectly on progressive 23.976 framerate. Here you said that I should have no stuttering if I were able to have 24hz refresh rate. Why the double framerate is an issue for the IVTC Film Mode currently?

Anyway, thank you very much again for your help and for the IVTC film mode which is already the greater option for the watch "live" telecine Blu-ray (and DVD) even if it seems that it can be improved again.

[Matching_Mole]

Originally Posted by leeperry

Indeed, and that's what's Reclock's for as it makes 24p@48.000Hz dead smooth on the HC3100.

Thanks for the tip, I will try also.

[nub234]

Originally Posted by madshi

I'm not sure, do we have a different definition of what "freeze" means? Please describe exactly what you mean. My understanding of a freeze is if the media player stops responding. The menu doesn't work, anymore, you can't press buttons anymore etc. Is that what happened? But usually such a freeze is "forever". Now you're saying there were 2 freezes. That doesn't really match my definition of a freeze. I guess I first need a detailed description of what happened exactly and what your definition of a "freeze" is.

In the debug log I can see that playback is paused about 1.5 seconds after madVR was created. Then about 7.5 seconds later you've pressed Ctrl+RightArrow to execute a frame step.

The player does indeed stop responding, and Windows 8 Displays a not responding window with End Task, but it then starts responding again and has skipped playback. I'll grab a screenie of exactly how it appears in a sec.

The exact steps I took to get the freeze are:

Start playback -> Pause -> Hit Frame Step a few times to try and get to a particular frame -> Freeze Occurs.

It occurs more often when holding down Frame Step to navigate over a longer period of time say a second or so, and like I mentioned in an earlier post this only happens with 0.85.2 onwards.

The screenshot: http://i.imgur.com/ytrOq.jpg