HAvsFunc

[Jindadil007]

The script looks good, maybe you are trying to load a 32bit dll and using the 64bit vspipe or the other way around.
Where did the dll come from? Where exactly did you put it?

Hi...Its 64 bit dll for vapoursynth downloaded from this forum only...

[Are_]

FFT3DFilter requires libfftw3f-3.dll to be in the search path. http://www.fftw.org/install/windows.html
Put it in system32 folder. I think the same directory as the main plugin should also work with vapoursynth.

Yeah, that's probably it.

[Jindadil007]

FFT3DFilter requires libfftw3f-3.dll to be in the search path. http://www.fftw.org/install/windows.html
Put it in system32 folder. I think the same directory as the main plugin should also work with vapoursynth.

Its already there in System32 as well as SysWOW64 folder. Would try to paste it in Vapoursyn th plugin folder and test the results. Thanks

[Jindadil007]

FFT3DFilter requires libfftw3f-3.dll to be in the search path. http://www.fftw.org/install/windows.html
Put it in system32 folder. I think the same directory as the main plugin should also work with vapoursynth.

Many Thanks Cking...Putting it in Vapoursynth plugin folder did the trick...:thanks:

Can you please share command for x264 output with 2 pass encoding ? I need to capture these parameters in output mkv file ... subq=6: partitions=all:8x8dct:me=umh:frameref=5:bframes=3:b_pyramid=normal:weight_b

:thanks:

[Mystery Keeper]

Many Thanks Cking...Putting it in Vapoursynth plugin folder did the trick...:thanks:

Can you please share command for x264 output with 2 pass encoding ? I need to capture these parameters in output mkv file ... subq=6: partitions=all:8x8dct:me=umh:frameref=5:bframes=3:b_pyramid=normal:weight_b

:thanks:My preset for x264 encoding with Y4M header in VapourSynth Editor r13:--demuxer y4m
--pass 1
--bitrate 3000
--stats "{sd}/{sn}.stats"
--slow-firstpass
--min-keyint 1
--b-adapt 1
--ref 5
--bframes 4
--vbv-init 1.0
--ratetol 100.0
--qcomp 0
--cplxblur 0
--qblur 0
--rc-lookahead 250
--aq-mode 2
--merange 64
--me tesa
--subme 11
--partitions all
--trellis 2
--no-dct-decimate
--stitchable
--direct auto
--no-psy
--no-progress
-
-o "{sd}/{sn}.mkv"

[Jindadil007]

My preset for x264 encoding with Y4M header in VapourSynth Editor r13:--demuxer y4m
--pass 1
--bitrate 3000
--stats "{sd}/{sn}.stats"
--slow-firstpass
--min-keyint 1
--b-adapt 1
--ref 5
--bframes 4
--vbv-init 1.0
--ratetol 100.0
--qcomp 0
--cplxblur 0
--qblur 0
--rc-lookahead 250
--aq-mode 2
--merange 64
--me tesa
--subme 11
--partitions all
--trellis 2
--no-dct-decimate
--stitchable
--direct auto
--no-psy
--no-progress
-
-o "{sd}/{sn}.mkv"

Many Thanks "MK" ... I would try it... Currently working on QTGMC script where i am getting errors...:thanks:

[Jindadil007]

Thank you so much.
Can't wait for testing QTGMC on vapoursynth.

Thanks for sharing

Hi...Script works fine till I introduce QTGMC in script. It is showing this error :

Failed to evaluate the script:
Python exception: There is no attribute or namespace named fmtc
Traceback (most recent call last):
File "src\cython\vapoursynth.pyx", line 1491, in vapoursynth.vpy_evaluateScript (src\cython\vapoursynth.c:26905)
File "C:/DVD/Sample.vpy", line 6, in
video = core.d2v.Source(r'C:\DVD\Sample.d2v')
File "C:\Program Files\Python35\lib\site-packages\havsfunc.py", line 943, in QTGMC
bobbed = Bob(clip, 0, 0.5, TFF)
File "C:\Program Files\Python35\lib\site-packages\havsfunc.py", line 4128, in Bob
clip = core.std.SeparateFields(clip, tff).fmtc.resample(scalev=2, kernel='bicubic', a1=b, a2=c, interlaced=1, interlacedd=0)
File "src\cython\vapoursynth.pyx", line 833, in vapoursynth.VideoNode.__getattr__ (src\cython\vapoursynth.c:16042)
AttributeError: There is no attribute or namespace named fmtc

