General bad performance for XF86_SVGA?

General bad performance for XF86_SVGA?

Post by Rainer Menzne » Wed, 23 Dec 1998 04:00:00



Hello Linux people,

I have a notebook using the * 9397 chip from Trident and a desktop
PC using S3Virge graphics card, both with 4MByte graphics memory. Both
systems run SuSE Linux 5.3 and the new XF86_SVGA (3.3.3), as binary
distributed by the XFree-project.

On both systems the server performs really bad. Two examples:

1) If I drag windows on a KDE desktop it takes a seconds that
   window follows the mouse / touchpad cursor. I use transparent mode
   for dragging windows

2) If I logout out under KDE it takes seconds until the display is
   shaded out and the confirmation window appears.

Both 1) and  2) work with an instantaneous feel if I run my desktop system
using XF86_S3V (which has other problems, though).

I know the acceleration code for the * 9397 has been disabled
hardcoded in contrast to what is said in the docs, but has the
acceleration also been disabled for the S3Virge?

And are there precompiled servers where the acceleration has not been
disabled? I assume it's not much fun to compile an X-server.

Thanks and best regards,

-Rainer

--

*** ____ ****** . * . ******* Dipl.-Ing. Rainer Menzner ********************
   ( /  \      /|  /|         Ruhr-Universitaet Bochum  
    /    |    / | / |         Institut fuer Neuroinformatik
   /____/    /  |/  |  __     D-44780 Bochum, Germany
  / \       /   '   | ( /     ----------------------------------------------
(/   \_ o (/        | -/- o  
********************* /--) ** Tel. +49-234/700-7978 ************************


WWW:   http://www.veryComputer.com/

 
 
 

General bad performance for XF86_SVGA?

Post by Antoni Lapinsk » Wed, 23 Dec 1998 04:00:00



> Hello Linux people,

> I have a notebook using the * 9397 chip from Trident and a desktop
> PC using S3Virge graphics card, both with 4MByte graphics memory. Both
> systems run SuSE Linux 5.3 and the new XF86_SVGA (3.3.3), as binary
> distributed by the XFree-project.

[snip]

Quote:

> On both systems the server performs really bad. Two examples:
> I know the acceleration code for the * 9397 has been disabled
> hardcoded in contrast to what is said in the docs, but has the
> acceleration also been disabled for the S3Virge?

> And are there precompiled servers where the acceleration has not been
> disabled? I assume it's not much fun to compile an X-server.

You could always go with a commercial server from www.xig.com.
Just download a demo from their site and see whether it's worth
the money.

-Antoni

 
 
 

General bad performance for XF86_SVGA?

Post by Manuel Munie » Wed, 23 Dec 1998 04:00:00


I just install the new XF86_SVGA (3.3.3) for use with my
S3 Virge MX. When I start X, this server enable many harcoded
features (XAA acceleration) by default. Using a "good old"
fvwm95 I get good performances with a PII 233.

Rainer Menzner a crit:

Quote:

> Hello Linux people,

> I have a notebook using the * 9397 chip from Trident and a desktop
> PC using S3Virge graphics card, both with 4MByte graphics memory. Both
> systems run SuSE Linux 5.3 and the new XF86_SVGA (3.3.3), as binary
> distributed by the XFree-project.

> On both systems the server performs really bad. Two examples:

> 1) If I drag windows on a KDE desktop it takes a seconds that
>    window follows the mouse / touchpad cursor. I use transparent mode
>    for dragging windows

> 2) If I logout out under KDE it takes seconds until the display is
>    shaded out and the confirmation window appears.

> Both 1) and  2) work with an instantaneous feel if I run my desktop system
> using XF86_S3V (which has other problems, though).

> I know the acceleration code for the * 9397 has been disabled
> hardcoded in contrast to what is said in the docs, but has the
> acceleration also been disabled for the S3Virge?

> And are there precompiled servers where the acceleration has not been
> disabled? I assume it's not much fun to compile an X-server.

> Thanks and best regards,

> -Rainer

 
 
 

General bad performance for XF86_SVGA?

Post by Manuel Munie » Wed, 23 Dec 1998 04:00:00


I just install the new XF86_SVGA (3.3.3) for use with my
S3 Virge MX. When I start X, this server enable many harcoded
features (XAA acceleration) by default. Using a "good old"
fvwm95 I get good performances with a PII 233.

Manu
--
Rainer Menzner a crit:

Quote:

> Hello Linux people,

> I have a notebook using the * 9397 chip from Trident and a desktop
> PC using S3Virge graphics card, both with 4MByte graphics memory. Both
> systems run SuSE Linux 5.3 and the new XF86_SVGA (3.3.3), as binary
> distributed by the XFree-project.

> On both systems the server performs really bad. Two examples:

> 1) If I drag windows on a KDE desktop it takes a seconds that
>    window follows the mouse / touchpad cursor. I use transparent mode
>    for dragging windows

> 2) If I logout out under KDE it takes seconds until the display is
>    shaded out and the confirmation window appears.

> Both 1) and  2) work with an instantaneous feel if I run my desktop system
> using XF86_S3V (which has other problems, though).

> I know the acceleration code for the * 9397 has been disabled
> hardcoded in contrast to what is said in the docs, but has the
> acceleration also been disabled for the S3Virge?

> And are there precompiled servers where the acceleration has not been
> disabled? I assume it's not much fun to compile an X-server.

> Thanks and best regards,

> -Rainer

 
 
 

1. RedHat + XF86_SVGA + TM5200 != Slackw + XF86_SVGA + TM5200

Ok, it's a silly subject but the point is this:

using the same XF86Config and the same machine, XF86_SVGA from Red Hat
does not work but XF86_SVGA from Slackware does.

I did an "X -configured" on both binaries and it seems that they are
both based on the same XF86 code, but the binaries are quite
different.

In any case, I copied the X binary from Slackware and it looks fine.

I'd like to get this working with the "distribution" setup though,
perhaps there is another problem I'm missing.

Bill
--
William York

2. pmax install media

3. XF86_SVGA -quiet fails to find PCI base address, XF86_SVGA works

4. Need help width shell script

5. Bad XF86_SVGA server :( shame on you Linux :)

6. 3CXSH572BT 3Com OfficeConnect 10/100 LAN PC Card

7. Symposium: CETS-2001 - Rich Marcello, Vice President and General Manager, Compaq High Performance Systems Division

8. UNIX Capacity Planning workshop

9. new color xterm and emacs -- also improves general X performance

10. Performance Issues and General Advice

11. General performance problem

12. Drive performance in general

13. C App Design, performance & general