[sdiy] Atari latency, was:Camel*ont* soft Da synth!

[Scott Gravenhorst]

Seb Francis <seb at burnit.co.uk> wrote:
>Ingo Debus wrote:
>> Hm, what *is* latency here? How is it defined?
>>
>> On a (hardware or software) synth, latency is the time between the 
>> arrival of the MIDI message and the actual starting of the sound. But 
>> we're talking about the Atari ST, thus certainly not about a soft 
>> synth, correct? What does latency mean for a MIDI sequencer? The 
>> difference between the time stamp value recorded along with a MIDI 
>> event and the actual time when this event occurred? As long as this 
>> 'latency' is constant, it could easily be compensated by the sequencer 
>> program.
>> And even if it couldn't, what's the difference between "tight timing" 
>> and "low MIDI latency" then?
>>
>
>I guess what is meant is the time between the MIDI data being received 
>by the MIDI input, passed through the sequencer (and assigned to 
>appropriate thru channel & port), and retransmitted out of the MIDI output.
>
>This seems to me the only MIDI latency that actually matters.  Anything 
>else can be easily compensated for in software as you say.  Personally 
>I've never had a problem with latency on the PC - I can't perceive any 
>delay from pressing a key on my MIDI (or USB for that matter) keyboard 
>and hearing the sound come from my analog (0 latency * :) modular.
>
>* Well ok, you could say my modular has 1ms latency as this is the time 
>taken to receive the MIDI note-on message by the MIDI->CV converter.

I concur, I use a PC, several actually, some pretty slow (200 MHz) and NONE of them gave
me what I would call "latency" problems.  Any timing problems I've ever had with MIDI
were MY OWN FAULT.  Trying to put 5 pounds of sh!t in a 2 pound bag - too much
continuous controller activity, trying to run too much stuff.  When I look at the
message volume on any one track, it's easy to see how problems like this can occur. 
It's why I use 2 MIDI interfaces with a total of 3 in/out ports so that I can balance
the load.  Splitting the load well can help use running status more efficiently.  I also
use spatial techniques, what I mean is that some instruments are intentionally delayed a
small amount to form a spatial field, more delayed instruments then sound like the are
somewhat farther away than instruments whose notes are not so delayed.  This helps to
alleviate timing problems and enhances the spatial nature of the soundscape.

IMHO, it's trying to push it too hard that causes the problems, not what computer
hardware is used to drive the system.











--------------------------------------------------------
- Where merit is not rewarded, excellence fades.
- Hydrogen is pointless without solar.
- What good are laws that only lawyers understand?
- The media's credibility should always be questioned.
- The only good terrorist is a dead terrorist.
- Governments do nothing well, save collect taxes.

-- Scott Gravenhorst | LegoManiac / Lego Trains / RIS 1.5
-- Linux Rex         | RedWebMail by RedStarWare
-- FatMan: home1.gte.net/res0658s/fatman/
-- NonFatMan: home1.gte.net/res0658s/electronics/
-- Autodidactic Master of Arcane and Hidden Knowledge.