I have following 64bit dll files in Vapoursynth plugins 64 folder : AddGrain, continuity, d2vsource, dctfilter, DFTTest, ffms2, FFT3DFilter, flash3kyuu_deband, KNLMeansCL, libawarpsharp2, libbifrost, libfftw3-3, libfftw3f-3, libfftw3l-3, libmvtools, libnnedi3, ReduceFlicker, scenechange, TDeintMod, temporalsoften, TNLMeans, vcmod, vs_sangnommod, vsavsreader & vslsmashsource. All these dll files have been copied to SysWOW64 as well as system32 folder.

Here is my script :

import vapoursynth as vs
import adjust
import havsfunc as haf
core = vs.get_core()
video = core.d2v.Source(r'C:\DVD\Sample.d2v')
video = haf.QTGMC(video, Preset='Medium', TFF=True)
video.set_output()

Any help in getting QTGMC working would be highly appreciated... :thanks:

[ChaosKing]

1. Don't copy the filters in system32.
2. As the error msg shows, you are missing another plugin: fmtc

Here's a nearly complete list of availible plugins: http://www.vapoursynth.com/doc/pluginlist.html
You need fmtconv – High quality bitdepth, colorspace conversion and resizing

[Jindadil007]

1. Don't copy the filters in system32.
2. As the error msg shows, you are missing another plugin: fmtc

Here's a nearly complete list of availible plugins: http://www.vapoursynth.com/doc/pluginlist.html
You need fmtconv – High quality bitdepth, colorspace conversion and resizing

Hi CK...Many Thanks :)... Looks I am just close... Now this error is showing up :

2017-01-09 19:40:43.605
Core freed but 4 filter instances still exist
Core freed but 4 filter instances still exist
Core freed but 1612800 bytes still allocated in framebuffers
Core freed but 1612800 bytes still allocated in framebuffers

What's this all about ?

Thanks

[Mystery Keeper]

Hi CK...Many Thanks :)... Looks I am just close... Now this error is showing up :

2017-01-09 19:40:43.605
Core freed but 4 filter instances still exist
Core freed but 4 filter instances still exist
Core freed but 1612800 bytes still allocated in framebuffers
Core freed but 1612800 bytes still allocated in framebuffers

What's this all about ?

ThanksIt is a leak caused by faulty plugins not freeing references on destruction. Please post your script.

[Jindadil007]

It is a leak caused by faulty plugins not freeing references on destruction. Please post your script.

Many Thanks MK...Resolved it by downloading nnedi3_weights :o

[Myrsloik]

Many Thanks MK...Resolved it by downloading nnedi3_weights :o

The leak happens in a few rare error paths on nnedi3. It will be fixed soon.

[Jindadil007]

The leak happens in a few rare error paths on nnedi3. It will be fixed soon.
That would be great...However while rendering a 23.976 fps clip with QTGMC with default command of preset and TFF, it gave an output which doubled the output fps, in Avisynth I used selecteven() or distributor() to keep original fps. How should it be used in python ? Thanks !!!

Sent from my GT-N8000 using Tapatalk

[VS_Fan]

From the python reference (http://www.vapoursynth.com/doc/pythonreference.html) at the vapoursynth website (silicing a clip (http://www.vapoursynth.com/doc/pythonreference.html#slicing-and-other-syntactic-sugar)):

Select even numbered frames:
video = clip[::2]

Select odd numbered frames:
video = clip[1::2]

Also, from the SelectEvery (http://www.vapoursynth.com/doc/functions/selectevery.html) function:

Return even numbered frames, starting with 0:
SelectEvery(clip=clip, cycle=2, offsets=0)

Return odd numbered frames, starting with 1:
SelectEvery(clip=clip, cycle=2, offsets=1)

Remember Search (https://www.google.com/search?q=selecteven+site%3Avapoursynth.com&ie=utf-8&oe=utf-8) is your fiend

[Jindadil007]

From the python reference (http://www.vapoursynth.com/doc/pythonreference.html) at the vapoursynth website (silicing a clip (http://www.vapoursynth.com/doc/pythonreference.html#slicing-and-other-syntactic-sugar)):

Select even numbered frames:
video = clip[::2]

Select odd numbered frames:
video = clip[1::2]

Also, from the SelectEvery (http://www.vapoursynth.com/doc/functions/selectevery.html) function:

Return even numbered frames, starting with 0:
SelectEvery(clip=clip, cycle=2, offsets=0)

Return odd numbered frames, starting with 1:
SelectEvery(clip=clip, cycle=2, offsets=1)

Remember Search (https://www.google.com/search?q=selecteven+site%3Avapoursynth.com&ie=utf-8&oe=utf-8) is your fiend
Hi Thanks... i read it but was confused with odd no. Or even no. Frames. What function i should use to keep intact 23.976 fps & 29.970 fps. It would be great if you explain this...


