Hey!
Good catch!. It really is a matter of buffering. When I set buffer size to 1 byte (-b:1, because 0 sets 1024) I get only 2 extra samples instead of 1024 at the _beginning_ of output file, but as a side effect a big portion of data (around 64kB) is lost at the end (or maybe in the middle, I'm not sure at the moment). And, of course, the processing is significantly slower (nonrt mode).
-- Artur Poniższą wiadomość napisał(a) Mr. Jeremy Hall <jhall@email-addr-hidden.com> w dniu 13.04.2010 13:41: > Hi, > > I think actually it is it rolls an extra "buffersize" on the end. If > you use the -b flag and make it something large you'll see my point, > and it's annoying because it includes bits from a live track not > intended. > > _J > > On Sun, Apr 11, 2010 at 10:01:52AM +0200, Artur wrote: >> Hello. >> >> Ecasound adds extra 1024 samples at the beginnig of output stream. >> For example, when I do something like: >> >> ecasound -t:0 -i:null -f:u8,1,48000 -o:output.wav >> >> I always get 1024 samples in output file (regardless of the sample >> rate, number of channels etc.) (version of Ecasound: git HEAD). >> >> Question to developer(s): would it be possible to eliminate this >> somehow ? Of course it is the lowest priority of the lowest >> priority requests. >> >> Regards. -- Artur >> >> ------------------------------------------------------------------------------ >> >> Download Intel® Parallel Studio Eval >> Try the new software tools for yourself. Speed compiling, find >> bugs proactively, and fine-tune applications for parallel >> performance. See why Intel Parallel Studio got high marks during >> beta. http://p.sf.net/sfu/intel-sw-dev >> _______________________________________________ Ecasound-list >> mailing list Ecasound-list@email-addr-hidden >> https://lists.sourceforge.net/lists/listinfo/ecasound-list > ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ Ecasound-list mailing list Ecasound-list@email-addr-hidden https://lists.sourceforge.net/lists/listinfo/ecasound-listReceived on Tue Apr 13 16:15:03 2010
This archive was generated by hypermail 2.1.8 : Tue Apr 13 2010 - 16:15:03 EEST