New tool - VapourBox

[aramkolt]

This is what I got - seems maybe that duplicate thing is the problem? Which should I delete or rename?

Still have the same half opaque no preview available and no thumbnails in 0.5.2

[flashandpan007]

0.3.0 release is out. It now handles 4:2:1 video sources (eg. NTSC DV footage) and up-converts to 4:2:2 so there's no loss of quality, PLUS the tool now supports queueing and batching of videos for processing.

4:1:1 not 4:2:1 right?

[tvd051955]

New user here, been testing for a couple of days.

I have found the software encode options to just be too slow (as I expected), so tried a couple of tests with the nvenc encoding.

Unfortunately, I am having very poor image quality and very small file sizes (just 20% of the size of a software encoded file) when using the nvenc option to encode files.

I'm using preset 7.

Bitrates are very low (1792 Kbs), Macro blocking is present on dark areas of the image, and fine detail is lacking.
Video card is an NVidia RTX 460.

While ffmpeg lists a number of options for the nvenc encoder including the following:

-preset p7 (or slow for maximum quality, though p7 is often optimal for quality-to-speed balance).
-tune hq to optimize for high quality.
-rc vbr (Variable Bitrate) with -cq 24 to 26 for constant quality mode (lower values = higher quality).
-b:v 0 to disable bitrate limiting, allowing the encoder to use optimal bitrate based on content.
-profile:v high and -level 4.1 for broad compatibility and quality.
-rc-lookahead 32 to improve rate control by analyzing future frames.
-bf 3 and -b_ref_mode middle for better B-frame utilization.
-refs 16 to improve prediction accuracy.
-spatial-aq 1 and -aq-strength 15 to enhance adaptive quantization.

the only ones I seem to have access to through the GUI are Preset and something called Quality which doesn't match any of the ffmpeg parameters.

I couldn't find any of the other ffmpeg settings visible through the GUI (or am simply missing them?), so could you either add them to the GUI, or tell me where I can hard code them to be the defaults for nvenc encodes?

MediaInfo results follow

H264

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference frames : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 7 min 7 s
Bit rate : 9 920 kb/s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 29.970 (30000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0 (Type 1)
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.160
Stream size : 506 MiB (96%)
Writing library : x264 core 165
Encoding settings : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x3:0x113 / me=hex / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=12 / lookahead_threads=2 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=15.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
Default : No
Forced : No
Color range : Limited

NVENC

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : Main@L4
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference frames : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 7 min 7 s
Bit rate mode : Variable
Bit rate : 2 105 kb/s
Maximum bit rate : 16.0 Mb/s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 29.970 (30000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.034
Stream size : 107 MiB (90%)
Writing library : Lavc62.23.102 h264_nvenc
Default : No
Forced : No
Color range : Limited


I have 2 systems, one is running Win 10 LTSC and the other Win 11.


Thanks for a great, easy to use tool.

[StuartCameronCode]

[Attachment 91219 - Click to enlarge]

This is what I got - seems maybe that duplicate thing is the problem? Which should I delete or rename?

Still have the same half opaque no preview available and no thumbnails in 0.5.2

Yeah, I'd just suggest deleting your ~/Library/Application Support/VapourBox folder and re-run the application, it'll re-download the dependencies. I've got no idea how it got into that state on your machine. There aren't too many MacOS users seemingly, it'd be interesting to know if anyone else runs into this issue.

[StuartCameronCode]

0.3.0 release is out. It now handles 4:2:1 video sources (eg. NTSC DV footage) and up-converts to 4:2:2 so there's no loss of quality, PLUS the tool now supports queueing and batching of videos for processing.

4:1:1 not 4:2:1 right?

Yep, typo .

[tvd051955]

Sorry for the dup, please excuse.....

[aramkolt]

[Attachment 91219 - Click to enlarge]

This is what I got - seems maybe that duplicate thing is the problem? Which should I delete or rename?

Still have the same half opaque no preview available and no thumbnails in 0.5.2

Yeah, I'd just suggest deleting your ~/Library/Application Support/VapourBox folder and re-run the application, it'll re-download the dependencies. I've got no idea how it got into that state on your machine. There aren't too many MacOS users seemingly, it'd be interesting to know if anyone else runs into this issue.

Doing that just creates the same ~/Library/Application Support/VapourBox folder when the dependencies download. Should I just delete the one that's actually within the app package contents instead maybe?

