Whats your Best Frame rate in UT?

[IronMan_Oz]

Was wondering how I compare to others in frame rate.

My best possible frame rate is around 71.45

This is in the ideal placement standing still looking at a blank wall, so I figure this must be the limit.

I usually run in 1024X768 32bit medium world and skin mode. Dynamic lights on.

I tested it in various screen sizes from 800 up to 1280 or 13xx can't remeber my highest. . There were no difference if I face a blank wall it still was around 71.45. Maybe up/down .01 lol

In Face my fps stays up there pretty good even running around like crazy. Then there's those maps where I'm lucky to get 30fps. lol

Oz

[Ghost_Recon]

consider urself lucky...

my Duron 700 and GF2MX400 runs at 20-30 fps in all maps... at 640x480... High stuff for everything.

[phalanx]

hmm i have never checked that:D

[Nova]

i'm not sure what this does, but i took mine down to 1. Whats it do?:dunno:

[Ghost_Recon]

ok if you wanna check it... open the console by pressing tilde key


THIS ----> ~

then you type

THIS ----> timedemo 1

(with the space)

and then your interface should have FPS on it displayed.

[phalanx]

thanks recon:D

[Moot]

Mines almost always pegged at 75 (vsync on). It used to go up to 175-200 avgerage with vsync off, but now it'll just go to 100ish. :scratch:

Dual Athlon 1800s & Voodoo 5, 1024X768 32bit

[IronMan_Oz]

I just hit toolmenu from the main screen then timedemo

Oz

[JohnnyBravo]

(Another way to turn TimeDemo on is go up to Tools on the main menu and select TimeDemo Stats.)

Playing online I keep a constant 45-55 fps at 1024x768@32 with everything on full. If I stare at a wall I can get up to around 90-100 but only if i stay still. No vsync either as I'm on an LCD monitor.


EDIT: Damn, beten to it.

[Doofus]

up to about 500 staring point blank at a wall with a Voodoo5 5500 AGP, with a 1.6a P4 processor pushed to about 2.3GHz. Same card with a 1GHz processor used to show about 300 under the same conditions. Under regular gameplay, High/High/dynamic 1024, multiple players (but other high bandwidth options turned off in advanced preferences), I get 50-130+

Voodoo cards don't do 32 bit, only 16.

With everything on, including 2 sample antialiasing on a high polygon map, I get about 30-40 fps. Better designed maps, more like 50-60.

I purchased a GeForce3 Ti200 to go with my new processor, but even when overclocking the GF3, the Voodoo is more playable (and colors appear more saturated), so I switched back. UT's optimization for Glide still pays off for Voodoo 5 owners.

Try the Wicked400 benchmark for realistic gameplay FPS stats (as opposed to the blank wall trick).

[DeadLamb]

Average around 90ish in games with alot of firefights unless the level SUCKS then even on my cpu it will drop to 35 but that is due to really, really poor level design more then anything else...

run around 125 avg just rolling around the level with out seeing any battles

I think I get sub 200 looking at a wall.. not sure have not tried that for along time

Most the time I just click on the timedemo in the menu area then bitch that I'm getting 40fps and that is just BS level design.. eh but then the levels are free and better then anything I could make sooooooo :)

p4 2.2 512 rambus
Ati 8500 card

[Chilly!!]

Here's the long winded answer - running an 850 athlon @1100 and a radeon 64mbDDR - yes its old, but good :) , 1024 x768, med, med , decals on , no dynamic lighting, music off - overall average of 71-75 online.
Highest average: Noon Raid - 83 fps
Lowest average: Euro2000 - 18 fps and lots of praying

You Voodoo guys are alright...my friend swears by his Voodoo5 to this day...best I saw him get was around 180 average about 2 mos. ago. Then his cooler leaked and ...pfffftttsk...doh...

[Chilly!!]

Did I say Athlon? I meant Duron :hammer:

[Ajenthavoc]

sometimes i go 10 fps during hardcore fights, but i usually run around 35 fps.. notice how most of the good players are 60+ fps constantly and rarely below that... almost makes me wanna buy a voodoo 5 just to narrow the handicap...

[ElectricHead]

120 or so...goes down to about 65 and peaks at 200+...

1024x768@32bit eveything on max
AMD XP1900+@1740 :D
GF3 225/530
D3D


EH

[Doofus]

Originally posted by Chilly!!

You Voodoo guys are alright...my friend swears by his Voodoo5 to this day...best I saw him get was around 180 average about 2 mos. ago. Then his cooler leaked and ...pfffftttsk...doh...

I compared with a GF3 Ti200, overclocked, and ended up going back to my Voodoo5. As far as your friend with the cooler, I saw no real benefit from overclocking the chip (from the stock 166 up to about 180) on a fast machine (I'm running my 1.6 P4 at about 2.3). When you go from say 80 to 83 fps, the gain is not worth the overclock instability. So I've stopped overclocking my V5, as most of my performance gain is processor related.

BTW, thanks to Lamb, Inkswitch, et al who have given me tech advice in the past, back before I upgraded my system and was trying to get an idea of what kind of performance was possible.

And no matter how good your system, we are all limited by the server and the size of the packets it sends out. Online performance is never going to be as good as LAN or playing offline with bots.

