Jonathan Morton wrote:
>
> >well i gave up on XFRee 4.0.1 cause after changing about every BIOS
> >setting, recompiling X with several mentioned improvements, switching
> >video cards, switching RTAI->RTL and aback again, selectively disabling
> >XAA parts, i still
> >had peaks of 1msec and more in mmy latency tests.
> >
> >So now i got XFree 3.3.6 (standard RH6.2 rpms) again. First thing i
> >noticed on the G400 is that it is slower than
> >XFree 4.0.1 (altough faster than XFree 4.0.1 without acceleration). The
> >latency is down to its normal level of 20usec
> >peaks and about 7usec normal. (i will try the s3Virge later , cause than
> >i have to rebuild my comp again :-/)
> >
> >I asume that the 3.3.x serie will not be developed on anymore so
> >sommetime next year users will be getting linux
> >distributions with XFree 4 on it and new cards won't be supported
> >anymore by XFree 3. So i think all ppl using
> >RT-linux/RTAI will have to be very carefull when switching to XFRee 4.
> >Also i hope some other ppl are willing
> >to try if they have the same problems with XFree 4.0.1 as i have.
> >(Michael did you tried X3 or X4 with that
> >nvidia card ? ).
> >
> >Well thanx to everybody that helped, and gave advice and so on.
> >
> >I have the bad feeling this won't be the last we hear from XFree 4:-/
>
> Perhaps the RT/Linux people would like to develop some guidelines on how to
> avoid latency in their setups. If this involves avoiding hardware with
> high-latency characteristics (including but not limited to video cards...),
> I am sure most RT users would understand. For those types of cards which
> are not inherently high-latency, perhaps a special mode could be introduced
> in v4.x drivers which avoids activating features of the card which
> introduce latency.
>
The design of XFree 4.x would make it rather easy the make patched
drivers and
distributing them. I am sure the XFree ppl feel the same about Realtime
as
the Linus does "won't be standard but you can patch what you like, and
if it ends
up being so good that i like it it might get part of the standard
distribution :-)"
> I'm not a RealTime user myself, but this seems like the most sensible path
> to me. After all, the goal of mainstream chip manufacturers is to increase
> 'normal' performance, which normally does not involve low-level latency
> measurements.
Yes ofcourse, on the other hand there end up being more and more
multimedia applications
to that want low latency, like audio/video player software, and having
latency upto 50msec
will mess things up.
>
> Another point: I heard various things about the Virge/DX series chips being
> incompatible with things like video-capture devices - perhaps this is
> related to the latency problems, or is simply indicative of overall bad
> design.
Hmmm can you say "crap" :-) i know that the s3 card is a bad one, it was
a temp
solution until i buy a new one. But the problem with the G400 bothers me
more, cause that
"should" be a good card, and one of the few multihead cards.
- Erwin
>
> --------------------------------------------------------------
> from: Jonathan "Chromatix" Morton
> mail: [EMAIL PROTECTED] (not for attachments)
> big-mail: [EMAIL PROTECTED]
> uni-mail: [EMAIL PROTECTED]
>
> The key to knowledge is not to rely on people to teach you it.
>
> Get VNC Server for Macintosh from http://www.chromatix.uklinux.net/vnc/
>
> -----BEGIN GEEK CODE BLOCK-----
> Version 3.12
> GCS$/E/S dpu(!) s:- a19 C+++ UL++ P L+++ E W+ N- o? K? w--- O-- M++$ V? PS
> PE- Y+ PGP++ t- 5- X- R !tv b++ DI+++ D G e+ h+ r- y+
> -----END GEEK CODE BLOCK-----
-- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
---
For more information on Real-Time Linux see:
http://www.rtlinux.org/rtlinux/
