UT2003 Demo Benchmark info

[Gargr]

http://www.rage3d.com/board/showthread.php?s=&threadid=33635900
"Brent
[H]ard|Video|Gawd

I am sure this has been posted, but i haven't read throught he forums yet, here are all the ways i have found out to benchmark in UT2003 Demo

just to show fps within the game hit Tab and type " stat fps " without the quotes, this will display max and average FPS


Also head over to the UT2003Demo/System directory, there you will find a program called Benchmark.exe this plays through 2 fly by scenes and 2 bot scenes and displays a final result

also you will find stuff in C:\UT2003Demo\Benchmark

nice...."

Now post your results
:)

[ExitWound]

ctf-citadel
3.214851 / 17.586548 / 36.378597 fps
Score = 17.612741

:D

[nin_fragile14]

1280x1024, no aniso or aa

botmatch antalus: 44.5 fps
flyby antalus: 118.4 fps

botmatch asbetos: 56.7 fps
flyby asbetos: 166.3 fps

[Crippler]

Played the first demo level.

Slow as on my Ti4600 with 2xaa and 4x anistropic filtering.

With the options disabled its more playable.

ATI's logo should be on startup. Its best on their card atm :p

Bench it later on.

[NeilY]

UT2003 Demo, benchmark app.

XP2100+, 1GB PC2100/KT266A, Audigy, XP Pro, Ti4600 default.

1024x768

No AA, no Aniso

Flyby: 130.536758 fps
Botmatch: 46.589367 fps

[nin_fragile14]

Tried benchmark app too, at 1024x768, no aa or aniso

Flyby: 160.42 fps

Botmatch: 54.89 fps

edit:

@ 1600x1200

Flyby: 113.34 fps

Botmatch: 54.19

Quite a CPU limited game. I don't think the flyby scores even matter; the game is only played with other people, which is more represenative of the "botmatch" scores.

[Grommit]

man i need an upgrade

640*480
flyby 21.96
botmatch 15.18

hey i just noticed i had aa *2 on
640*480
flyby 44.36
botmatch 27.25
but nv30 cant come soon enough:)

xp 1800
asus a7v133c
384 mb ram
gforce 2mx 64mb(200core 180mem)
sbl value
det 4041

[Nutty]

1024x768x32

2x Aniso
No AA.

Flyby: 92
Botmatch: 34

[sytaylor]

ctf-citadel
12.128448 / 32.933201 / 67.661026 fps
Score = 33.139965

dm-antalus
3.067780 / 27.078579 / 53.090290 fps
Score = 27.094137

dm-asbestos
16.388601 / 35.378353 / 63.396591 fps
Score = 35.390633

br-anubis
4.549019 / 51.526283 / 93.489777 fps
Score = 51.556042

dm-antalus - flyby
4.004732 / 90.984131 / 375.635040 fps
Score = 91.031944

dm-asbestos - flyby
4.449244 / 112.025116 / 322.898010 fps
Score = 112.123650

ctf-citadel - flyby
3.316787 / 79.336212 / 219.344299 fps
Score = 79.483627


if you look in the benchmark folder theres a batch file that runs each level individually :D rather cooooool :) those results are on a

1.2ghz tbird B
512mb 133mhz ram
geforce 4ti4400 (stock)
abit kt7a raid

[Kev1]

Flyby Avg: 101.117 fps

Botmatch Avg: 46.144 fps

Game looks awesome using GF3. Can't Imagine what it will be like with a newer video card.

Still, it plays very fast and is very fun :D

[Kev1]

Using 1024 x 768, 32 bit color, No AA but am using patched 30.30 Dets with a -LOD of 1.5.

[cricket]

ran benchmark at 1024 x 768 results were: flyby 149.395157

: botmatch:57.247272

system spec's : Dell Demension 8200 P4 2.26GHz with 533MHz system bus/ 512K L2 Cache ~256MB PC800 RDRAM~80GB Ultra ATA/100 Hard Drive~128MB DDR NVIDIA GeForce4™ Ti 4600~with 30.82 driver's~Direct X 8.1~24x/10x/40x CD-RW Drive~ Harman Kardon HK-395 Speakers with Subwoofer ~Sound Blaster Live~ WinXP Home edition~withSP1

[volt]

Flyby: 77 fps
Botmatch: 46 fps
Resolution: 1280x960 2X AA

[Matthyahuw]

111.9
45.6
1800+, & GF3 @ 10x7

not bad, thought it would've been worse, especially after playing the betas

This code is TIGHT!!!

[JamesDax]

No AA No AF No overclock

1024x768x32
Flyby: 78fps
Botmatch: 32fps

Not to bad on my dated system. While playing the game runs smooth and looks great.

