[PD-ot] how low (latency) can you go?

[Hans-Christoph Steiner]

On Dec 16, 2006, at 6:18 PM, IOhannes m zmoelnig wrote:

> hi.
>
> Hans-Christoph Steiner wrote:
>>
>>>
>>> as always when it comes to latency and people tell numbers, i  
>>> have to
>>> ask my question: have you measured this? how? or was it just  
>>> "putting a
>>> number somewhere and assuming that it's the real latency"?
>>> how much load do your settings allow?
>>
>> Yes, its true, I am too lazy to do an actual measurement.  I am just
>> talking about the buffer setting.
>
> thank you for this clarification.
> i think it is important to have such things articulated correctly.
> as soon as the word "latency" can be found on the web, i am pretty  
> sure
> that a lot of myth and rumour are involved.
> people keep telling that they can go as low as 3ms in pd, which is
> simply not possible.
>
>> For what I do, I don't really need
>> much lower.
>
> this is fine for you.
>
>>
>> And always, when talking about latency, I feel the need to point  
>> out the
>> speed of sound: 340 m/s or .34 m/ms.  So if your speakers are 2m away
>> from you, that's 6ms of latency.  You could spend days tweaking your
>> machine to get 3ms less latency, or you could move 1m closer to your
>> speakers.  Puts things into perspective...
>
> this is correct.
> but unfortunately perspectives can change. as a musician you might be
> happy with one machine adding latency of 3ms.
> however, it is common practice to stack several machines  
> (synthesizers,
> effect-hardware, PCs), all adding a small latency. if you have 3
> machines, all with a perfectly low latency of 10ms, you get a  
> resulting
> latency of 30ms which might be intolerably low.

> that is just music. as soon as you get into technical applications,
> things get far worse (e.g. there might really be a need to get as  
> low as
> 2ms).

Just out of curiosity, what kind of applications need this?



> so again: it is my concern to get clarity when people use the term
> "latency".
> i have done a lot of latency measurements over the last years; usually
> the numbers i saw in various control panels (be it jack, pd, rme
> hammerfall's settings,...) are not identical with the actual latency;
> though the often (not always) directly related.
>
> mf.asdr
> IOhannes
>
> PS: thanks to all of you for your explanations about latency; my  
> remark
> was really not about latency (what it is; what you need) but about  
> what
> we think when we hear the buzzword "latency".

Yeah, that is also a necessary clarification.  "latency" is a simple  
number that people can use to sell things.  I think that's why it is  
such a buzzword.

.hc

------------------------------------------------------------------------

Computer science is no more related to the computer than astronomy is  
related to the telescope.      -Edsger Dykstra