[aramkolt]

Actually, looking at it, the duplicate files appear to be within the app package itself:

/Applications/VapourBox.app/Contents/Frameworks/rhttp.framework/Versions/A/rhttp
/Applications/VapourBox.app/Contents/Frameworks/rhttp.framework/rhttp





Deleting any of those seems to make the app think it isn't compatible with the current OS and just doesn't allow the app to launch, but also not sure what I could try renaming them to?

[tvd051955]

I just found another case with a source media format that delivers unexpected (incorrect) results:

DVD IVTC sources are being rendered at the incorrect frame rate. IE, a 23.976 FPS telecine'd (3-2 pulldown applied) source file is actually being rendered at 29.970 FPS rather than at a proper 23.976 FPS using IVTC.

Notice that the running time on the processed file is also incorrect: 48 minutes vs 39 minutes.

Here is the MediaInfo source media data:

Video
ID : 1
ID in the original source medium : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings : BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Default
Format settings, GOP : M=3, N=12
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 48 min 8 s
Bit rate mode : Variable
Bit rate : 3 993 kb/s
Maximum bit rate : 7 500 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.482
Time code of first frame : 09:59:59:00
Time code source : Group of pictures header
Stream size : 1.34 GiB (95%)
Language : English
Default : No
Forced : No
Original source medium : DVD-Video

And the here is the MediaInfo target media data:

ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference frames : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 38 min 15 s
Bit rate : 24.6 Mb/s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 29.970 (30000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0 (Type 1)
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.396
Stream size : 6.58 GiB (78%)
Writing library : x264 core 165
Encoding settings : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x3:0x113 / me=hex / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=12 / lookahead_threads=2 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=250 / keyint_min=25 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=10.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
Default : No
Forced : No
Color range : Limited

This a very common format on DVD and it needs to be handled correctly for conversion to a BD.

Thanks in advance for taking a look at this.

[HemLok]

New user here, been testing for a couple of days.

I have found the software encode options to just be too slow (as I expected), so tried a couple of tests with the nvenc encoding.

Unfortunately, I am having very poor image quality and very small file sizes (just 20% of the size of a software encoded file) when using the nvenc option to encode files.

You aren't alone. I noticed but hadn't gotten around to posting. IMO the hevc_nvenc encodes via VapourBox are poorer quality than they should be and I don't know why. I did a test where I only used VapourBox to convert a file from H264 to HEVC at p7. I then used the same ffmpeg from the deps folder and did a conversion using hevc_nvenc with p7 from the command line. The VapourBox output was like half the bitrate and, obviously, undersized.

Note: Personally, I normally use rigaya's NVEnc for hardware encoding and am very happy with it so my experience with ffmpeg's hardware encoding is lacking. I love VapourBox's features and usability. The software encoding is great but time consuming. I had hoped the hardware encoding would be good enough even if it didn't match the quality. For now I'm sticking with rigaya's for straight encoding and VapourBox with software encoding if I need to denoise, sharpen, upconvert, etc.

[tvd051955]

Thanks for the info.
Personally, I think Vaporbox has outstanding potential, and is really good for a brand new product.

It's not surprising that there are a few rough spots that need polishing.

[HemLok]

Thanks for the info.
Personally, I think Vaporbox has outstanding potential, and is really good for a brand new product.

It's not surprising that there are a few rough spots that need polishing.

I agree completely!

[StuartCameronCode]

Hi everyone, you can stop yer whinging , I've just published a new 0.5.3 release that I'm hoping addresses most of the issues you're all describing. There's still the frame rate issue I'll need to look into.

Changes

- Fixed bugs and added more options to compression codecs, including ffmpeg overrides
- Add zoom & pan support to the preview window
- Disable bilinear filtering on split preview for pixel-accurate zoom
- Fixed progress tracking of processing jobs (this was a deep rabbit hole!)
- DMG based distribution for MacOS builds, fixes symlink issues with rhttp library

You can find the new build here: https://github.com/StuartCameronCode/VapourBox/releases

[tvd051955]

Looking good.

Testing IVTC now.

However, another minor issue: looks like the resize kernel is not being properly saved / restored in the preset. I specified spline 36, saved the preset and when I loaded the preset, the value was set back to the default. The saved preset file actually contains "kernel":"lanczos".

[tvd051955]

IVTC is not working properly.

Frame rate of processed file is incorrect (14.985 FPS) Running time is incorrect.

IVTC is a totally different process from deinterlacing. The source frames are progressive @ 23.976 FPS. The telecine process handles the first two frames as progressive, but splits every third frame into 2 interlaced fields. A standard deinterlace algorithm will not work.