[fasedww]

I got, Flyby 142.0144, botmat 53.5, at 1024/768 32 bit, no AA Etc.;)

[mcortz_2000]

Hi All,

This is the first set of mini benchmarks run on my system. The settings were set to maximum quality, or at least what I could set in the options section, including HQ 3D hardware sound + EAX extension. I am using the Omega KX 1.1.82 driver with quality settings on. All of these are the more strenious botmatch bechmarks so don't be scared by the results. The resolution was set to 1024x768, max.

UT2003 Build UT2003_Build_[2002-09-13_17.31]
Windows XP 5.1 (Build: 2600)
GenuineIntel Unknown processor @ 2396 MHz with 511MB RAM
NVIDIA GeForce4 Ti 4600 (Omega KX 1.1.82) (3082)

dm-antalus?spectatoronly=true?numbots=12?quickstart=t rue -benchmark -seconds=77 -exec=..\Benchmark\Stuff\botmatchexec.txt

13.998052 / 37.235203 / 70.214691 fps rand[2992]
Score = 37.249538

br-anubis?spectatoronly=true?numbots=12?quickstart=tr ue -benchmark -seconds=77 -exec=..\Benchmark\Stuff\botmatchexec.txt

21.706738 / 55.652126 / 96.608856 fps rand[11570]
Score = 55.660789

dm-asbestos?spectatoronly=true?numbots=12?quickstart= true -benchmark -seconds=77 -exec=..\Benchmark\Stuff\botmatchexec.txt

22.989298 / 48.633221 / 84.101189 fps rand[21047]
Score = 48.655060

ctf-citadel?spectatoronly=true?numbots=12?quickstart=t rue -benchmark -seconds=77 -exec=..\Benchmark\Stuff\botmatchexec.txt

17.912407 / 42.310249 / 77.634651 fps rand[29105]
Score = 42.355621

My system spec is listed at the bottom of my post so u can compare. Let me know if there are any resolution/specific settings that u would like me to run. I will run them in the evenings and post it later that night.

Comments are also welcome

[SnakeEyes]

Wow. After playing with some of the tweaks listed around the nVnews site and elsewhere, I reran the benchmark.exe.

First, the settings-
(taken directly from my ut2003.ini file, so I hopefully don't miss anything):
[WinDrv.WindowsClient]
WindowedViewportX=800
WindowedViewportY=600
FullscreenViewportX=1024
FullscreenViewportY=768
MenuViewportX=640
MenuViewportY=480
Brightness=0.800000
Contrast=0.700000
Gamma=1.000000
UseJoystick=False
CaptureMouse=True
StartupFullscreen=True
ScreenFlashes=True
NoLighting=False
MinDesiredFrameRate=35.000000
Decals=True
Coronas=True
DecoLayers=True
Projectors=True
NoDynamicLights=False
ReportDynamicUploads=False
TextureDetailInterface=UltraHigh
TextureDetailTerrain=Normal
TextureDetailWeaponSkin=Normal
TextureDetailPlayerSkin=Normal
TextureDetailWorld=Normal
TextureDetailRenderMap=Normal
TextureDetailLightmap=Normal
TextureMaxLOD=0
TextureMinLOD=0
NoFractalAnim=False
ScaleHUDX=0.000000

[ALAudio.ALAudioSubsystem]
UseEAX=False
Use3DSound=True
UseDefaultDriver=True
CompatibilityMode=False
UsePrecache=True
ReverseStereo=false
Channels=32
MusicVolume=0.000000
AmbientVolume=0.300000
SoundVolume=1.000000
DopplerFactor=1.000000
Rolloff=0.500000
TimeBetweenHWUpdates=15.000000
DisablePitch=False
LowQualitySound=False

[D3DDrv.D3DRenderDevice]
DetailTextures=True
HighDetailActors=False
SuperHighDetailActors=False
UsePrecaching=True
UseTrilinear=True
AdapterNumber=-1
ReduceMouseLag=True
UseTripleBuffering=False
UseHardwareTL=True
UseHardwareVS=True
UseCubemaps=True
DesiredRefreshRate=60
UseCompressedLightmaps=True
UseStencil=False
Use16bit=False
Use16bitTextures=False
MaxPixelShaderVersion=255
UseVSync=False
LevelOfAnisotropy=1
DetailTexMipBias=0.800000
DefaultTexMipBias=-0.500000
UseNPatches=False
TesselationFactor=1.000000
CheckForOverflow=False
DecompressTextures=False
UseXBoxFSAA=False
DescFlags=0
Description=

With these settings, in-game I'm seeing what appears to be a minimum (NOT average) framerate in the upper 50's lower 60's in all the maps, including the DM-Antalus map (up until now the one that usually pushes my system the hardest).

