GPAC MP4Box Information, Feature Requests & Discussions [No bug reports]

[dimzon]

Just noticed the gpac updated yesterday.
@dimzon, did the following CVS update solved your problem?

Yeah! Seems like problem fixed (i have prodice a little quick tests)

Great work!

[leowai]

Yeah! Seems like problem fixed (i have prodice a little quick tests)Good to hear that. :)
Great work!Indeed, all credits should go to the code writer (author of the mp4box program in GPAC project), jeanlf. ;)

Cheers. :cool:

[SeeMoreDigital]

Does anybody have anything new to report?


Cheers

[yaz]

a new build is up on cd's site
Latest CVS Changes 20/09/05
- changed module naming - all modules are now prefixed with "gm_", and module is loaded/unloaded at run-time
- changed isomedia file open API for better support of temporary directories
- simplified 2D blitter for non DirectX output: no more in-middle surface used, direct stretch, blit and yuv conversion to back buffer
- added software and hardware support for MaterialKey in 2D renderer
- added all project files for evc4
- added support for OpenGL-ES in 3D renderer, in DX and in GAPI plugins
* tested with Hybrid Graphics, Vincent3D. Suggested library is Hybrid implementation
* Klimt could compile but no decent results
* Result only tested on PocketPC with Win PPC2003
- added support for GDI drawing on WinCE to avoid weird menu behaviour in windowed mode
- added support for OpenGL drawing on WinCE to avoid weird menu behaviour in windowed mode
- added osmophone to gpac/applications, demo player for Smartphone devices (tested on SPV C500)
- fixed redraw bug in 2D renderer for simple video drawing (background was always repaint regardless of video transparency)
- added LASeR XML to SVG loader
- fixed UTF8 vs Win-CP text handling (BT and XMT parsers, subs->bifs and subs->ttxt converters)
- improved SRT -> 3GPP convertion - now accepts any number of <b>, <i>, <u> tags at random places. Also cleaned up TTXT->SRT conversion.
- fixed bug in unknown stsd boxes handling
- attempt to fixe gpac compilation on 64 bits platforms

the bests
y

[dimzon]

Is there any ability to change track names in MP4?
Haali Splitter reports "ISO GPAC Video Handler" if file was muxed by MP4BOX but there are another string if file produced by Nero Recode

[JoeBG]

Is there any ability to change track names in MP4?
Haali Splitter reports "ISO GPAC Video Handler" if file was muxed by MP4BOX but there are another string if file produced by Nero Recode

You can do this with dumpster (ftp://ftp.apple.com/developer/Quicktime/Windows_Tools/Programmers_Tools/Dumpster/Dumpster.zip) r. Here is a link to a useful threat (http://forum.doom9.org/showthread.php?t=91879) about dumpster. I use it for every video.

[Doom9]

Is there any ability to change track names in MP4?You could reimport into another mp4 and using the language tags mp4box offers..

[JoeBG]

How dumpster works:

Open the file with right mouseclick\open with dumpster.
The tracks are in "moov"
Open a track, go to mdia/hdlr and you will see GPAC ISO MEDIA HDLR
You can change this with:
Main Video AVC: $4D61696E 20566964 656F2041 56430000 000000
Forced Sub TTXT: $466F7263 65642053 75622054 54585400 000000
Sub TTXT: $53756220 54545854 00000000 00000000 000000
Englisch TTXT:$456E676C 69736368 20545458 54000000 00000000 000000
Englisch 5.1 AAC: $456E676C 69736368 20352E31 20414143 00000000 000000

to change the language in the header information go to trak/mdia/mdhd
For a trcklist look here: http://forum.doom9.org/showthread.php?t=91879

[yakima]

i've noticed that mp4box uses strange timescale values when importing some video tracks, i.e. 24999 for 25 fps or 24000 for 23,976, resulting in quite some scew.
this happens only with some (avi)-files, and only since 0.3.0 (0.2.4dev works fine). -nodrop and -fps have no effect on this & fps is set correctly in the source avi.
did anyone else notice this? could it be some rounding error? afaik, there is no option to set the timescale, is there?

EDIT: after some more tests, i've found that 0.2.4 doesn't always work either; the source files are quite old and a bit messed up. it seems within one of the avis, strh gives 24.9998 while avih says 25 fps, resulting in a mp4 with 25 fps and a timescale of 24999 for all versions of mp4box. however, for the other file that gives a strange timescale, both avi headers say 23.976 and in this case, version 0.2.4 gives the correct timescale of 23976.
i wonder why mp4box doesn't take fps as reference for choosing a timescale ...