Here is what I use to IVTC Telecined media when using Vapoursynth:

video = core.d2v.Source(input=r'.\title00_Track01.m2v.d2v' ) # Use the name of the m2v.d2v file
video = core.std.AssumeFPS(video, fpsnum=30000, fpsden=1001)
video = core.vivtc.VFM(video, order=1, cthresh=10)
video = core.vivtc.VDecimate(video)
HDvideo = video.resize.Lanczos(width=1920, height=1080, filter_param_a=3)
HDvideo = core.std.CropRel(HDvideo, top=0, bottom=0, left=0, right=0)
HDvideo = finesharp.sharpen(HDvideo, 2, sstr=2.0, xstr=0.2)
HDvideo.set_output()

[StuartCameronCode]

IVTC is not working properly.

Yep, it's the one main thing I hadn't addressed in the previous release. And guess what? There's a fresh new 0.6.0 release ready for you that introduces true IVTC inverse telecine support via the VIVTC vapoursynth plugin. I've done some work cleaning up all the control parameters in the UI tool so once you select "deinterlace" you've got a choice of algorithm / plugin to use.

https://github.com/StuartCameronCode/VapourBox/releases

Enjoy .

[HemLok]

Hi everyone, you can stop yer whinging , I've just published a new 0.5.3 release that I'm hoping addresses most of the issues you're all describing...

Wow. Thank you! The hardware encoding so much improved from some quick testing and I can now tweak it even more with custom arguments.

Edit: I did my testing with 0.6.0 and the new deps.

[aramkolt]

For me, it still doesn't update the preview window (the semitransparent "no preview available" shows up on the left) or give thumbnails still with 0.6.0 on macOS Tahoe despite new dependencies also downloaded.

The preview window does seem to work for cropping effects (it updates the whole window instead of just the right pane of the slider like I think it would work?).

A sort of feature request would also be when selecting whether to double interlaced to progressive or not when using QTGMC, it would be nice if it showed the identified field rate instead of always 50i->50p. I thought it had misidentified my frame rate, but when I actually ran it, it produced the correct 59.94p with an NTSC source, so I think it must be just saying 50i->50p as an "example" as opposed to what the input file actually is.

Definitely appreciate you putting this tool out there, it just isn't quite optimized for Mac just yet, or perhaps there's something else going on with my system that it still doesn't like, but not sure what it could be.

[StuartCameronCode]

For me, it still doesn't update the preview window (the semitransparent "no preview available" shows up on the left) or give thumbnails still with 0.6.0 on macOS Tahoe despite new dependencies also downloaded.

I tried this on another Mac that hasn't had any VapourBox development on it, so it was a fresh install on a machine running MacOS 26.2 and it worked just fine. I'm officially confused! When I get a chance I'll try running another MacOS instance in a VM with a fresh clean OS install and see how that goes.

[tvd051955]

IVTC is still not working properly. Running 0-6-0. Source running time is 47:44 and the target running time is 38:11.

Source MediaInfo;

Video
ID : 1
ID in the original source medium : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings : BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Default
Format settings, GOP : M=3, N=12
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 47 min 44 s
Bit rate mode : Variable
Bit rate : 3 997 kb/s
Maximum bit rate : 7 500 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.482
Time code of first frame : 09:59:59:00
Time code source : Group of pictures header
Stream size : 1.33 GiB (95%)
Language : English
Default : No
Forced : No
Original source medium : DVD-Video

Target MediaInfo:

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L4
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference frames : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 38 min 11 s
Bit rate : 7 906 kb/s
Width : 1 920 pixels
Height : 1 080 pixels
Display aspect ratio : 16:9
Frame rate mode : Constant
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0 (Type 1)
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.159
Stream size : 2.11 GiB (77%)
Writing library : x264 core 165
Encoding settings : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x3:0x113 / me=hex / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=12 / lookahead_threads=2 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=250 / keyint_min=23 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=18.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
Default : No
Forced : No
Color range : Limited

[tvd051955]

I suspect this may be the problem:

clip = core.vivtc.VDecimate(clip,

[StuartCameronCode]

For me, it still doesn't update the preview window (the semitransparent "no preview available" shows up on the left) or give thumbnails still with 0.6.0 on macOS Tahoe despite new dependencies also downloaded.

I tried this on another Mac that hasn't had any VapourBox development on it, so it was a fresh install on a machine running MacOS 26.2 and it worked just fine. I'm officially confused! When I get a chance I'll try running another MacOS instance in a VM with a fresh clean OS install and see how that goes.

