Timothy Normand Miller wrote:
On 4/19/07, James Richard Tyrer <[EMAIL PROTECTED]> wrote:
So, it would be much better if the sync interrupt could be serviced
with RT.
Do you have examples of people complaining about this in the past?
Only on 386s with DOS. :-)
I've been working with graphics cards and drivers for a long time, and
only in one extreme case did we find that there wasn't enough blanking
time to get done all that we needed to do. And that was truly an
unusual situation.
I don't know of any specific instances and I only have a 1024 x 768
display running at 70 Hz (which would be V blank of about 2 ms -- IIUC V
blank goes down as a % of field time as the number oh H lines increases)
but I have AGP which I think means a dedicated interrupt. My old system
isn't really fast enough to do 30 fps video so I wouldn't notice if the
interrupt latency was an issue anyhow.
I think that this problem would only show up on a PCI video card and at
high bit rate video playing. We are talking about both, but it may not
be an issue. And, I didn't realize that even new PCs are short on
hardware interrupts. One more reason that someone should make a PC
designed to run *NIX.
--
JRT
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