Sent from my GT-N8000 using Tapatalk

[Boulder]

As far as I know, QTGMC has the option to output single-rate video. The FPSDivisor parameter controls the output. There's also the option to use some motion blur.

[buchanan]

Hi,

I tried to run QTGMC with the latest version of havsfunc.py (9th november) from github and on Vapoursynth R35, but i get a "No attribute with the name focus2 exists" in relation to the line "return core.focus2.TemporalSoften2 etc"
What could be the problem ?

[Are_]

http://forum.doom9.org/showthread.php?t=166769 and on the second page there is a 64bit version.

[buchanan]

http://forum.doom9.org/showthread.php?t=166769 and on the second page there is a 64bit version.

I use the 64bit version and now that I loaded temporalsoften.dll and scenechange.dll from the x64 package, I get a similar error:
"No attribute with the name focus exists", related to core.focus.TemporalSoften :-S

[Are_]

For focus you need https://github.com/dubhater/vapoursynth-temporalsoften

[buchanan]

For focus you need https://github.com/dubhater/vapoursynth-temporalsoften

Thanks ! It's working now.
I don't remember that I needed those plugins with Vapoursynth R33.1, which I used until now.

[Jindadil007]

[QUOTE=HolyWu;1603132]https://github.com/HomeOfVapourSynthEvolution/havsfunc

Hi...Thanks for sharing this script. I have successfully render the file with QTGMC. However, I do not know how I can keep the original frame rate intact. After rendering, I am getting double frame rate instead of original.

Pl. guide how to fix this. In avisynth I used functions but its not working in python.

Thanks...:thanks:

[Jindadil007]

My preset for x264 encoding with Y4M header in VapourSynth Editor r13:--demuxer y4m
--pass 1
--bitrate 3000
--stats "{sd}/{sn}.stats"
--slow-firstpass
--min-keyint 1
--b-adapt 1
--ref 5
--bframes 4
--vbv-init 1.0
--ratetol 100.0
--qcomp 0
--cplxblur 0
--qblur 0
--rc-lookahead 250
--aq-mode 2
--merange 64
--me tesa
--subme 11
--partitions all
--trellis 2
--no-dct-decimate
--stitchable
--direct auto
--no-psy
--no-progress
-
-o "{sd}/{sn}.mkv"

Hi MK...this is 1 pass or 2 pass ?

[Mystery Keeper]

Hi MK...this is 1 pass or 2 pass ?This is a template for multipass.
"--pass 1" for the first pass.
Change it to "--pass 3" for subsequent passes or "--pass 2" for the final pass. I personally just use "--pass 3" for every subsequent pass even if I only do 2 passes. The difference is that "--pass 3" updates the stat files and "--pass 2" doesn't.

[Jindadil007]

This is a template for multipass.
"--pass 1" for the first pass.
Change it to "--pass 3" for subsequent passes or "--pass 2" for the final pass. I personally just use "--pass 3" for every subsequent pass even if I only do 2 passes. The difference is that "--pass 3" updates the stat files and "--pass 2" doesn't.

Many Thanks MK...Is this right...Bitrate can be changed based on target file size...

--demuxer y4m
--pass 3
--bitrate 3000
--stats "{sd}/{sn}.stats"
--slow-firstpass
--min-keyint 1
--b-adapt 1
--ref 5
--bframes 4
--vbv-init 1.0
--ratetol 100.0
--qcomp 0
--cplxblur 0
--qblur 0
--rc-lookahead 250
--aq-mode 2
--merange 64
--me tesa
--subme 11
--partitions all
--trellis 2
--no-dct-decimate
--stitchable
--direct auto
--no-psy
--no-progress
-
-o "{sd}/{sn}.mkv"

One more thing...is it incorrect to mention these command parameters ?