[dimzon]

a new build is up on cd's site
Can't find it!

[yaz]

Can't find it!hmm ... i can :) try this (http://www.aziendeassociate.it/./cd//gpac/dev/MP4Box.7z) !
the bests
y

[SeeMoreDigital]

Can't find it!There's a link to Celtic Druid's development builds on page one of this thread ;)

[SeeMoreDigital]

A few days ago Jean Le Feuvre was asked the following: -Date: 2005-09-22 06:28

May I ask if there has been any headway with VOBsub support?To which Jean replied: -Date: 2005-09-27 00:38
Sender: jeanlf

Not really, I tried to find some info on the Decoder Specific Info used in such streams, without success. The rest shouldn't be too hard to implement, but without the DSI syntax I cannot do anything.

Anyone having the slightest clue what's in there is more than welcome.So it would seem we've reached a brick wall for the time being.... Me thinks the next stop will be asking... the Ateme guys!


Cheers

[bond]

strange, maybe he would need more sample files?

[SeeMoreDigital]

strange, maybe he would need more sample files?Well, if he requires longer "NeroDigital" samples generated using Recode2, I can send him some on disc (or upload them) I guess....


Cheers

[SeeMoreDigital]

Jean Le Feuvre has provided us with a little more information with regard to the "VOBsub" DecoderSpecificInfo issue and MP4Box Date: 2005-10-13 05:20
Sender: jeanlf
Logged In: YES
user_id=815447

If you look at such a VobSub track, there is some decoder intialization data (which can be seen with MP4Box -nhnt). This is what is called in MPEG-4 the DecoderSpecificInfo or "DSI" for short.

Its syntax is codec dependent, and in this case I have no idea what has been placed there, and thus cannot generate it at import time...Can anybody help Jean out?


Cheers

[Kurtnoise]

For those who are interessed by the last mp4box improvements (http://forum.doom9.org/showthread.php?p=728093#post728093) and defunkt hack (http://forum.doom9.org/showthread.php?p=721387#post721387), here is (http://kurtnoise.free.fr/mp4tools/MP4Box_20051026+Defunkt_hack.zip) a build...

Otherwhise the latest mp4box : here (http://kurtnoise.free.fr/mp4tools/MP4Box_20051026.zip) or there (http://www.aziendeassociate.it/cd.asp?dir=/gpac/dev)...:)

[vsv]

Java player inside mp4 container is possible?

[Drachir]

There is MPEG-J : http://www.chiariglione.org/mpeg/faq/mp4-sys/sys-faq-mpegj.htm

But I don't know any free available implementation.

GPAC only supports JavaScript as far as I know.

[vsv]

Drachir, Thank you for clarification and link. Will be nice to have only one mp4 file with menu, subs, chapters and embeded interactive player:)

[Kurtnoise]

Can anyone confirm that there is a bug with -fps switch concerning avc-in-mp4 streams ?

I tried this :
mp4box -fps 23.976 -add input.mp4 output.mp4
and
mp4box -fps 23.976 -add input.mp4#1 output.mp4
the fps for the output file seems to be always 25.000.

btw, by doing 2 steps (extraction of the raw stream and then remux), it works fine:

mp4box -raw 1 input.mp4

mp4box -fps 23.976 -add input_track1.h264 output.mp4

In addition, fps switch seems to work fine with asp video files. So, if this bug can be reproduced, I'll submit a bug report in SF.

[bond]

its indeed a bug

[SeeMoreDigital]

I am able to change the fps of AVC "video only" streams with your latest YAMB build.

I've just changed an 25.000 fps stream to 50.000fps and 23.976fps...


Here are some samples (http://81.98.148.105/Uploaded_Files/Doom9_Forum_files/FPS_Tests.7z). Dunno if that helps you!

[Drachir]

I get the same problem with CVS from today. FPS is ignored if I import from a mp4 file. RAW import is working.


