r_antialias 0 -- No antialiasing
r_antialias 1 -- Only the level is antialiased (walls, floor, etc.)
r_antialias 2 -- Weapons antialiased
r_antialias 3 -- Level, weapons, and monsters antialiased
r_antialias 4 -- Particles antialiased
r_antialias 5 -- Particles and level antialiased
r_antialias 6 -- Monsters, weapons, and particles antialiased
r_antialias 7 -- Monsters, level, weapons, particles antialiased (The Works!)
And my scores are:
Note: The percentage of the maximum framerate (at r_antialias 0) is given
in parentheses after each score
r_antialias timerefresh (at start position) timedemo demo2
=======================================================================
0 21.8 (100%) 21.8 (100%)
1 19.8 (90.8%) 20.3 (93.1%)
2 20.3 (93.1%) 21.3 (97.7%)
3 18.5 (84.9%) 20.0 (91.7%)
4 21.2 (97.2%) 21.7 (99.5%)
5 19.8 (90.8%) 20.3 (93.1%)
6 20.4 (93.6%) 21.3 (97.7%)
7 18.4 (84.4%) 19.9 (91.3%)
========================================================================
The tests were all run from dos 6.22, without emm386 loaded. The
timerefresh scores were run twice, and their results averaged. Since it
took a good 45 seconds or so, I ran each timedemo only once.
I hope all this answers some questions, or at least gives people an idea
of performance. Mail me if you have any questions (real questions
please, not "How do I use a config file?").
Also, I have never seen anything besides screenshots of GLQuake, so
please don't bug me about that, or drag me into flame wars. :)
Michael J. Raiter "I would rather be an optimist
Physics major at Saint Olaf College and a fool than a pessimist
E-Mail:rai...@stolaf.edu and right"
http://www.stolaf.edu/people/raiter -Albert Einstein
I got 19.2 frames in demo2.
I ran VQuake from Win95 and I have a Pent133, 24meg, Tyan Tomcat m/b,
and Intergraph Reactor. Overall I saw a dramatic difference in the
antialias effects with very little difference in framerate.
--
John Hope | City and Regional Planning
jh...@pop.calpoly.edu | Cal Poly San Luis Obispo
http://www.netcom.com/~gtiracer | http://www.calpoly.edu/~crp
>And my scores are:
> The tests were all run from dos 6.22, without emm386 loaded. The
>timerefresh scores were run twice, and their results averaged. Since it
>took a good 45 seconds or so, I ran each timedemo only once.
> I hope all this answers some questions, or at least gives people an idea
>of performance. Mail me if you have any questions (real questions
>please, not "How do I use a config file?").
> Also, I have never seen anything besides screenshots of GLQuake, so
>please don't bug me about that, or drag me into flame wars. :)
>Michael J. Raiter "I would rather be an optimist
>Physics major at Saint Olaf College and a fool than a pessimist
>E-Mail:rai...@stolaf.edu and right"
>http://www.stolaf.edu/people/raiter -Albert Einstein
Thanks for the framerates!
>And my scores are:
Those numbers seem awfully low. On my system (P166, 430fx, 32mb FPM
RAM, 256k pipeline burst cache, Canopus Total3D) in Timedemo 2, I'm
getting 24.8 fps @640x480 full screen, no AA, and 24.7 fps @640x480
full screen, r_antialias 4 (round particles). I see virtually no
performance hit with round particles enabled.
Try increasing your monitor's refresh rate. In my experience with
Vquake, I get an extra 1.5-2 fps when going from 60Hz to 75Hz or
greater.
Ed
These are almost identical to what I get with a iPentium75 (oc'd to 100 at
1.5 x 66)+ or - 2%. My only question is how to increase the refresh rate
of the monitor? By the way, why in the world did this game not have round
particles in the first place....it is an extremely nice effect.
TIA,
Chuck Putman
> > Hi all! I just downloaded the last vquake, and wanted to know how
much
> >the anti-aliased particles affected performance. I also noticed how no
> >one had posted any scores yet, so I thought I'd post mine.
> > I few words about how I tested it first: I tested the framerate with
> >the settings I normally play with.
<snip>
>
> >r_antialias timerefresh (at start position) timedemo demo2
> >=======================================================================
> > 0 21.8 (100%) 21.8 (100%)
> > 1 19.8 (90.8%) 20.3 (93.1%)
> > 2 20.3 (93.1%) 21.3 (97.7%)
> > 3 18.5 (84.9%) 20.0 (91.7%)
> > 4 21.2 (97.2%) 21.7 (99.5%)
> > 5 19.8 (90.8%) 20.3 (93.1%)
> > 6 20.4 (93.6%) 21.3 (97.7%)
> > 7 18.4 (84.4%) 19.9 (91.3%)
<snip>
> I ran VQuake from Win95 and I have a Pent133, 24meg, Tyan Tomcat m/b,
> and Intergraph Reactor. Overall I saw a dramatic difference in the
> antialias effects with very little difference in framerate.
I agree, the AA in this new version is great. The new levels are great
also.
--
Don Biz
don...@ix.netcom.com
Ed wrote in article <5fopkf$eg...@integrix.support.nl>...
> Mike Raiter <rai...@stolaf.edu> wrote:
> Those numbers seem awfully low. On my system (P166, 430fx, 32mb FPM
> RAM, 256k pipeline burst cache, Canopus Total3D) in Timedemo 2, I'm
> getting 24.8 fps @640x480 full screen, no AA, and 24.7 fps @640x480
> full screen, r_antialias 4 (round particles). I see virtually no
> performance hit with round particles enabled.
You have a Canopus board, take note. Due to the faster memory and the fact
they've overlocked the Verite slightly, it's anywhere from 10-30% faster
than other Verite boards.
It's depend upon your card utilities what refresh rate you can use in
dos.
I play quake at 160Hz! (Millenium+Viewsonic 17ps+640*480)