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

Re: latency (was: Ebay echoplexi...)

>At 70 degrees F the speed of sound is 1129.5 ft per sec.  So figure 
>1.1 ft per millisecond.  A device like the Repeater, with a 11 msec 
>latency will sound as if it's 12.1 feet away from you.  The fact 
>that you don't hear an electric guitar from the source means the 
>guitar has a natural latency based on the fact that the sound source 
>isn't in your hands.  So add another 5 feed in my studio, more when 
>I play out.  My mind seems to instantly compensate for this.  I 
>wonder what makes some people so much more sensitive to this than 
>others?  Maybe you get used to it.  Not a single instrument I own 
>has 0 latency due to digital conversion or MIDI latency.  The plus 
>side is it makes my studio seem bigger to me in a subliminal way, I 
>guess.  People have always told me that I'm not "all there" so maybe 
>that's it.  I'm over there a bit.

Important remark mark mark...

I guess it depends a lot on the kind of music
Thomas Diethelm used to say that he could not play with his speaker 
more than 50cm away from his head. I believe him because he did very 
quick and precise rhythmic music based on short loops

also, one thing is to hear yourself while you play (no loops),
another is to hear the loop with the same latency,
a third is to hear the loop with no latency, and thus the actual 
playing not where it will end up in the loop.
Did anyone investigate where the problem really is?

in praxis with a laptop you get more latency for the direct signal 
than for the loop, being that the one of the loop can be adjusted, 
depending on the software we use...

>Natural and unnatural latency will always exist. It's important to 
>this latency where possible. I think that analog pedals will always have a
>minimal amount of latency compared to digital pedals. I haven't measured
>that though. I should try that out sometime. We all need to be aware that
>the more digital processing we do to our signal, the more latency will
>exist. Effects manufacturers aren't likely to advertise this or even make 
>obvious in their specs. It's a good thing to keep in mind though.

analog delay is so short it never was an issue.
It may be interesting to have a few miliseconds of delay on some 
frequencies, while only the treble range is important for timing 

simple digital effect machines, be it the old delays that did not 
process but just store sound, or the ones with DSP, have extremely 
low latencies of a few samples.
More modern converters added buffers with like 16 samples to the DSP 
machines, still below 1ms anyway. But yes, if you go through 15 of 
them, you may notice... %-/

latency only becomes an issue where the sound goes through a complex 
OS or a comunication system like Ethernet.

>On Jun 12, 2004, at 11:53 PM, David J. Grossman wrote:
>>In theory, the idea of a digitally processed and modelled guitar signal 
>is a
>>great idea. In reality, I feel that any significant latency in the 
>signal is
>>unacceptable for my purposes. Especially since, as a bass player, I 
>>to nail the beat and be in time with my drummer. That is the primary 
>>I've stayed away from the V-Bass. Before I buy one, I'd have to be 100% 
>>that the latency is completely inaudible.
>>- Dave


          ---> http://Matthias.Grob.org