> MP4Box -add video.264:fps=23.976 out.mp4
> MP4Box -info out.mp4
* Movie Info *
Timescale 600 - Duration 00:06:24.191
Fragmented File no - 1 track(s)
File Brand isom - version 1

File has root IOD
Scene PL 0xff - Graphics PL 0xff - OD PL 0xff
Visual PL: AVC/H264 Profile (0x15)
Audio PL: No audio capability required (0xff)
No streams included in root OD

Track # 1 Info - TrackID 1 - TimeScale 23975 - Duration 00:06:24.191
Media Info: Language "und" - Type "vide" - Sub Type "avc1" - 9211 samples
MPEG-4 Config: Visual Stream - ObjectTypeIndication 0x21
AVC/H264 Video - Visual Size 192 x 144 - Version 1 Profile 0x4d Level 0x28
Self-synchronized


> MP4Box -add out.mp4#1:fps=15.000 out2.mp4
> MP4Box -info out2.mp4
* Movie Info *
Timescale 600 - Duration 00:06:24.191
Fragmented File no - 1 track(s)
File Brand isom - version 1

File has root IOD
Scene PL 0xff - Graphics PL 0xff - OD PL 0xff
Visual PL: AVC/H264 Profile (0x15)
Audio PL: No audio capability required (0xff)
No streams included in root OD

Track # 1 Info - TrackID 1 - TimeScale 23975 - Duration 00:06:24.191
Media Info: Language "und" - Type "vide" - Sub Type "avc1" - 9211 samples
MPEG-4 Config: Visual Stream - ObjectTypeIndication 0x21
AVC/H264 Video - Visual Size 192 x 144 - Version 1 Profile 0x4d Level 0x28
Self-synchronized


No change, still using TimeScale 23975.

[Kurtnoise]

According to Jean this is not a bug...:)

jeanlf Project Admin
Logged In: YES
user_id=815447

This is not a bug, or maybe a documentation bug. -fps is
used when importing any visual media except IsoMedia
tracks. To change the fps of a visual track, you must
export/re-import it.

http://sourceforge.net/tracker/index.php?func=detail&aid=1372338&group_id=84101&atid=571738

[mplaure]

Hi folks!
Has anyone tried to import a scene description stream and an audio stream into a new file with mp4box?
example:
-keepsys -add bifsfile.mp4 -add audioacc.acc output.mp4
The output.mp4 is not a valid file!
I have done the following: I have exported the audio of an bifsfile.mp4 that had audio by using -raw. After that, I have deleted the audio track from the bifsfile.mp4 and I have imported again this audio (audio.aac obtained with -raw) using -add, with the -keepsys option to avoid the destruction of the scene description track. The file obtained (output.mp4) is very similar to the original one. The tracks are the same and if I use the -info command line I can see that in the output, the audio is sincronized with the bifs track. But It doesnt sound!!!! In any player!!!! I can only see the rendered images.
Does anyone know why? Would it be a bug? I dont think so, but I have used almost all the mp4box -options to obtain what I want (that is, obviusly, to mux a scene description stream with audio) and it seems to be impossible.
Thank you for all your posts. I am learning a lot with you. :thanks:

[Drachir]

If you use -kepsys and add an audio track, the track will not be included in the SceneDescrition. -> No Systems Player will play it.

Extract the SceneDescription with MP4Box.

>MP4Box -bt yourFile.mp4

Open the BIFS text(*.bt) file with an editor and adjust the Scene and the OD.

Example: Video only mp4 file

Scene:
OrderedGroup {
children [
Shape {
appearance Appearance {
texture MovieTexture {
url [od:20]
}
}
geometry Bitmap {}

}
]
}

OD:
RAP AT 0 {
UPDATE OD [
ObjectDescriptor {
objectDescriptorID 20
esDescr [
ES_Descriptor {
ES_ID 201
OCR_ES_ID 1
decConfigDescr DecoderConfigDescriptor {
objectTypeIndication 33
streamType 4
bufferSizeDB 5737
maxBitrate 167224
avgBitrate 93336
decSpecificInfo DecoderSpecificInfo {
info %01%4D%00%28%FD%E1%00%18%67%4D%00%28%9A%71%01%82%74%20%00%00%03%00%20%00%00%03%03%11%E3%06%31%10%01%00%04%68%EE%BC%80
}
}
slConfigDescr SLConfigDescriptor {
useAccessUnitStartFlag true
useAccessUnitEndFlag true
useRandomAccessPointFlag true
useTimeStampsFlag true
timeStampResolution 25000
timeStampLength 32
}
muxInfo MuxInfo {
fileName "yourFile.mp4#201"
streamFormat "MP4"
}
}
]
}
]

