View Single Post
  #18  
Old 2013-02-09, 23:09
joshuajohn89 joshuajohn89 is offline
 
Join Date: Dec 2009
Posts: 341
Quote:
Originally Posted by SteveDiverse View Post
...

In order to prevent the sound from stopping and automatically stepping down the rate (from say 96k to 88k) the computer would have to 'look ahead', which would mean calculating a buffer of some time duration (perhaps only a few milliseconds) and then playing the buffer back 'behind' the calculations.

if the buffer length shrinks, it means that playback is catching up with the calculations, which means the calculation is starting to fall behind (or already has fallen behind) which would trigger the step down in sample rate.

...

With a live or interactive situation like Reason, you would notice it because you'd start triggering sounds with your controller, the calculations would be done in real time, but the playback would lag by that delay that is the buffer.
Would this buffer be required in addition to the buffer already in place? Could not a system work wherein the processor steps down the sampling rate (or switches 'playheads', rather) when it detects that the (current) buffer is dangerously depleted and/or is being depleted at an unsustainable pace?

I imagine the setup being something akin to playing the same project on two different computers running at two different rates while identical messages from the controller are sent to both rigs, while playback from only one of the pair is heard at any time. The default player would be that of the higher sample rate, but would switch to the second whenever it is detected that the primary is running too hot. The only difference, then, between such a setup & my interpretation of cyclic's suggestion is that the latter would be managed internally by a single computer. The multiple cores of a modern workstation are essentially multiple computers working together, are they not?

It seems to me, though, you are saying that if I possess processing power sufficient to run my system at 96 & 88kHz simultaneously -effectively a sample rate of 184kHz- then I might as well just run at 184k alone, making the fail-safe needlessly redundant. Whether I use dual rates or a single rate less than or equal to the sum of the two makes no difference. Is this right?