[FFmpeg-user] V_QUICKTIME used as codec ID for DPX/TIFF rewrapped to Matroska

[Kieran O Leary]

Hi,

When re-wrapping DPX or TIFF images to Matroska, it appears that FFmpeg
uses the V_QUICKTIME codec ID. Why is this used? Would something like
V_UNCOMPRESSED (for raw TIFF anyhow) b more appropriate/

The reason why I'm investigating this is that DPC and TIFF rewrappings to
Matroska appear to be fully reversible with matching whole-file checksums.

To replicate using ffmpeg and mediainfo:

$ ./ffmpeg -f lavfi -i testsrc -c:v tiff -compression_algo raw -y -t 1
1_tiff.mkv && mediainfo --Details=1 1_tiff.mkv|grep -i codec
ffmpeg version N-92521-g737b5f5869 Copyright (c) 2000-2018 the FFmpeg
developers
built with gcc 7 (Ubuntu 7.3.0-27ubuntu1~18.04)
configuration:
libavutil 56. 24.101 / 56. 24.101
libavcodec 58. 40.100 / 58. 40.100
libavformat 58. 23.100 / 58. 23.100
libavdevice 58. 6.100 / 58. 6.100
libavfilter 7. 46.100 / 7. 46.100
libswscale 5. 4.100 / 5. 4.100
libswresample 3. 4.100 / 3. 4.100
Input #0, lavfi, from 'testsrc':
Duration: N/A, start: 0.000000, bitrate: N/A
Stream #0:0: Video: rawvideo (RGB[24] / 0x18424752), rgb24, 320x240
[SAR 1:1 DAR 4:3], 25 tbr, 25 tbn, 25 tbc
Stream mapping:
Stream #0:0 -> #0:0 (rawvideo (native) -> tiff (native))
Press [q] to stop, [?] for help
Output #0, matroska, to '1_tiff.mkv':
Metadata:
encoder : Lavf58.23.100
Stream #0:0: Video: tiff (tiff / 0x66666974), rgb24, 320x240 [SAR 1:1
DAR 4:3], q=2-31, 200 kb/s, 25 fps, 1k tbn, 25 tbc
Metadata:
encoder : Lavc58.40.100 tiff
frame= 25 fps=0.0 q=-0.0 Lsize= 5638kB time=00:00:00.96
bitrate=48061.9kbits/s speed=58.3x
video:5636kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB
muxing overhead: 0.034428%
0001AE CodecID - V_QUICKTIME (13 bytes)
0001CB CodecPrivate (93 bytes)
$ ./ffmpeg -f lavfi -i testsrc -c:v dpx -y -t 1 1_dpx.mkv && mediainfo
--Details=1 1_tiff.mkv|grep -i codec
ffmpeg version N-92521-g737b5f5869 Copyright (c) 2000-2018 the FFmpeg
developers
built with gcc 7 (Ubuntu 7.3.0-27ubuntu1~18.04)
configuration:
libavutil 56. 24.101 / 56. 24.101
libavcodec 58. 40.100 / 58. 40.100
libavformat 58. 23.100 / 58. 23.100
libavdevice 58. 6.100 / 58. 6.100
libavfilter 7. 46.100 / 7. 46.100
libswscale 5. 4.100 / 5. 4.100
libswresample 3. 4.100 / 3. 4.100
Input #0, lavfi, from 'testsrc':
Duration: N/A, start: 0.000000, bitrate: N/A
Stream #0:0: Video: rawvideo (RGB[24] / 0x18424752), rgb24, 320x240
[SAR 1:1 DAR 4:3], 25 tbr, 25 tbn, 25 tbc
Stream mapping:
Stream #0:0 -> #0:0 (rawvideo (native) -> dpx (native))
Press [q] to stop, [?] for help
Output #0, matroska, to '1_dpx.mkv':
Metadata:
encoder : Lavf58.23.100
Stream #0:0: Video: dpx (dpx / 0x20787064), rgb24, 320x240 [SAR 1:1
DAR 4:3], q=2-31, 200 kb/s, 25 fps, 1k tbn, 25 tbc
Metadata:
encoder : Lavc58.40.100 dpx
frame= 25 fps=0.0 q=-0.0 Lsize= 5668kB time=00:00:00.96
bitrate=48312.5kbits/s speed=57.6x
video:5666kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB
muxing overhead: 0.033439%
0001AE CodecID - V_QUICKTIME (13 bytes)
0001CB CodecPrivate (93 bytes)
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-***@ffmpeg.org with subject "