Add the Audio Track:
Scene:
OrderedGroup {
children [
Sound2D {
source DEF Audio AudioSource {
url [od:10]
}
}
Shape {
appearance Appearance {
texture MovieTexture {
url [od:20]
}
}
geometry Bitmap {}

}
]
}

OD:
RAP AT 0 {
UPDATE OD [
ObjectDescriptor {
objectDescriptorID 10
esDescr [
ES_Descriptor {
ES_ID 101
OCR_ES_ID 1
muxInfo MuxInfo {
fileName "audio.aac"
}
langDescr LanguageDescriptor {
languageCode "eng"
}
}
]
}
ObjectDescriptor {
objectDescriptorID 20
esDescr [
ES_Descriptor {
ES_ID 201
OCR_ES_ID 1
decConfigDescr DecoderConfigDescriptor {
objectTypeIndication 33
streamType 4
bufferSizeDB 5737
maxBitrate 167224
avgBitrate 93336
decSpecificInfo DecoderSpecificInfo {
info %01%4D%00%28%FD%E1%00%18%67%4D%00%28%9A%71%01%82%74%20%00%00%03%00%20%00%00%03%03%11%E3%06%31%10%01%00%04%68%EE%BC%80
}
}
slConfigDescr SLConfigDescriptor {
useAccessUnitStartFlag true
useAccessUnitEndFlag true
useRandomAccessPointFlag true
useTimeStampsFlag true
timeStampResolution 25000
timeStampLength 32
}
muxInfo MuxInfo {
fileName "yourFile.backup.mp4#201"
streamFormat "MP4"
}
}
]
}
]

Create a backup from your file(yourFile.backup.mp4) und run "MP4Box -mp4 yourFile.bt"

This will make only sense if you want to adjust a mp4 file with a complex SceneDescription. For A/V only mp4 file create a new Scene(BIFS)/OD with -isma.

[mplaure]

Thanks a lot!!! :thanks:
It works!! Well, if you have a bifs file and you obtain the bt and you follow the steps... it works. But I have convert a *.swf to bifs with mp4box, and after that, I have obtained the *.bt, followed the steps... and despite the obtained file has the audio track... it doesnt sound!!!!
Thanks anyway

[DeeGee]

I have made some rips of my dvd's into 2cd mp4's. I use mp4box for the splitting, but it seems like i must set the size for the splitting larger than i need. eq. if i tell mp4box to split the file into 716800kb parts the end result usually are something like 2x ~714000kb parts and one ~4000kb part. I know that the splitting can only be done at keyframes, but still this seems flawed to me as upping the desired split size makes me end up with the result i wanted (2x ~716800kb parts). The annoying thing is that you need to experiment with what required part size gives you the one you really want...

D:\DVD\2001_A_SPACE_ODYSSEY>mp4box -splits 718000 "2001 - A Space Odyssey.mp4"
Storing split-file 2001 - A Space Odyssey_001.mp4 - duration 3989.24 seconds (714 252 kb)
Storing split-file 2001 - A Space Odyssey_002.mp4 - duration 4321.60 seconds (713 899 kb)
Storing split-file 2001 - A Space Odyssey_003.mp4 - duration 254.72 seconds (4 809 kb)

D:\DVD\2001_A_SPACE_ODYSSEY>mp4box -splits 720000 "2001 - A Space Odyssey.mp4"
Storing split-file 2001 - A Space Odyssey_001.mp4 - duration 4003.76 seconds (716 255 kb)
Storing split-file 2001 - A Space Odyssey_002.mp4 - duration 4509.12 seconds (715 726 kb)
Storing split-file 2001 - A Space Odyssey_003.mp4 - duration 52.68 seconds (979 kb)

D:\DVD\2001_A_SPACE_ODYSSEY>mp4box -splits 720500 "2001 - A Space Odyssey.mp4"
Storing split-file 2001 - A Space Odyssey_001.mp4 - duration 4007.16 seconds (716 742 kb)
Storing split-file 2001 - A Space Odyssey_002.mp4 - duration 4556.32 seconds (716 182 kb)
Storing split-file 2001 - A Space Odyssey_003.mp4 - duration 2.08 seconds (36 kb)

D:\DVD\2001_A_SPACE_ODYSSEY>mp4box -splits 720600 "2001 - A Space Odyssey.mp4"
Storing split-file 2001 - A Space Odyssey_001.mp4 - duration 4007.68 seconds (716 843 kb)
Storing split-file 2001 - A Space Odyssey_002.mp4 - duration 4557.88 seconds (716 115 kb)

And on top of that, the build I have at the moment seems to do the split inproperly. Or then it can't do it proprely due to the large ammount of reference frames (15). Part two of the split is unplayable due to reference frames missing (mplayer says this "[h264 @ 0098CCE8]reference picture missing during reorder".

