[
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]
[ General ] [ Speakers ] [ Tubes ] [ Vinyl ] [ Digital ] [ Hi-Rez ] [ Video Asylum ] [ Cables ] [ Tweaks/DIY ] [ Music ] [ Films ]
65.19.76.104
In Reply to: RE: Design goal is to minimize mobo traffic to just the audio samples posted by cics on July 10, 2008 at 20:17:53
"cMP, cMP² and other players that load into RAM before playback would eliminate this conflict as no wav data traffic occurs. It however doesn't end here as some devices and applications have a polling nature. This would create such conflict and is audible."
Aren't hardware interfaces that send the audio samples out of the box buffered? Assuming the samples make it across the PCI bus to the interface before the buffers underrun, why would a slight delay affect the sound? Do you have a physical mechanism in mind and, if so, have you been able to verify this?
(I am talking about the occasional delay, not the situation when there is lots of conflict for the PCI bus in which catchup from small delays does not happen and the buffer fill level gradually depletes.)
"This all happens at frantic speeds but for ultra low latency playback, this conflict will have impacts."
A related question is why the need for ultra low latency playback? Given that some processing or data movement has to be done, why does it matter if it is done earlier rather than later? (I am talking about processing that is done while the music is playing, not processing that could be moved before playback starts, such as loading RAM.) I can understand the need for low latency in pro-audio applications where additional tracks are being laid down by musicians who have to stay in sync, but why does this matter for end-user playback?
Tony Lauck
"Perception, inference and authority are the valid sources of knowledge" - P.R. Sarkar
Follow Ups:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2008, Audio Asylum, all rights reserved ]