Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Bug: Blank seek pts in Flac when using Convertr (Read 8528 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Bug: Blank seek pts in Flac when using Convertr

Flac files created with the Converter have an excessive number of seek points (max out at 1217).

metaflac.exe --list encoded_from_cmdline.flac:
Code: [Select]
METADATA block #0
 type: 0 (STREAMINFO)
 is last: false
 length: 34
 minumum blocksize: 4608 samples
 maximum blocksize: 4608 samples
 minimum framesize: 14 bytes
 maximum framesize: 13645 bytes
 sample_rate: 44100 Hz
 channels: 2
 bits-per-sample: 16
 total samples: 173531736
 MD5 signature: fe037483aed4f9b799a80eaf3f01288b
METADATA block #1
 type: 3 (SEEKTABLE)
 is last: false
 length: 7074
 seek points: 393
   point 0: sample_number=0, stream_offset=0, frame_samples=4608
   point 1: sample_number=437760, stream_offset=839360, frame_samples=4608
   point 2: sample_number=880128, stream_offset=1735713, frame_samples=4608
   point 3: sample_number=1322496, stream_offset=2587232, frame_samples=4608
   point 4: sample_number=1764864, stream_offset=3451039, frame_samples=4608
   point 5: sample_number=2207232, stream_offset=4350277, frame_samples=4608
   point 6: sample_number=2644992, stream_offset=5193427, frame_samples=4608
   point 7: sample_number=3087360, stream_offset=6045617, frame_samples=4608
   .
   .
   .
   point 390: sample_number=172205568, stream_offset=432912851, frame_samples=4608
   point 391: sample_number=172647936, stream_offset=433971129, frame_samples=4608
   point 392: sample_number=173085696, stream_offset=435016634, frame_samples=4608
METADATA block #2
 type: 4 (VORBIS_COMMENT)
 is last: false
 length: 40
 vendor string: reference libFLAC 1.1.2 20050205
 comments: 0
METADATA block #3
 type: 1 (PADDING)
 is last: true
 length: 4096


metaflac.exe --list encoded_from_foobar.flac:
Code: [Select]
METADATA block #0
 type: 0 (STREAMINFO)
 is last: false
 length: 34
 minumum blocksize: 4608 samples
 maximum blocksize: 4608 samples
 minimum framesize: 14 bytes
 maximum framesize: 13645 bytes
 sample_rate: 44100 Hz
 channels: 2
 bits-per-sample: 16
 total samples: 173531736
 MD5 signature: fe037483aed4f9b799a80eaf3f01288b
METADATA block #1
 type: 3 (SEEKTABLE)
 is last: false
 length: 21906
 seek points: 1217
   point 0: sample_number=0, stream_offset=0, frame_samples=4608
   point 1: sample_number=437760, stream_offset=839360, frame_samples=4608
   point 2: sample_number=880128, stream_offset=1735713, frame_samples=4608
   point 3: sample_number=1322496, stream_offset=2587232, frame_samples=4608
   point 4: sample_number=1760256, stream_offset=3441619, frame_samples=4608
   point 5: sample_number=2202624, stream_offset=4341444, frame_samples=4608
   point 6: sample_number=2644992, stream_offset=5193427, frame_samples=4608
   point 7: sample_number=3087360, stream_offset=6045617, frame_samples=4608
   .
   .
   .
   point 391: sample_number=172486656, stream_offset=433586111, frame_samples=4608
   point 392: sample_number=172924416, stream_offset=434631868, frame_samples=4608
   point 393: sample_number=173366784, stream_offset=435642974, frame_samples=4608
   point 394: sample_number=173810299, stream_offset=0, frame_samples=0
   point 395: sample_number=174251442, stream_offset=0, frame_samples=0
   point 396: sample_number=174692585, stream_offset=0, frame_samples=0
   .
   .
   .
   point 1214: sample_number=535547471, stream_offset=0, frame_samples=0
   point 1215: sample_number=535988614, stream_offset=0, frame_samples=0
   point 1216: sample_number=536429757, stream_offset=0, frame_samples=0
METADATA block #2
 type: 4 (VORBIS_COMMENT)
 is last: false
 length: 40
 vendor string: reference libFLAC 1.1.2 20050205
 comments: 0
METADATA block #3
 type: 1 (PADDING)
 is last: true
 length: 4096


Won't speculate on cause, but figured it was worth mentioning. Can anyone else confirm? Thanks.
f to c to f to c

Bug: Blank seek pts in Flac when using Convertr

Reply #1
Are you sure that you used the exact same settings to create both files?

I have a hunch that you encoded encoded_from_cmdline.flac from a wave file, while for the creation of encoded_from_foobar.flac flac.exe was fed via pipe.
A riddle is a short sword attached to the next 2000 years.

Bug: Blank seek pts in Flac when using Convertr

Reply #2
I can recreate the output with foobar2000 0.9's converter; %s -o %d gives normal results but - -o %d results in lots of seek points.

However, I can't reproduce the abnormal behaviour when I run

cat input.wav | flac -V8 -s - -o from_stdin.flac

Bug: Blank seek pts in Flac when using Convertr

Reply #3
Apparently handling of truncated source WAV files in FLAC tools is broken - encoding from stdin causes WAV file with header signaling maximum possible length to be generated, since exact total length of stream being encoded is not known in advance; then "truncated file" behavior is simulated when out of data to encode.
This behavior will not be changed on our side; use %s if this problem disturbs you.
Microsoft Windows: We can't script here, this is bat country.

Bug: Blank seek pts in Flac when using Convertr

Reply #4
does this make the file lossy?
or it doesn't matter for quality?

Bug: Blank seek pts in Flac when using Convertr

Reply #5
No.
Yes.
Your file has simply more seek points (and a few additional/unecessary Kb).

Bug: Blank seek pts in Flac when using Convertr

Reply #6
so this is more of an esthetical file issue than quality issue that is (user not liking 1217 seek points). why wvpack has a astronomical speed when it seeks?