![]() |
![]() ![]() ![]() ![]() ![]()
|
next newest topic | next oldest topic |
Author | Topic: Why is it that DSP load increases so much |
mathis Member |
![]() ![]() ![]() when working on high sample rates? I mean the difference between 96k and 48k is about four times DSP load. In my small brain I could understand doubled DSP load but not four times. So how come? Where to post a question like this in the TWiki? Bests, IP: Logged |
SSC Administrator |
![]() ![]() ![]() The increase in processor usage grows in a nonlinear way with the sample rate. One reason is that each processor has to be running a minimal program at least to get samples in from its neighbor and pass samples along to the next neighbor. This program is a fixed length so for low SRs, it is a smaller percentage of the total available processing time than it is for high SRs. Another reason is the "granularity" of Sounds. Most Sounds can be split into several parts and scheduled across multiple processors, but there is a smallest block of code that cannot be split (the microSound). As the SR goes up, fewer microSounds can be scheduled on a single processor, but since there is a "quantum" size, there is some point at which microSound that used to fit on the same processor with another microSound will suddenly have to be scheduled on its own processor. Also, some microSounds require memory and that further complicates the scheduling process. IP: Logged |
mathis Member |
![]() ![]() ![]() OK, I see. Thanks for the explanation! IP: Logged |
All times are CT (US) | next newest topic | next oldest topic |
![]() ![]() |
This forum is provided solely for the support and edification of the customers of Symbolic Sound Corporation.