[hpsdr] Announcement of a CudaSharpDSP package for HPSDR: doing parallel DSP processing on your GPU

Jeremy McDermond mcdermj at xenotropic.com
Wed May 5 14:07:46 PDT 2010


On May 1, 2010, at 1:53 AM, Hermann wrote:
> Interestingly, if I run the application with
> CudaSharpDSP, turn on AGC, Filter, and have 192 kHz sample rate, and
> 8k input buffer, 4k output buffer, the CPU still takes 30% (averged on
> the cores). Must all come from shuffling data around from Ozy to GPU
> and back, or?

My experience with MacHPSDR is that most of the processor right now is taken up by merely shuffling the USB data back and forth between userspace and the kernel.  The next biggest consumer of processor seems to be the AGC code.

> Any hints, comments and suggestions are welcome!
> 
> vy 73s,
> 
> Hermann
> 
> DL3HVH

--
Jeremy McDermond (NH6Z)
Xenotropic Systems
mcdermj at xenotropic.com




 1273093666.0


More information about the Hpsdr mailing list