On 2005-11-12 03:17, cjw1378 wrote:
... butif lower latency is better, then why is there even a ULLI control panel in Scope 4? ...
because some old farts like me don't even think to give up their even older Pulsar One cards (no, it's the other way round... ) and ride them with a funky 25ms setting.
Btw, this card cannot go less than 13ms at 44.1k and the 25ms are required by the TripleDat module.
Anwyway I wouldn't bother about 80ms either.
I monitor at the proper position and know what is time-aligned and what is not, so when exactly the thing is written to disk isn't my cup of tea.
I prefer the classic 'to tape approach' with VDAT and edit with Triple - in this context (Asio)latency doesn't show up at all
My words are of course bare nonsense if you have to produce (commercially) under economic constraints, with schedules, deadlines, data compatibility and all that stuff.
I do not consider my system a 'hobby thing', tho - my quality demands are identical to those in any commercial production, yet it's a creativity setup, not a factory alike.
Even with the fastest setup you still have to be AWARE about the timing of your signal flow, it's absolutely crucial if you stack sounds (for example).
Try it with artificially introduced errors by delaying one or the other source a few samples (up to 44, as that covers 1 ms).
You'll be amazed at the number of variations a multi-sourced sound can take by different sample delays.
Since ASIO latency (ULLI) is CPU dependent and (as written above) the exact value of the latency isn't the most important fact in fidelity context, there's this control panel that lets you trade some CPU cycles for time
cheers, tom
<font size=-1>[ This Message was edited by: astroman on 2005-11-12 05:17 ]</font>