Subject: Re: [linux-audio-dev] lowish-latency patch and toolchain
From: Tom Pincince (stillone_AT_snowcrest.net)
Date: Mon Jul 10 2000 - 22:15:03 EEST
A common technique for developing an understanding of mathematical
functions is to investigate their behavior at the extremes. All of that
"as n aproaches infinity" or "as n aproaches 0" stuff found in college
text books. The purpose of my letter was to illustrate the feasability
of using regular Linux for audio by investigating the extreme. As the
total number of simultanious processes aproaches 0, what kind of latency
performance will Linux exhibit for the few remaining (audio) processes.
This example was not targeted at the LAD people, it was targeted at the
people at linux-kernel who seemed to be having way too much fun finding
all of the different ways of saying "if you want to do audio then Linux
is not for you". I followed the thread at l-k triggered by pbd's
letter. The signal to noise ratio was so bad that I eventually found it
too toxic to tollerate. I started to think of pbd as a saint just for
his willingness to hang in there till the end. When the same kind of
noise started to develop in the threads that were trying to
constructively address solutions to the issue, I felt like doing
something about it. If even one person on l-k woke up this morning
ready to spit venom, and realized that low latency audio on regular
Linux is at least possible in the extreme case after reading my post,
then it was worth doing. I was just trying to improve the s/n ratio at
l-k while some people there are actually willing to work on our behalf.
Tom
This archive was generated by hypermail 2b28 : Mon Jul 10 2000 - 22:45:49 EEST