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

Re: latency questions

My latency opinions: I can DEAL with a 256 sample buffer size, but I need
128 to be happy, and 64 feels even better but I probably wouldn't notice
32....and I'm not a great guitar player.  Professional guitar players who
I've plugged into my rig (without telling them about latency) will
immediately notice and hate 256 samples, and about half will do the same
for 128.

For keyboards you get 2 latency bonus': 1) You are not running though the
soundcard input so latency is cut in half.  2) A little extra latency
translates in your brain as the weight of the keys.  If you are very in
touch with your keyboard weighting you may not like it, but I've been able
to play at 512 and deal with it.

ASIO4All is GREAT and I have frequently used it at live performances for
output only.  It's less of an option for guitar because you need some sort
of preamp to set the input volume with - using the mixer software will
effectively reduce the bitrate.  It also ends up having a few extra
samples that it must round up to a standard buffer size, so it's hard to
get it to 64 or 128 in my experience.  And of course you need connecters
in interface a 1/8" (3.5mm)headphone output...and non pro connectors are
often not gig-worthy.

I put up a little explanation of latency at


> On the latency topic, here are my observations.
> The cheap solution: Terratec Phase26 USB.
> Works perfect and stable with an output latency of 16ms and input latency
> of
> 5ms = global latency of 21ms.
> Very seldomly clicks at 9ms/4ms = 13ms.
> Unuseable at its lowest setting (4ms/3ms = 7ms).
> The better solution: RME Digiface connected to Hammerfall CardBus.
> works perfect and stable with 2ms/2ms = 4ms.
> unuseable at its lowest setting (1ms/1ms = 2ms).
> I havent' tried optimizing my system for using ultra low latency, as
> suggested by RME. Still, 4ms is MUCH better than 21ms. You get what you
> pay
> for...;-)
> Please be advised that I did not measure the total latency of, say 
> from the A/Ds to the ADAT interface into the Digiface into the Hammerfall
> CardBus through the buffers through Live's I/O through several effects
> through Live's I/O through the buffers through the Hammerfall into the
> Digiface into the ADAT interface and out to the D/As. And this is
> bascially
> what we're after if we are looking at latency.
> Per, you mentioned you always listen to your instrument's signal before
> going through the laptop chain. Now my question: how do you make sure 
> way that you are "in time" with what comes out of your laptop if you
> actually record a loop or something? I'll try to explain what I mean.
> Let's say we have three timing systems running at the same speed but with
> different offsets.
> 1) the sound of your instrument (the input)
> 2) "the inside of your laptop" - the timing system of e.g. Mobius
> 3) what comes out of the speakers (the output)
> I think it's safe to say that 3 is later than 2 is later than 1. Let's
> call
> the respective delays d12 and d23.
> Note that there are different possibilities in which timing system you
> perceive your own playing. If you listen to the signal going through the
> converters, through the laptop and then coming back out of the speakers,
> you
> listen to it in system 3 - case A. If you monitor your
> unprocessed/converted
> signal, you listen to it in system 1 - case B.
> Case A:
> You listen to a loop playing back (in timing system 3) and play along 
> it (in timing system 1). Everything you play reaches your ear some time
> t=d12+d23 after you play it. (nb: this is not at all unusual for people
> e.g.
> playing old church organs, both from mechanical, pneumatical and acoustic
> delays). Hence, you start playing everything a little earlier, namely by
> t.
> So your playing arrives at 2 (in Mobius for an overdub) actually d23
> before
> you hear it, but that is ok because the loop happens in Mobius by
> definition
> d23 before you hear it, too. So your overdub is perfectly in sync with 
> original loop.
> Case B:
> You listen to a loop playing back (in timing system 3) and play along 
> it (in timing system 1). Everything you play reaches your ear instantly,
> so
> you don't have to "correct" anything. So something you overdub onto your
> loop takes the time d12 after you play it to get to Mobius, and Mobius
> actually played something d23 ahead of what you heard when you played 
> overdub - hence your overdub happens with a delay of t - which, with
> crappy
> hardware can be more than 20ms. (something like 32T at moderate house
> tempo).
> So while it's perfectly cool to put your "unprocessed" instrument signal
> into the mix together with any non-looper effects you might be using 
> reverb or chorus, which usually even have a predelay of their own, which
> you
> can then shorten accordingly), I would recommend against doing so when
> actually overdubbing loops.
>       Rainer