[Carl Eugen Hoyos]

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

Post by Kieran O Leary
Why is this used?
Would something like V_UNCOMPRESSED (for raw TIFF anyhow)
be more appropriate/

If you use -vcodec rawvideo instead of tiff or dpx (which are not
rawvideo) you get V_UNCOMPRESSED but note that matroska
is not a useful container for rawvideo because it is not well
defined, use nut instead.

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-***@ffmpeg

[Kieran O Leary]

Hi Carl,

Post by Carl Eugen Hoyos

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

I don't think they exist - this remuxing is most likely not common/never
used. I'll get in touch with the CELLAR/IETF list and see if anyone has any
ideas. Also, as per my last email, DPX is fine - V_MS/VFW/FOURCC / dpx . I
just had a crappy command line :(

Post by Carl Eugen Hoyos

Post by Kieran O Leary
Why is this used?
Would something like V_UNCOMPRESSED (for raw TIFF anyhow)
be more appropriate/

If you use -vcodec rawvideo instead of tiff or dpx (which are not
rawvideo) you get V_UNCOMPRESSED but note that matroska
is not a useful container for rawvideo because it is not well
defined, use nut instead.

Cool, but why is a QuickTime ID used ? Is this some sort of generic
fallback that ffmpeg uses?

Best,

Kieran.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-***@ffmpeg.org with subject "unsubscribe"

[Kieran O Leary]

Post by Kieran O Leary
Hi Carl,

Post by Carl Eugen Hoyos

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

I don't think they exist - this remuxing is most likely not common/never
used. I'll get in touch with the CELLAR/IETF list and see if anyone has any
ideas.

P.S - Carl - not sure if you're on CELLAR, but here's a link to the thread
I started - https://mailarchive.ietf.org/arch/browse/cellar/
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-request@

[Carl Eugen Hoyos]

Post by Kieran O Leary
Hi Carl,

Post by Carl Eugen Hoyos

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

I don't think they exist - this remuxing is most likely not common/never
used. I'll get in touch with the CELLAR/IETF list and see if anyone has any
ideas.
Also, as per my last email, DPX is fine - V_MS/VFW/FOURCC / dpx . I
just had a crappy command line :(

I don't think it's fine: You are using a very unusual fourcc to put something
in matroska that is not meant to be there, expect massive compatibility
issues.

Post by Kieran O Leary

Post by Carl Eugen Hoyos

Post by Kieran O Leary
Why is this used?
Would something like V_UNCOMPRESSED (for raw TIFF anyhow)
be more appropriate/

If you use -vcodec rawvideo instead of tiff or dpx (which are not
rawvideo) you get V_UNCOMPRESSED but note that matroska
is not a useful container for rawvideo because it is not well
defined, use nut instead.

Cool, but why is a QuickTime ID used ? Is this some sort of generic
fallback that ffmpeg uses?

Just like the V_MS/VFW/FOURCC codec id above for avi, there is
also a QuickTime fallback defined for Matroska (not by FFmpeg),
tiff happens to be defined for QuickTime but not in avi, therefore
the QuickTime code point is used for Matroska.
Contrary to dpx above, this can at least be argued to be following
some standard.

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-***@ffmpeg.org with subj

[Kieran O Leary]

Post by Carl Eugen Hoyos

Post by Kieran O Leary
Hi Carl,

Post by Carl Eugen Hoyos

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

I don't think they exist - this remuxing is most likely not common/never
used. I'll get in touch with the CELLAR/IETF list and see if anyone has

any

Post by Kieran O Leary
ideas.
Also, as per my last email, DPX is fine - V_MS/VFW/FOURCC / dpx . I
just had a crappy command line :(

I don't think it's fine: You are using a very unusual fourcc to put something
in matroska that is not meant to be there, expect massive compatibility
issues.

