<Homer voice>
Wow Bob Katz
</Homer voice> :-)
Hi Bob,
I'd never heard of wav tags until you mentioned them either but I've checked and Steinberg's Wavelab 6 will write tags that my Sqeezeboxes can read.
Surely they gave you a free copy ;-)
Welcome aboard
Craig
Results 11 to 20 of 48
-
2007-02-12, 12:42 #11MC2Slim - Windows Shell and J River Media Center Integration for Squeezebox.
http://www.duff-zapp.co.uk
-
2007-02-12, 14:03 #12Senior Member
- Join Date
- Apr 2006
- Posts
- 130
From what I understood of the .wav format is that it is more a kind of container than a simple file. there is the obligatory header and audio data, but a lot of other blocks are possible. So in theory you could add tags without interfering with normal playback.
Has anyone looked at the byte-level what happens?
Going to look at that application. This is a find!
-
2007-02-12, 14:18 #13Member
- Join Date
- Nov 2006
- Posts
- 64
WAV uses a simple RIFF structure:
http://en.wikipedia.org/wiki/RIFF
It actually mentions a quite old INFO chunk type at the beginning of the file, which would be more official than a ID3v1 tag at the end of the file, which shouldn't interfer with audio anyway, since the data chunk has a defined size and reading wouldn't spread further into an attached tag. But Wikipedia clearly says that INFO chunks have to be at the beginning.
I've written a RIFF parser once, and if someone has a properly tagged file (meaning an INFO chunk at the beginning), I could use that information to add support for those tags to TagsRevisited (the official INFO chunk style information, not the ID3 tricks that app from the first post mentions
).
Last edited by CCRDude; 2007-02-12 at 14:20.
-
2007-02-12, 14:21 #14
Not quite: WAV is a header and a data block. At the start of the data block, there is a length. In theory, a player will read the length and use that to determine the end of the data and not send anything past that point.
http://mural.uv.es/samecues/wav.htm shows a picture of how they append the id3 to the end of the file, and http://www.ringthis.com/dev/wave_format.htm shows the format of a WAV file. (Note the "size of data chunk / 4 bytes - DWORD / Number of bytes of data is included in the data section." part.)
Of course, using id3 tags on WAVs is completely non-standard. As long as you are being non-standard you may as well use Vorbis comments, since at least those suck less.
(And I still don't understand why not just FLAC the files... it isnt like FLAC takes much time at all and then you don't have to worry about whether undocumented/undefined behavior will continue to work.)
-
2007-02-12, 14:41 #15Member
- Join Date
- Nov 2006
- Posts
- 64
Except that that documentation mentions only one chunk/subchunk: WAVE/fmt; while actually there could be multiple chunks
-
2007-02-12, 14:51 #16
I've looked at the file I tagged in Wavelab with a hex editor, it starts with RIFF then the tags then padding then the data.
Slimserver sees them ok though
CraigMC2Slim - Windows Shell and J River Media Center Integration for Squeezebox.
http://www.duff-zapp.co.uk
-
2007-02-12, 18:52 #17
Duhhhh..
Hi, Craig, no comment on the free, but yes I did know that Wavelab will write tags that Slimserver can read. But Wavelab will rewrite the entire audio file which is a pain, while tagging programs find a way to insert themselves in the header and only rewrite the header part of the file, saving a lot of time.
-
2007-02-12, 19:04 #18
I agree with snarly that there are many good reasons to use FLAC. My choice not to is a workflow issue and perhaps an ego issue as well. I take my 24 bit files and copy them to the server, done.
When clients come in to listen to 96 kHz/24 bit originals I want to say that's what I'm playing without going into a discussion of lossless and all that. I even turned off the FLAC streaming option in the server so that uncompressed wavs are sent by the server. Eats up more ethernet bandwidth than FLAC, but nothing significant in our studio.
Matter of choice. I certain agree with a poster who said that there is or should be no sonic difference between WAVs and FLAC. The poster who heard a difference needs to do some blind testing.
BK
-
2007-02-12, 19:26 #19
Uhoh, don't mention double blind testing.
(Though I agree: I am of the school that my brain is easily confused by subjective clues and expectations... When logic says something makes no sense, I want proof that shows logic is wrong since I know how easy it is to fool the brain.)
-
2007-02-12, 21:31 #20

Reply With Quote

