Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: Live Gig Rig Logistics



uhm... interesting... are you running your own software?

ciao
leo


At 18.30 29/10/99 -0700, you wrote:
>At 11:14 AM -0700 10/29/99, Ken Melms wrote:
>>How did you solve the latency issue?>What OS? What
>>hardware?
>
>If you read the fine print I was careful to say I only dealt with latency
>in a looping situation, which is a special circumstance.
>What I meant was, if the computer is only used to output the loop, not any
>"dry" signal, then my scheme works. If you also want to replace a bunch of
>stomp boxes and reverbs with software, I agree you are stuck with a fixed
>minimum delay (the latency) that cannot be turned off. (Although a good
>fast computer plus a sound card with very small buffers can bring this 
>down
>to just a few milliseconds.)
>
>The idea is pretty simple but seems to be a little hard to believe. 
>Several
>people said it wouldn't work, and I still doubt it sometimes even after
>demonstrating it to 40 people. Let's say you want a 10.000 second loop and
>your computer's total latency (from audio input through the OS, through
>your application, and back down out to the audio output) is 0.2 seconds--
>enough to throw off even a viola player or an accordionist or a guitarist
>or whatever (duck!). You set up a 10 second delay line, with two delay
>taps-- one at 10.000 seconds and one at 9.800 seconds. The 10 second tap 
>is
>used for feedback or regeneration, which happens inside the software every
>sample so there is no latency problem there. The 9.8 second tap is what 
>you
>listen to. If you play along in perfect sync with what you hear, it ends 
>up
>back inside the computer in sync with the 10.0 s internal feedback tap.
>Note that even though you are technically listening to a 9.8 s tap the
>delay that you hear, and the total loop length, is exactly 10.0 s.
>
>In this scheme the computer's internal sense of "now" is still slightly 
>off
>from the outside world, but it is nonetheless possible to overdub in
>perfect sync (barring musicianship like mine, of course!). If this
>discrepancy bothers you, perhaps for existential reasons, I can't help. 
>But
>I wouldn't be surprised if our mind-body connections work in a similar
>manner when we're keeping time anyway. In any case, for me the concept of
>multiple nows fits right in with my interest in altering time perception
>through music.
>
>Anyway, I worked this out when I had a system with over 300 msec latency,
>and it took care of that nicely. I am currently using a MOTU 2408 card in 
>a
>400Mhz blue-and-white G3 Mac.  The latency with this setup is under 6
>milliseconds end-to-end, which is less time than it takes for the sound
>from my speakers to reach my ear. So I'm not sure if it is still worth the
>trouble, unless you want to get into compensating for the air too. These
>newer systems make me hopeful that I can eventually get rid of the 
>outboard
>compressors, distortion, etc. too.
>
>-Alex S.
>
>
>
>