Good point.

Post by Carl Eugen Hoyos

Post by Kieran O Leary

Post by Carl Eugen Hoyos

Post by Kieran O Leary
Why is this used?
Would something like V_UNCOMPRESSED (for raw TIFF anyhow)
be more appropriate/

If you use -vcodec rawvideo instead of tiff or dpx (which are not
rawvideo) you get V_UNCOMPRESSED but note that matroska
is not a useful container for rawvideo because it is not well
defined, use nut instead.

Cool, but why is a QuickTime ID used ? Is this some sort of generic
fallback that ffmpeg uses?

Just like the V_MS/VFW/FOURCC codec id above for avi, there is
also a QuickTime fallback defined for Matroska (not by FFmpeg),
tiff happens to be defined for QuickTime but not in avi, therefore
the QuickTime code point is used for Matroska.
Contrary to dpx above, this can at least be argued to be following
some standard.

This all makes sense to me and it's incredibly helpful. I see the TIFF
mapping in the QuickTime spec now.
So I guess that TIFF, and DPX need to be mapped in the Matroska
specification?

Best,

Kieran.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffm

[Carl Eugen Hoyos]

Post by Kieran O Leary
So I guess that TIFF, and DPX need to be mapped in the Matroska
specification?

Not sure if I am the right person to answer but why would you want to
put images into another container?

Shouldn't you keep the original image files and use ffmpeg to produce
a view copy in the requested quality?

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
f

[Kieran O Leary]

Post by Carl Eugen Hoyos

Post by Kieran O Leary
So I guess that TIFF, and DPX need to be mapped in the Matroska
specification?

Not sure if I am the right person to answer but why would you want to
put images into another container?
Shouldn't you keep the original image files and use ffmpeg to produce
a view copy in the requested quality?

Image sequences are difficult to deal with,especially with LTO tape
libraries. Archives commonly use zip in either a compressed or uncompressed
form to alleviate this issue.
It's preferable to have an uncompressed or lossless single file for ease of
use, compatibility with tape libraries and digital asset management systems.
I would prefer to retain the original image sequence but it is not common
practise I'm archives from what I understand,for the reasons outlined above.
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-re

[Carl Eugen Hoyos]

Post by Kieran O Leary

Post by Carl Eugen Hoyos

Post by Kieran O Leary
So I guess that TIFF, and DPX need to be mapped in the Matroska
specification?

Not sure if I am the right person to answer but why would you want to
put images into another container?
Shouldn't you keep the original image files and use ffmpeg to produce
a view copy in the requested quality?

Image sequences are difficult to deal with,especially with LTO tape
libraries. Archives commonly use zip in either a compressed or
uncompressed form to alleviate this issue.

How does this contradict what I write above?
Assuming a current high-quality, widely adopted video codec for the
view copy would be obsolete after a decade (that's too soon, no?),
that would mean one necessary access to the file system every ten
years: Is that too difficult to deal with?

Post by Kieran O Leary
It's preferable to have an uncompressed or lossless single
file for ease of use

I don't think so...
(But fortunately, it is your decision!)

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-us

[Kieran O Leary]

Post by Kieran O Leary
Hi,
When re-wrapping DPX or TIFF images to Matroska, it appears that FFmpeg
uses the V_QUICKTIME codec ID. Why is this used? Would something like
V_UNCOMPRESSED (for raw TIFF anyhow) b more appropriate/

Ugh, This only happens with tiff - my command line was wrong for the DPX
mediainfo check :(((((
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmpeg-user-***@ffmpeg.org with subject "

[Carl Eugen Hoyos]

Post by Kieran O Leary
When re-wrapping DPX or TIFF images to Matroska, it appears that
FFmpeg uses the V_QUICKTIME codec ID.

Where can I find the specification for tiff in matroska and dpx in matroska?

Post by Kieran O Leary
Why is this used?

The answer depends on above...

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
ffmpeg-***@ffmpeg.org
http://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
ffmp