Subject: RE: [ecasound] need some buffering advice
From: Jamie Lutch (jlutch_AT_altairinc.com)
Date: Thu Sep 18 2003 - 02:14:49 EEST
seems to be working well now, I simplified and am just running the following
with no xruns:
ecasound -f:s32_le,2,96000,i -i
alsa,hw:PDAudioCF -z:intbuf -f:s24_le,2,96000,i -o test.wav
a few questions:
1. in the syntax I am running above, am I correct to assume that the engine
is reading 32bit from the soundcard (bottom 8 are zeros), then it just
truncates the 8 zero bits before writing to disk? there is no dithering or
digital processing going on, right?
2. in the case I need to use -z:db, 100000, I am confused by this statement:
"The default is'-z:nodb,100000', which is 100000 samples of buffering per
non-realtime
audio object (files, pipes, etc). For 32bit, 2ch audio this means 0.8MB of
buffering per audio object... in other words quite a lot."
so 100000 samples of buffering is not the same as 100000 samples,
apparently. it seems from your statement that a 'sample of buffer' is 8KB.
is this fixed, or is it a function of the soundcard buffer size?
This archive was generated by hypermail 2b28 : Thu Sep 18 2003 - 02:12:53 EEST