Ok, I did a fresh install of MacOS 26.3 into a VM, did nothing but install VapourBox 0.6.0, it downloaded the dependencies and can open input files, generates thumbnails and previews generate. Plus it processes videos properly. Sorry, I think there's something specific to your machine, I'm not sure where to start looking. When the previews fail to generate, have you tried clicking the log that appears down the bottom left? Can you send that to me?

[StuartCameronCode]

IVTC is still not working properly. Running 0-6-0. Source running time is 47:44 and the target running time is 38:11.

Thanks, my test video seemed to look OK, but I'll take another look when I get a chance and compare frame counts / playback duration.

[StuartCameronCode]

IVTC is still not working properly. Running 0-6-0. Source running time is 47:44 and the target running time is 38:11.

Source MediaInfo;

Video

...
Duration : 47 min 44 s
Bit rate mode : Variable
Bit rate : 3 997 kb/s
Maximum bit rate : 7 500 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
...
Original source medium : DVD-Video

Ah, actually - shouldn't your source DVD have been 29.97fps, not 23.976? I don't deal with a lot of NTSC DVDs. To my understanding that'd be a soft telecine DVD, meaning it's already effectively progressive, just telecined to 29.97 for playback from the DVD player, but the source is at film fps progressive. Am I missing something?

[tvd051955]

No, this says it all (from running Mediainfo on the MKV file generated from the DVD MakeMKV)

Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown

https://en.wikipedia.org/wiki/Three-two_pull_down

I tried watching the output file from Vapourbox and it looked like an old silent movie with everything speeded up.

[tvd051955]

Here's the Vapoursynth script I have been using for DVD for many years:

import vapoursynth as vs
import finesharp as fs
core = vs.get_core()
video = core.d2v.Source(input=r'.\title_t00_Track01.m2v.d2 v')
video = core.std.AssumeFPS(video, fpsnum=30000, fpsden=1001)
video = core.vivtc.VFM(video, order=1, cthresh=10)
video = core.vivtc.VDecimate(video)
HDvideo = video.resize.Lanczos(width=1920, height=1080, filter_param_a=3)
HDvideo = core.std.CropRel(HDvideo, top=0, bottom=0, left=0, right=0)
HDvideo = fs.sharpen(HDvideo)
HDvideo.set_output()

Note that all DVD's are encoded at 30 FPS as DVD does not natively support 24 FPS.

[tvd051955]

Here's the difference between your script

# IVTC: Inverse Telecine (VFM field matching + VDecimate)
clip = core.vivtc.VFM(clip, order=0,)
clip = core.vivtc.VDecimate(clip,)


and my script

video = core.std.AssumeFPS(video, fpsnum=30000, fpsden=1001)
video = core.vivtc.VFM(video, order=1, cthresh=10)
video = core.vivtc.VDecimate(video)

Since Mediainfo is returning

Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown

I suspect the VDecimate is removing valid frames resulting in the shorter run time.

[tvd051955]

OK, more info:
I modified your template to:

{{/DEINT_QTGMC}}
{{#DEINT_IVTC}}
# IVTC: Inverse Telecine (VFM field matching + VDecimate)
IVTC: Inverse Telecine (VFM field matching + VDecimate)
clip = core.std.AssumeFPS(clip, fpsnum=30000, fpsden=1001)
clip = core.vivtc.VFM(clip, order=1, cthresh=10
)

and no change. I ran a plain IVTC process with no scaling for the test.

I can't find where you are writing the MKV, but somehow the video clip duration is being changed somewhere in the process while the MKV duration remains correct. Both the MKV and the video clip frame rates remain correct.

The MKV data from the target file:

General
Unique ID : 302143673945228067893244814201141147054 (0xE34EC14229EC056AA58065703EB135AE)
Complete name : F:\MKV\Season 6\B1_t00 (1)-001_processed.mkv
Format : Matroska
Format version : Version 4
File size : 73.1 MiB
Duration : 5 min 0 s
Overall bit rate : 2 043 kb/s
Frame rate : 23.976 FPS
Writing application : Lavf62.8.102
Writing library : Lavf62.8.102
ErrorDetectionType : Per level 1