[bond]

I have made some rips of my dvd's into 2cd mp4's. I use mp4box for the splitting, but it seems like i must set the size for the splitting larger than i need. eq. if i tell mp4box to split the file into 716800kb parts the end result usually are something like 2x ~714000kb parts and one ~4000kb part. I know that the splitting can only be done at keyframes, but still this seems flawed to me as upping the desired split size makes me end up with the result i wanted (2x ~716800kb parts). The annoying thing is that you need to experiment with what required part size gives you the one you really want...i dunno how mp4box chooses the splitpoint, but propably it does it in a way so you never get a file thats bigger than what you specified (which makes sense imho)

if its possible by specifying a larger filesize than wanted but still getting a file thats smaller than the really wanted size than its a bug

And on top of that, the build I have at the moment seems to do the split inproperly. Or then it can't do it proprely due to the large ammount of reference frames (15). Part two of the split is unplayable due to reference frames missing (mplayer says this "[h264 @ 0098CCE8]reference picture missing during reorder". this is either a bug in the encoder you used or in mp4box not splitting at idr-frames only

run the mp4videoinfo tool from mpeg4ip over the second split and tell us what frametype the first frame is

also did you use b-pyramid?

[DeeGee]

Well at least it never seems to do a larger slit than you requested, but still it seems at least somewhat bugged. If you look at those tests of splitting I had on the first post you can see how i got with 718000 a file sized 714 252 kb and also with 720000 it was still only 716 255 kb.

The file was encoded with x264 build 384H with this commandline:

x264.exe --pass 1 --bitrate 1218 --stats ".stats" --bframes 3 --b-pyramid --filter -2:-2 --subme 1 --weightb --analyse none --me dia --progress --no-psnr --sar 64:45 --zones 0,4836,q=40/125695,129651,q=40/202080,214139,q=40 --output NUL "D:\DVD\2001_A_SPACE_ODYSSEY\VIDEO_TS\2001 - A Space Odyssey.avs"

x264.exe --pass 3 --bitrate 1218 --stats ".stats" --bframes 3 --b-pyramid --ref 15 --filter -2:-2 --analyse all --weightb --me umh --subme 7 --b-rdo --mixed-refs --8x8dct --trellis 1 --cqmfile "C:\Program Files\x264\soulhunter_v2.cfg" --sar 64:45 --progress --zones 0,4836,q=40/125695,129651,q=40/202080,214139,q=40 --output "D:\DVD\2001_A_SPACE_ODYSSEY\VIDEO_TS\2001 - A Space Odyssey_video.mp4" "D:\DVD\2001_A_SPACE_ODYSSEY\VIDEO_TS\2001 - A Space Odyssey.avs"


So I did use 3 b-frames and b-pyramid.

And mp4videoinfo gives this:

I need to find compiled version somewhere first... There seems to be references to this mp4videoinfo on mpeg4iptools site, but can't find even sources to it.

[bond]

Well at least it never seems to do a larger slit than you requested, but still it seems at least somewhat bugged. If you look at those tests of splitting I had on the first post you can see how i got with 718000 a file sized 714 252 kb and also with 720000 it was still only 716 255 kb.thats indeed not good, you might want to tell the gpac devs about it

And mp4videoinfo gives this:well you can also try extracting the avc stream from the 2nd mp4 to raw and than run h264_parse from mpeg4ip over it
tell me what it says there after the NAL with the SEI (i guess there must be some I frame, but what does it say in detail)

[DeeGee]

As the output of h264_parse is complete gibberish to me, i'll just give you a link to the whole log (http://students.oamk.fi/~t3paja01/sekalaista/h264_parse.7z)...

