rec.autos.simulators

I think we have been totally confusing people out there with our video
score reports.  We are mixing apples, oranges & bananas, with a few
doughnuts thrown in besides.  What I propose is that we stop using GP2's
logon text and subjective fps/CPU% and use Chris' 3d Benchmark programs.
This way we can all be on the same page and make some sense out of this
mess.  I am posting the suite of benchmarks in
alt.binaries.simulators.autos under the heading
        Chris' 3d Benchmark

I have also made the program available via anonymous ftp at
        ftp://ftp.teleport.com/users/mcarver/cbench.zip

Please run the test in VGA mode (bench.exe) and in SVGA mode
(svgabnch.exe) and report your results.  Hopefully, this way people will
be able to make an informed decision on upgrading their system (if
needed).

Okay, my results:

Gateway2000 P5-90 (Neptune Chip Set 2) with 16meg memory
BIOSName = American Megatrends
BIOSVersion = BIOS Version 1.00.12.AX1T
Video Card: ATI PCI SuperVGA (Mach64 88800GX-D) with 2 Mb VRAM
           (OEM version)
RAM DAC:    ATI68860 24 bit DAC
Clock Chip: ATI 18818-1 Clock

CBench Score     Frames/sec  
VGA    93.7      56.2
VGA    93.8      56.3 (with mergeon.exe)

SVGA   28.3      17.0
SVGA   39.0      23.4 (with SDD52 or m64vbe2.0 w/acc -4 switches)

--
**************************** Michael E. Carver *************************
     Upside out, or inside down...False alarm the only game in town.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=

BENCH:          VGA     342.9 score
                        205.7 frames per second

SVGABNCH:       SVGA    77.0 score
                        46.2 frames per second

JEB in Vegas

 >> Gateway2000 P5-90 (Neptune Chip Set 2) with 16meg memory
 >> BIOSName = American Megatrends
 >> BIOSVersion = BIOS Version 1.00.12.AX1T
 >> Video Card: ATI PCI SuperVGA (Mach64 88800GX-D) with 2 Mb VRAM
 >>            (OEM version)
 >> RAM DAC:    ATI68860 24 bit DAC
 >> Clock Chip: ATI 18818-1 Clock
 >>
 >> CBench Score     Frames/sec
 >> VGA    93.7      56.2
 >> VGA    93.8      56.3 (with mergeon.exe)
 >>
 >> SVGA   28.3      17.0
 >> SVGA   39.0      23.4 (with SDD52 or m64vbe2.0 w/acc -4 switches)
 >>
 >> --
 >> **************************** Michael E. Carver *************************
 >>      Upside out, or inside down...False alarm the only game in town.

 >> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=
 >I ran the bench.exe and the svgabnch.exe on my P6/200
 >and it resulted in the following scores.
 >
 >BENCH:          VGA     342.9 score
 >                        205.7 frames per second
 >
 >SVGABNCH:       SVGA    77.0 score
 >                        46.2 frames per second
 >
 >JEB in Vegas

My results:
VGA     210.3
SVGA    61.0

Intel P54C/166MHz, Matrox Millenium

Clark

Good call. Here are my results.

P5-150 oc'd to 166 24megs ram + Lightspeed 128 w/2.25megs

VGA  SCORE=200.3  fps=120.1
SVGA SCORE=62.4   fps=37.4

Hey JEB if you are out there can you run this with and without fastvid and give
us the results? I am seriously considering getting the CPU and Mobo but want to
be certain.

_____________________________________

: >
: > I think we have been totally confusing people out there with our video
: > score reports.  We are mixing apples, oranges & bananas, with a few
: > doughnuts thrown in besides.  What I propose is that we stop using GP2's
: > logon text and subjective fps/CPU% and use Chris' 3d Benchmark programs.
: > This way we can all be on the same page and make some sense out of this
: > mess.

: Since this newsgroup is partly about GP2 and how GP2 utilises your video
: and CPU power, I think it is most suitable to use logon files. Although
: a new
: reader of this newsgroup, I don't think the idea of this gourp is to
: brag
: with what results I get with Chris' 3D BM prog, but to figure out the
: best
: hardware setup for GP2.

The point is to brag.  The point is that the logon file does not seem to
truly reflect performance in the game.  Everyone is using different
levels of graphics and have set their frame rate to different levels
within GP2.  Therefore posting this information is biased and purely
subjective.  There are many people buying video cards based on this
information only to find little or no improvement.  The point of using
an outside benchmark program is to shed a little objective light on the
subject.  

Since GP2 doesn't use vesa 2.0 extensions, running the 3d Benchmark
without sdd52 or other vesa 2.0 tsr's does show how well the video card
peforms and puts the subject on an objective basis.  Based on this
information and information on an individuals card/motherboard/cpu, we
can make valid decisions on how to get good hardware setups.  Not only
for GP2 but other applications.

--
**************************** Michael E. Carver *************************
     Upside out, or inside down...False alarm the only game in town.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=

I am sorry, my earlier reply had a typo in it that totally
misrepresented what I meant to say.  I have deleted, but if you read it
already...  The point is NOT to brag!