Video
ID : 1
Format : AVC
Format/Info : Advanced Video Codec
Format profile : High@L3
Format settings : CABAC / 4 Ref Frames
Format settings, CABAC : Yes
Format settings, Reference frames : 4 frames
Codec ID : V_MPEG4/ISO/AVC
Duration : 4 min 0 s
Bit rate : 1 810 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 3:2
Frame rate mode : Constant
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0 (Type 1)
Bit depth : 8 bits
Scan type : Progressive
Bits/(Pixel*Frame) : 0.218
Stream size : 51.8 MiB (71%)
Writing library : x264 core 165
Encoding settings : cabac=1 / ref=3 / deblock=1:0:0 / analyse=0x3:0x113 / me=hex / subme=7 / psy=1 / psy_rd=1.00:0.00 / mixed_ref=1 / me_range=16 / chroma_me=1 / trellis=1 / 8x8dct=1 / cqm=0 / deadzone=21,11 / fast_pskip=1 / chroma_qp_offset=-2 / threads=12 / lookahead_threads=2 / sliced_threads=0 / nr=0 / decimate=1 / interlaced=0 / bluray_compat=0 / constrained_intra=0 / bframes=3 / b_pyramid=2 / b_adapt=1 / b_bias=0 / direct=1 / weightb=1 / open_gop=0 / weightp=2 / keyint=250 / keyint_min=23 / scenecut=40 / intra_refresh=0 / rc_lookahead=40 / rc=crf / mbtree=1 / crf=18.0 / qcomp=0.60 / qpmin=0 / qpmax=69 / qpstep=4 / ip_ratio=1.40 / aq=1:1.00
Default : No
Forced : No
Color range : Limited


The MKV data from the source file:

General
Unique ID : 223606134092021486945389122429003431127 (0xA838F995A514C73F7D7742363BC6FCD7)
Complete name : F:\MKV\Season 6\B1_t00 (1)-001.mkv
Format : Matroska
Format version : Version 4
File size : 153 MiB
Duration : 5 min 0 s
Overall bit rate mode : Variable
Overall bit rate : 4 271 kb/s
Frame rate : 23.976 FPS
Encoded date : 2026-02-26 15:00:45 UTC
Writing application : mkvmerge 89.0 ('And the Melody Still Lingers On (Night in Tunisia)') 64-bit
Writing library : libebml v1.4.5 + libmatroska v1.7.1

Video
ID : 1
ID in the original source medium : 224 (0xE0)
Format : MPEG Video
Format version : Version 2
Format profile : Main@Main
Format settings : BVOP
Format settings, BVOP : Yes
Format settings, Matrix : Default
Format settings, GOP : M=3, N=12
Codec ID : V_MPEG2
Codec ID/Info : MPEG 1 or 2 Video
Duration : 5 min 0 s
Bit rate mode : Variable
Bit rate : 4 076 kb/s
Maximum bit rate : 7 500 kb/s
Width : 720 pixels
Height : 480 pixels
Display aspect ratio : 16:9
Frame rate mode : Variable
Frame rate : 23.976 (24000/1001) FPS
Color space : YUV
Chroma subsampling : 4:2:0
Bit depth : 8 bits
Scan type : Progressive
Scan order : 2:3 Pulldown
Compression mode : Lossy
Bits/(Pixel*Frame) : 0.492
Time code of first frame : 09:59:59:00
Time code source : Group of pictures header
Stream size : 146 MiB (95%)
Language : English
Default : No
Forced : No
Original source medium : DVD-Video

I hope this helps (and hopefully you have at least one NTSC DVD to test with).

[tvd051955]

OK this looks like a bug with ffdshow that apparently affects BestSource.

You'll need to use L-Smash-Works to read the source file and that's the reason I never have seen this issue before.

https://github.com/HomeOfAviSynthPlusEvolution/L-SMASH-Works

In the end, doing a src = core.bs.VideoSource(source='F:\MKV\Season 6\B1_t00_sample.mkv') and previewing the result of a 5 minute clip shows an incorrect duration of 4 minutes!!!!

and doing a src = core.lsmas.LWLibavSource(source='F:\MKV\Season 6\B1_t00_sample.mkv') and previewing the result of a 5 minute clip shows a duration of 5 minutes!

This issue showed up before doing any IVTC processing.

It even shows up on the Vapoursynth preview page time line.

Wow, I can't believe I just spent many many hours finding this, including having to rebuild my Python / Vapoursynth environments on my system as they were very much out of date.

But it was bugging heck out of me....

[tvd051955]

After more research looking for a possible fix, I guess that's what comes from relying on a codec (ffdshow) whose last update was in 2014. And it seems that Bestsource should probably be renamed after depending on an outdated codec.....