[bond]

As the output of h264_parse is complete gibberish to me, i'll just give you a link to the whole log (http://students.oamk.fi/~t3paja01/sekalaista/h264_parse.7z)...ok, you have just found a bug in mp4box! the 2nd file doesnt start with a keyframe, which means mp4box is cutting at the wrong place

please make a bug report reporting all this here (http://sourceforge.net/tracker/?group_id=84101&atid=571738)

[Hyper Shinchan]

I hate making request (and I hate it even more if I don't know if they are intelligent request), but I'd like to know if it is possible to upgrade the 3GPP release to the 6 one. Actually I think that the only player that really handle 3gp6 is QT (the outputted 3gpp made with avc video show 3gp6 media type, and another application that outputs this type of file is the reference AAC++ encoder of 3GPP.org), and actually it's pretty useless to do it (because there aren't 3GPP 6 enabled mobile phone), but it can be useful start thinking about it now.

[bond]

afaik qt doesnt handle ttxt in 3gp6 files, or something crappy like that

[Kurtnoise]

Can anyone confirm that adding A/V mp4 files to a new mp4 file crashes mp4box now ?

I tested my own compile and the new Celtic_Druid builds and mp4box crashes...:(

Such as :
mp4box -add audio_input.mp4 output.mp4

It works fine with builds compiled before December 22th...

[Sharktooth]

It makes even x264 crash when outputting to mp4...
Im still using the old gpac to compile x264... so yes, there must be a problem with it.

[Kurtnoise]

This is due maybe to the new enhancements in gpac libraries....I'll investigate.

[Zero1]

Can anyone confirm that adding A/V mp4 files to a new mp4 file crashes mp4box now ?

I tested my own compile and the new Celtic_Druid builds and mp4box crashes...:(

Such as :
mp4box -add audio_input.mp4 output.mp4

It works fine with builds compiled before December 22th...

Yes, mine crashed when adding a HE-AAC .mp4 track. It looks like it didn't set the SAR on my video either, fortunately I had an older build to roll back to.

[Kurtnoise]

It seems to work fine now...I'm testing fresh cvs builds from January 14th/16th and all seem to be ok.

I'll upload a build tonight for my part.

[Daniel Mott]

MP4Box crashes just after importing audio when using MeGUI 0.2.3.2024 as a frontend to mux a .264 with an AAC .mp4 audio file. When approximately will a working MP4Box be released and where is the best place to download it? :)

[bond]

MP4Box crashes just after importing audio when using MeGUI 0.2.3.2024 as a frontend to mux a .264 with an AAC .mp4 audio file. When approximately will a working MP4Box be released and where is the best place to download it? :)first of all what mp4box compile did you use?

[Daniel Mott]

Sorry for not specifying! I used the latest one on this page:
http://www.aziendeassociate.it/cd.asp?dir=/gpac/dev (MP4Box - GPAC version 0.4.1-DEV)

Maybe there's a better place to download ready-compiled versions?

[bond]

indeed, the compile available there is borked

celtic_druid has already made a new one today, which works fine
you will have to wait till it gets posted on that page

[Kurtnoise]

Maybe there's a better place to download ready-compiled versions?
http://kurtnoise.free.fr/mp4tools/

[Daniel Mott]

Great, thanks a lot! I was getting really eager to do some muxing... :D

[Zero1]

Hmm, just wondering if anyone else is interested in Variable Frame Rate MP4, by using MKV timecode files? (No, I'm not volunteering, unfortunately I can't program, otherwise I'd gladly help :()

There was a request put in a while ago but nothing as of yet.

The suggestion posted was that you would encode and mux your decimated video, and add the MKV timecode file at time of muxing and have MP4box alter the frame durations of the MP4 file (I'm assuming that's how it works, I recall the stts atom being mentioned).

I think it's a good idea since you will be able to use existing software to create VFR MP4, as long as the feature exists in MP4box.

The other thing I look forward to is reduced encoding times, you can decimate quite a lot of frames in anime, and encoding H.264 at high quality it really saves time as well as helps with playback requirements :D

[bond]

Zero1:

there has been already a feature request for adding this timecode thingie to mp4box, with no response from jeanlf till now tough:
http://sourceforge.net/tracker/index.php?func=detail&aid=1356516&group_id=84101&atid=571741

the next thing would be being able to demux vfr mkv to .mp4 with mkvextract, but mosu doesnt seem to be willing to do that

the next thing would be getting vfr output to work within x264. i am pretty sure the gpac lib already does vfr, so it would need to be added on the x264 side (frame dropping aso...)

[Zero1]

Hmm, I originally posted thinking that it had been added, seems it has not. That's a bit of a let down lol, but still, there's always the next day...