On 09/13/2007 09:27 AM, Clemens Ladisch wrote: > Rene Herman wrote: >> On 09/12/2007 05:39 PM, Clemens Ladisch wrote: >>> Rene Herman wrote: >>>> Latency-wise, I expect softvol is a non-starter always but if this is not a >>>> concern, I don't know. >>> Nothing needs to be scheduled separately; there shouldn't be any noticeable >>> latency. >> Well, there _has_ to be a difference between (at least in principle) being >> able to DMA from application to card and needing to process each individual >> sample, no? > > Yes, but the application processes each individual sample anyway, and > the processing can be done synchronously when the application uses an > ALSA function to write to the buffer or to tell ALSA that it has done > writing to the DMA buffer. > > The actual time needed to multiply each sample by some value is > negligible. Okay, that's fair enough I guess. Thanks for the comment. I'll for now believe that softvol doesn't matter all that much generally, even less than that when you do still have a master volume control, and ofcourse not at all if you only ever adjust volume externally at an amplifier anyway and disable the softvol plugin. Add in that fewer blocks in the signal path == better, and it need not even (necessarily) be considered a little lame... Rene. ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2005. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