: >
: > I think we have been totally confusing people out there with our video
: > score reports.  We are mixing apples, oranges & bananas, with a few
: > doughnuts thrown in besides.  What I propose is that we stop using GP2's
: > logon text and subjective fps/CPU% and use Chris' 3d Benchmark programs.
: > This way we can all be on the same page and make some sense out of this
: > mess.

: Since this newsgroup is partly about GP2 and how GP2 utilises your video
: and CPU power, I think it is most suitable to use logon files. Although
: a new
: reader of this newsgroup, I don't think the idea of this gourp is to
: brag
: with what results I get with Chris' 3D BM prog, but to figure out the
: best
: hardware setup for GP2.

The point isn't to brag.  The point is that the logon file does not seem
to truly reflect performance in the game.  Everyone is using different
levels of graphics and have set their frame rate to different levels
within GP2.  Therefore posting this information is biased and purely
subjective.  There are many people buying video cards based on this
information only to find little or no improvement.  The point of using
an outside benchmark program is to shed a little objective light on the
subject.  

Since GP2 doesn't use vesa 2.0 extensions, running the 3d Benchmark
without sdd52 or other vesa 2.0 tsr's does show how well the video card
peforms and puts the subject on an objective basis.  Based on this
information and information on an individuals card/motherboard/cpu, we
can make valid decisions on how to get good hardware setups.  Not only
for GP2 but other applications.

--
**************************** Michael E. Carver *************************
     Upside out, or inside down...False alarm the only game in town.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=

And to figure out which hardware setup is best, you need to know the
performance of each - you need a benchmark.

A couple of weeks ago I wrote a piece for next month's Sim Racing News
about CBench, because I reckon it's a great program, and which I've used
for a couple of months now. It works well with sims because it is based
upon SVGA texture mapped 3D polygons, exactly what we use.

I would suggest though that rather than post results to the newsgroup,
they should be posted to the WWW page. It cuts down traffic here, and it
allows non-sim fans to see and take profit of our experience.

Cheers!
John

                     _________________________________
         __    _____|                                 |_____    __
________|  |__|    :|           John Wallace          |     |__|  |________

  \    :|  |::|    :|        Team WW Racing TSW       |     |::|  |     /
    >  :|  |::|    :|_________________________________|     |::|  |   <
  /    :|__|::|____/       * Sim Racing News *         \____|::|__|     \
/______:/  \::/ http://sneezy.dcn.ed.ac.uk/simnews/index.htm \::/  \._____\
               http://www.math.ohio-state.edu/~harmon/simnews

  Can anyone tell me where I can find some of these benchmark
programs, particularily CBench.

thanks
russ

Gateway2000 P5-90 (Neptune Chip Set 2) with 16meg memory
CBench Score     Frames/sec
VGA   242.2     145.3
VGA   340.3     204.2 (with fastvid 011 4)

SVGA   46.9     204.2
SVGA   77.7      46.6 (with fastvid 011 4)

The "real benchmark" in this case (GP2) gets:

All track side detail:                          With FASTVID:
Without FASTVID
ALL forward objects, mirrors as forward:        Estimated FR: 14.2
13.4
ALL forward objects, mirrors (no texture):      Estimated FR: 15.0
15.0

Probably, Cris's 3D benchmark is a better  measure of relative FPS
speed than the GP2LOG since the PPro200 GP2LOG video score with
fastvid is over 600 compared to 108 for a P90 with Mach64. Certainly
not 6 times the video speed.

Thanks for suggesting to use a more defined test Michael.

VGA :   208.3  
        125.0 Frames per second

SVGA:   38.4
        22.8  Frames per second

As you can see the Write Combining and Linear
Frame buffering being enabled makes quite a
difference.

JEB

I agree that cbench is a much better way to check for hardware speed and
that it's important not to use Vesa 2.0. The problem I see is for Matrox
Millenium, ET6000 cards, and new cards coming out with on board support
for VESA 2.0. Unless it's possible to disable VESA 2.0 support on these
cards, the benchmark numbers won't relate to GP2 video performance.

Claude

: A couple of weeks ago I wrote a piece for next month's Sim Racing News
: about CBench, because I reckon it's a great program, and which I've used
: for a couple of months now. It works well with sims because it is based
: upon SVGA texture mapped 3D polygons, exactly what we use.

: I would suggest though that rather than post results to the newsgroup,
: they should be posted to the WWW page. It cuts down traffic here, and it
: allows non-sim fans to see and take profit of our experience.

http://www.dfw.net/~sdw/
contains many benchmark results including CBench.

My machine Pentium 150 overclocked to 166mhz, 16mb RAM, Triton II chipset
on AFX Motherboard with Diamond Stealth Video 2001 card with 2mb RAM

Still don't run sky and track in GP2 even less detail at monaco.

: The results I posted were with FASTVID enabled.
: Without FASTVID the scores are:

: VGA : 208.3  
:       125.0 Frames per second

: SVGA: 38.4
:       22.8  Frames per second

: As you can see the Write Combining and Linear
: Frame buffering being enabled makes quite a
: difference.

Looks like we need Geoffr 'n team to go back to the drawing board and
fix GP2 to use Linear Frame Buffering and R/W's!!!!!  Should make a hell
of a difference.

--
**************************** Michael E. Carver *************************
     Upside out, or inside down...False alarm the only game in town.

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=<[ /./.  [-  < ]>=-=-=-=-=-=-=-=-=-=-=-=-=-=