Subject: Re: [ecasound] Two ewf-file bugs
From: Kai Vehmanen (k_AT_eca.cx)
Date: Thu May 29 2003 - 20:17:56 EEST
On Wed, 28 May 2003, Vegard Lima wrote:
> during a rather long session with ecasound 2.2.3 (latest from CCRMA)
> I found what appears to be a couple of bugs in ecasound's handling
> of ewf-files. The following should show how to reproduce them:
Thanks for these bug reports! In fact, these reports are nice examples of
good bug reports: 1) bug description + 2) simple instructions how to
reproduce the bug. Very often (2) is missing from reports. Having an easy
way to reproduce the problem makes it _so_ much easier (and faster!) for
me to fix bugs.
> # BUG #1: wrong length of ewf-file.
[...]
> $ ecasound -f:16,2,48000 -a:1 -i alsahw,0 -o 48kHz.wav -t:10
Yups, there was a ewf-specific bug in handling files with srate that
differed from the initial chainsetup rate (usually 44100). Now fixed in
CVS.
> # BUG #2: Hello feedback.
[...]
> $ ecasound -a:1 -i 48kHz.ewf -a:2 -f:16,2,48000 -i null \
> -a:1 -el:lcrDelay,500,25,490,25,510,30,50,50,50,25,1 -a:1,2 -o alsahw,0
[...]
> As soon as 48kHz.ewf is finished playing the delay-effect
> gets increasingly louder.
Now this was an interesting one. Ecasound's engine didn't properly erase
the samplebuffer after 48kHz.ewf had ended, and as a result, lcrDelay kept
getting the last block over and over again. And as it was the same
identical block, the plugin's feedback lines kept amplifying the signal.
Also fixed in CVS!
-- http://www.eca.cx Audio software for Linux!
This archive was generated by hypermail 2b28 : Thu May 29 2003 - 20:18:51 EEST