--level 4.1 --preset slower --deblock -1:-1 --qpmin 10 --qpmax 51 --vbv-bufsize 78125 --vbv-maxrate 62500

Thanks !!!

[Mystery Keeper]

Many Thanks MK...Is this right...Bitrate can be changed based on target file size...

--demuxer y4m
--pass 3
--bitrate 3000
--stats "{sd}/{sn}.stats"
--slow-firstpass
--min-keyint 1
--b-adapt 1
--ref 5
--bframes 4
--vbv-init 1.0
--ratetol 100.0
--qcomp 0
--cplxblur 0
--qblur 0
--rc-lookahead 250
--aq-mode 2
--merange 64
--me tesa
--subme 11
--partitions all
--trellis 2
--no-dct-decimate
--stitchable
--direct auto
--no-psy
--no-progress
-
-o "{sd}/{sn}.mkv"

One more thing...is it incorrect to mention these command parameters ?

--level 4.1 --preset slower --deblock -1:-1 --qpmin 10 --qpmax 51 --vbv-bufsize 78125 --vbv-maxrate 62500

Thanks !!!Sorry, but I'm not expert either. I made my preset based on how I understood the help.

[HolyWu]

Hi...Thanks for sharing this script. I have successfully render the file with QTGMC. However, I do not know how I can keep the original frame rate intact. After rendering, I am getting double frame rate instead of original.

Pl. guide how to fix this. In avisynth I used functions but its not working in python.

Thanks...:thanks:

By default QTGMC outputs video at "double-rate", twice the frame rate of the source. By setting FPSDivisor to 2, every second frame is dropped and the output frame rate is the same as the source ("single-rate").

[Jindadil007]

By default QTGMC outputs video at "double-rate", twice the frame rate of the source. By setting FPSDivisor to 2, every second frame is dropped and the output frame rate is the same as the source ("single-rate").
Hi...Thanks...would appreciate if you share the command example...i am a newbie to python...


Sent from my GT-N8000 using Tapatalk

[tObber166]

@Jindadil007:

import vapoursynth as vs
import havsfunc as haf
core = vs.get_core()
v = core.lsmas.LWLibavSource(source = r'F:\clip.mkv')
v = haf.QTGMC(v, Preset='Fast', TFF=True, FPSDivisor=2)
v.set_output()

Or:

v = haf.QTGMC(v, Preset='Fast', TFF=True)
v = v[::2]

[Jindadil007]

@Jindadil007:

import vapoursynth as vs
import havsfunc as haf
core = vs.get_core()
v = core.lsmas.LWLibavSource(source = r'F:\clip.mkv')
v = haf.QTGMC(v, Preset='Fast', TFF=True, FPSDivisor=2)
v.set_output()

Or:

v = haf.QTGMC(v, Preset='Fast', TFF=True)
v = v[::2]

Many Thanks tObber...It was really helpful:thanks:

[AMED]

Would it be possible to add in the prefilter 1 and prefilter 2 modes for SMDegrain please?

(prefilter==1) ? input8.MinBlur(1,Chr,planar)
(prefilter==2) ? input8.MinBlur(2,Chr,planar)

[HolyWu]

Would it be possible to add in the prefilter 1 and prefilter 2 modes for SMDegrain please?

I replace MinBlur with sbr for prefilter 1&2 on purpose. sbr is faster on 16-bit input, especially prefilter 2. You can still feed the prefiltered clip with your favorite denoiser to SMDegrain like this:

foo = haf.SMDegrain(foo, prefilter=haf.MinBlur(foo, 2))

[AMED]

My apologies, I didn't notice that it had been put in the else statement.

[KingLir]

Hi, I am having a problem with srestore. Maybe I just don't know how to use it properly. I am using latest VapourSynth and HAvsFunc.
I am running the following script on PAL DVD (tv animation) and it finish encoding on the first frame - but runs OK (on all frames) when I remove the srestore row.

clip = core.ffms2.Source(source=filename)
clip = haf.QTGMC(clip, Preset='Placebo', TFF=True)
clip = haf.srestore(source=clip)
clip.set_output()

Anyone have an idea ?

