- From: Noah Mendelsohn <nrm@arcanedomain.com>
- Date: Thu, 08 Aug 2013 08:08:42 -0400
- To: Marcus Geelnard <mage@opera.com>
- CC: Jer Noble <jer.noble@apple.com>, "K. Gadd" <kg@luminance.org>, Srikumar Karaikudi Subramanian <srikumarks@gmail.com>, Chris Wilson <cwilso@google.com>, Alex Russell <slightlyoff@google.com>, Anne van Kesteren <annevk@annevk.nl>, Olivier Thereaux <Olivier.Thereaux@bbc.co.uk>, "robert@ocallahan.org" <robert@ocallahan.org>, "public-audio@w3.org" <public-audio@w3.org>, www-tag@w3.org
On 8/8/2013 5:29 AM, Marcus Geelnard wrote: > I think that most people on this list have agreed that the time/speed > implications of a memcpy is not really an issue. OK, great. If that's really true and it's been thought through, then you've already done what I was suggesting. I'm always suspicious when streaming media interfaces involve copying, but if it's been added up and not just the single program performance but the impact on other applications has been evaluated, then that's terrific. (It's easy to build systems that run just fine if nothing else is happening on the system, but that take more than a reasonable share of memory bandwidth.) Noah
Received on Thursday, 8 August 2013 12:09:01 UTC