Now, as far as the benchmark settings go, I left the MaxDetail.ini and MinDetail.ini files mostly alone, except for one change- I changed the ReduceMouseLag variable to false in both.

As far as video card settings go, I've been playing the game most of the weekend without any aniso, and with 4xs FSAA. But I decided to tweak those too, and used RivaTuner to adjust the D3D LOD setting to a -1.5 as suggested and used by Kev1 (thanks for the tip :)), and then dropped to Quincunx AA again. For the benchmark results, I'm now getting:

http://www.nvnews.net/vbulletin/attachment.php?s=&postid=18211

I'm preparing to copy my current benchmark MaxDetail ini to a backup location, and then make a copy of my current game ut2003.ini in it's place, to see what it reports for those settings. I expect a decent increase in performance over what I've got here. Hopefully I'll get time tonight, and can post THOSE results here.

[sk8bloke22]

lol, this cant be right.


/me looks at my flyby results. this benchmark was taken at 1024x768x32 with Quincunx FSAA

either ive got some super incredible setup (which i dont) or somethings odd here.

[SnakeEyes]

That's a great flyby score. Now you just need to work on the botmatch results a bit. :D

As far as my tests have gone, I did the bit where I replaced the MaxDetail.ini with my current one, and reran the benchmark executable. It appeared to be running a bit quicker in the botmatches, but the end results only showed a 15fps increase in the flyby, with no (well, inconsequential, as in .02000) change in the botmatch score. I know that in-game it is much better, because until I'm satisfied I've tweaked it out as well as I can for both performance and IQ, I've been playing with the fps display on. The new settings (listed above) have made no less than a 20-25 fps difference in the minimum framerates in all maps. (This is equivalent to around a 120% improvement in the minimum I had at max detail / 4XS AA in DM-Antalus, btw. :p)

[Gator]

1024x768 @ 32bit color

Flyby 123
Botmatch 43

(me shrugs) works for me
:cool:

[volt]

I ran another one this time at 1024x768 / 32bit / 2x AA / OpenGL mode:

Flyby: 99.5
Botmatch: 49

This time I enabled vsync and set the sound to "low quality"':

Flyby: 114
Botmatch: 49

My ini is a bit different though:

[WinDrv.WindowsClient]
WindowedViewportX=800
WindowedViewportY=600
FullscreenViewportX=1280
FullscreenViewportY=1024
MenuViewportX=640
MenuViewportY=480
Brightness=0.800000
Contrast=1.000000
Gamma=0.800000
UseJoystick=False
CaptureMouse=True
StartupFullscreen=True
ScreenFlashes=True
NoLighting=False
MinDesiredFrameRate=20.000000
Decals=True
Coronas=True
DecoLayers=True
Projectors=True
NoDynamicLights=False
ReportDynamicUploads=False
TextureDetailInterface=UltraHigh
TextureDetailTerrain=UltraHigh
TextureDetailWeaponSkin=UltraHigh
TextureDetailPlayerSkin=UltraHigh
TextureDetailWorld=UltraHigh
TextureDetailRenderMap=UltraHigh
TextureDetailLightmap=UltraHigh
TextureMaxLOD=0
TextureMinLOD=0
NoFractalAnim=False
ScaleHUDX=0.000000

[Engine.Engine]
;RenderDevice=D3DDrv.D3DRenderDevice
;RenderDevice=Engine.NullRenderDevice
RenderDevice=OpenGLDrv.OpenGLRenderDevice
AudioDevice=ALAudio.ALAudioSubsystem
NetworkDevice=IpDrv.TcpNetDriver
DemoRecordingDevice=Engine.DemoRecDriver
Console=XInterface.extendedconsole
GUIController=XInterface.GUIController
Language=int
Product=UT2003Demo
GameEngine=Engine.GameEngine
EditorEngine=Editor.EditorEngine
DefaultGame=XGame.XDeathmatch
DefaultServerGame=XGame.XDeathmatch
ViewportManager=WinDrv.WindowsClient
;ViewportManager=SDLDrv.SDLClient
Render=Render.Render
Input=Engine.Input
Canvas=Engine.Canvas

I will change the texture details to 'normal' and see what comes up hehe

[volt]

With SnakeEye's config I only get 99 in flyby and 45 in botmatch...weird.

[thcdru2k]

geesh this game is totally cpu limited.

1024x768x32 max details, eax+3d sound, no af, quincunx aa

114 flyby and 50 botmatch

[SnakeEyes]

Originally posted by d1rX
With SnakeEye's config I only get 99 in flyby and 45 in botmatch...weird.
Is that in OpenGL, or in D3D? Looking at your earlier scores for OpenGL, I may have to give that a go, just to see how it works/looks.