[Selur]

Using Interframe:
# Imports
import os
import sys
import vapoursynth as vs
core = vs.get_core()
# Import scripts folder
scriptPath = 'G:/Hybrid/vsscripts'
sys.path.append(os.path.abspath(scriptPath))
# Loading Plugins
core.std.LoadPlugin(path="G:/Hybrid/vsfilters/FrameFilter/Interframe/svpflow2_vs64.dll")
core.std.LoadPlugin(path="G:/Hybrid/vsfilters/FrameFilter/Interframe/svpflow1_vs64.dll")
core.std.LoadPlugin(path="G:/Hybrid/vsfilters/SourceFilter/LSmashSource/vslsmashsource.dll")
# Import scripts
import havsfunc as havsfunc
# Loading F:\TestClips&Co\test.avi using LWLibavSource
clip = core.lsmas.LWLibavSource(source="F:/TESTCL~1/test.avi", format="YUV420P8", cache=0)
# making sure input color matrix is set as 470bg
clip = core.resize.Point(clip, matrix_in_s="470bg")
# making sure input color range is set to TV (16-235) scale.
clip = core.std.SetFrameProp(clip=clip, prop="_ColorRange", intval=1)
# adjusting frame count with Interframe/SVP
clip = havsfunc.InterFrame(clip, Preset="fastest", NewNum=60, NewDen=1, GPU=True)
# Output
clip.set_output()
I get: "SVSmoothFps: unable to init GPU-based renderer - code 255" any clue what I'm missing? (Got an NVIDIA Geforce 980 Ti with driver version 381.65)
(Interframe works fine when using through Avisynth)

Cu Selur

[HolyWu]

Hi, I am having a problem with srestore. Maybe I just don't know how to use it properly. I am using latest VapourSynth and HAvsFunc.
I am running the following script on PAL DVD (tv animation) and it finish encoding on the first frame - but runs OK (on all frames) when I remove the srestore row.

clip = core.ffms2.Source(source=filename)
clip = haf.QTGMC(clip, Preset='Placebo', TFF=True)
clip = haf.srestore(source=clip)
clip.set_output()

Anyone have an idea ?

It works fine here. Can you preview your script in VapourSynth Editor?

[HolyWu]

I get: "SVSmoothFps: unable to init GPU-based renderer - code 255" any clue what I'm missing? (Got an NVIDIA Geforce 980 Ti with driver version 381.65)
(Interframe works fine when using through Avisynth)

Cu Selur

I just tried it on my GTX 660 with driver 378.78 and it works fine. Since it's an initialization error in the filter itself, I guess you need to report the issue to the SVP team rather than here.

[Selur]

@HolyWu: Thank for testing. Will do. -> https://www.svp-team.com/forum/viewtopic.php?pid=64820

[KingLir]

Can you preview your script in VapourSynth Editor?

Thank you for checking. I don't know how to use VapourSynth Editor - can you instruct me on macOS ?

Maybe you can just figure it out by looking at the mkv ? I tested with other mkvs (from NTSC DVDs) - and the issue didn't happen.

As my script shows, I am trying to run srestore after QTGMC x2 (50 fps) and if I just comment out the srestore row - it runs OK.

I test encoding with:
vspipe --y4m --arg "filename=input.mkv" deinterlacer_QTGMC_PAL_script.vpy - | ffmpeg -thread_queue_size 1024 -i pipe: -map 0:0 -aspect 720:540 -c:v h264 -preset ultrafast -crf 0 input1.mkv

And getting:
Stream #0:0 -> #0:0 (rawvideo (native) -> h264 (libx264))
Finishing stream 0:0 without any data written to it.

Here is download link for a small test (that still reproduce the issue):
https://mega.nz/#!4tZxmYxD!lvoe7etj_Oelx_P8_xl7m_u6-PKwJ1ryoaadAuA6NCA

And here is it's mediainfo:
https://pastebin.com/LfYr6gpr

[Selur]

SVSmoothFps: unable to init GPU-based renderer - code 255
Was fixed by reinstalling the drivers,..

[Pat357]

I get: "SVSmoothFps: unable to init GPU-based renderer - code 255" any clue what I'm missing? (Got an NVIDIA Geforce 980 Ti with driver version 381.65)
(Interframe works fine when using through Avisynth)
Cu Selur
I suspect that you use the SVP 4.x DDL's, aren't you ?
They have a different syntax then the older DLL's from the Interframe project.

This works on my system :


import vapoursynth as vs
core = vs.get_core(threads=8)
core.std.LoadPlugin(r"c:\Program Files (x86)\VapourSynth\filters\svpflow-4.0.0.128\x64\svpflow1_vs64.dll")
core.std.LoadPlugin(r"c:\Program Files (x86)\VapourSynth\filters\svpflow-4.0.0.128\x64\svpflow2_vs64.dll")

clip = core.ffms2.Source(r"k:\film\Beyond.2014.1080p.BluRay.H264.AAC-RARBG\Beyond.2014.1080p.BluRay.H264.AAC-RARBG.mp4")

super_params="{pel:2,gpu:1}"
analyse_params="{}"
smoothfps_params="{rate:{num:5,den:2}}"

super = core.svp1.Super(clip,super_params)
vectors= core.svp1.Analyse(super["clip"],super["data"],clip,analyse_params)
smooth = core.svp2.SmoothFps(clip,super["clip"],super["data"],vectors["clip"],vectors["data"],smoothfps_params)

# smooth = core.std.AssumeFPS(smooth,fpsnum=smooth.fps_num,fpsden=smooth.fps_den)
smooth.set_output()

[Selur]

I suspect that you use the SVP 4.x DDL's, aren't you ?
I used the latest plugins download from their homepage: (http://www.svp-team.com/files/gpl/svpflow-4.0.0.128.zip)
and like I post reinstalling the drivers fixed the issue for me, no need to change the script.

[HolyWu]

I test encoding with:
vspipe --y4m --arg "filename=input.mkv" deinterlacer_QTGMC_PAL_script.vpy - | ffmpeg -thread_queue_size 1024 -i pipe: -map 0:0 -aspect 720:540 -c:v h264 -preset ultrafast -crf 0 input1.mkv

And getting:
Stream #0:0 -> #0:0 (rawvideo (native) -> h264 (libx264))
Finishing stream 0:0 without any data written to it.

Here is download link for a small test (that still reproduce the issue):
https://mega.nz/#!4tZxmYxD!lvoe7etj_Oelx_P8_xl7m_u6-PKwJ1ryoaadAuA6NCA

And here is it's mediainfo:
https://pastebin.com/LfYr6gpr

The core resize filter complains about unrecognized transfer characteristics and unrecognized color primaries with your mkv, so the script can't be evaluated at all. I'm not sure why _Transfer=5 and _Primaries=5 are not accepted. Anyway, I update srestore to only do format conversion when necessary. Grab the latest raw file from GitHub then.

[ChaosKing]

I just downloaded the newest version of your script.

But I get this error:
567: fmask = core.misc.Hysteresis(core.std.Median(sobelm, planes=[0]), sobelm, planes=[0])
AttributeError: There is no function named Hysteresis

I works when I change misc to generic. I'm using genericfilters from R32.
Is there a newer version available somewhere?

[HolyWu]

But I get this error:
567: fmask = core.misc.Hysteresis(core.std.Median(sobelm, planes=[0]), sobelm, planes=[0])
AttributeError: There is no function named Hysteresis

I works when I change misc to generic. I'm using genericfilters from R32.
Is there a newer version available somewhere?

There is something wrong in your VS installation. It loads a misc plugin that doesn't have the Hysteresis filter, which is not possible as it's already included since R36.

[ChaosKing]

hmm I re-downloaded the R37 coreplugins and removed all dlls from the plugins folder. Only left ffms2.
Also tried a fresh R37 x64 portable. Same error :/

Genericfilters is not a seperate plugin (dll) anymore, correct?

[HolyWu]

hmm I re-downloaded the R37 coreplugins and removed all dlls from the plugins folder. Only left ffms2.
Also tried a fresh R37 x64 portable. Same error :/

Genericfilters is not a seperate plugin (dll) anymore, correct?

Search for filename "miscfilter" in your entire hard drive and see what you get.

GenericFilters is obsolete and not included in the windows installer anymore.

[ChaosKing]

Found nothing. But I'm only using the portable version. So maybe it is not included?

[TheFluff]

MiscFilters.dll appears to be missing from the portable .7z, yes (it should be in the coreplugins folder).

[KingLir]

The core resize filter complains about unrecognized transfer characteristics and unrecognized color primaries with your mkv, so the script can't be evaluated at all. I'm not sure why _Transfer=5 and _Primaries=5 are not accepted. Anyway, I update srestore to only do format conversion when necessary. Grab the latest raw file from GitHub then.

Thank you! Seems to work well.