[JohnnyBravo]

Originally posted by DvNA-Ajenthavoc
sometimes i go 10 fps during hardcore fights, but i usually run around 35 fps.. notice how most of the good players are 60+ fps constantly and rarely below that... almost makes me wanna buy a voodoo 5 just to narrow the handicap...


You would hate yourself if you did. The UT engine was built for Glide which is why they perform so much better than alot of D3D cards (Radeons, Geforces etc) but these aren't the same results you would get with other games. More and more you're going to see game developers saying that Voodoo cards are not going to be supported with new titles due to the card's lack of driver support past Directx 7. NWN uses OpenGL 1.2, therefore meaning Voodoo cards were dropped from support. (Granted some people got it to run using a Voodoo 5, but they're using 3rd party drivers, and in reality they could have spent $25 on a TNT2 and gotten better performance.)

It's unfortanate that Voodoos are slowly getting phased out, but I suppose thats just how the industry goes.

[Nova]

i get around 33.

[Chilly!!]

Originally posted by Doofus:

I saw no real benefit from overclocking the chip (from the stock 166 up to about 180) on a fast machine (I'm running my 1.6 P4 at about 2.3). When you go from say 80 to 83 fps, the gain is not worth the overclock instability

True enough. After water cooling chip and card he produces about a 45fps gain with no instability which, to him, was significant.

FPS is less important than decreased fluctuation in framerates and that's more of what he was after. A good example of this was achieved by doing something as simple as installing a new driver (on my rig).

On Cola Wars the frames would fluctuate from 19 in the central area with the enemy base in fov to 77 within the bases, producing an average of around 47 based on my play style of broad map coverage. After the new driver, average fps only jumped by 4 or 5 fps, but the lows went from 19 to about 27, resulting in smoother overall play. The overall peaks of the card didn't change but the capacity was more stabilized.

I've played on his machine with a V5 and a Radeon8500 and the V5 was by far the better card despite the 8500's higher fps.

I agree with you Doofus , on the appreciable gains. As far as I'm concerned once you get past lows of 110 improvements aren't really worth the effort and risk.

Still trying to figure out whatsits of this posting thing tho...hope i quoted you right...:angst:

[Doofus]

Originally posted by Chilly!!


True enough. After water cooling chip and card he produces about a 45fps gain with no instability which, to him, was significant...

Still trying to figure out whatsits of this posting thing tho...hope i quoted you right...:angst:

Quoted just fine.

I was not able to see that significant a gain in fps (under gameplay conditions) with the stock cooling and a raise of clock frequency from 166 to 180 (maybe he's pushing it a lot higher with the added cooling). I stress gameplay, because a simple flyby with no players and action doesn't give a realistic assessment of how framerates will be in a firefight with 3 or 4 other players. That is why I use the wicked400 benchmark (a recorded demo that has a firefight with multiple bots always on screen). I can get 500 staring at a wall, and 200+ in an average flyby, but real action framerates are more in the 50- 100 range.

[Chilly!!]

Originally posted by Doofus:

(maybe he's pushing it a lot higher with the added cooling). I stress gameplay, because a simple flyby with no players and action doesn't give a realistic assessment of how framerates will be in a firefight with 3 or 4 other players. That is why I use the wicked400 benchmark

Yup. He's cranked up the speed of the card considerably, he's big into that.

I'd like to try that wicked400 benchmark, where can I grab it? It would certainly be more reliable.

My numbers come from gameplay on populated servers, no empty flybys or 1vs1bot coret...( which btw I get 90 on..:P which means ..NOTHING ..:D

[Doofus]

http://www.ve3d.com/3dpulpit/Unreal/Unreal.html

He has several other benchmarks there too. It states the wicked400 won't work with UT versions greater than the original retail 400, but I have no problems running it upgraded to v436. I haven't tried his other benchmarks yet.

[Bobjua]

---->8 FPS<----

[yomama]

just press F5 or F6 in game

[FBMachine]

I tested it out offline with Disposable Heroes, I got about 40 fps average, and 85 fps ( my monitors refresh rate ) staring at 1 poly. Usually online I only get between 20-30 fps, and every 20 seconds or so I freeze for like 3 or 4 seconds. Hopefully it won't be glitchy like that in UT2003.
Dan
EDIT: I was playing with 15 bots when I did the fps test.

[SkaarjMaster]

Best that I actually wrote down was 179.78 with an Athlon1800+, Dragon+ and GeF4Ti4400 card.

[_F7_]

My top was a few points over 200. Granted there wasn't much happening on screen, but I was happy given that my old PC was hard pressed to top 60 FPS. Heavy action nets 115-145.

XP 2000+
Leadtek GeForce4 128MB
512MB Samsung DDR
Abit KX7-333

[HERO Maxwell]

well, I have a 1.8 ghz(crappy lcache..vcache...er some crap)
Geforce ti 200
soyo dragon mobo
and my frames usually stay about 30 on stealths avg. I run my resolution at 1024x768 and set to 16 bit color.
er, and for some reason my mouse and keyboard continually decides to lock up on me. I think its my mother board or something. Anyway, I just hope ut2003 will run smoother.

[D.J.P.]

20 gajillion fps...

[blito3]

just played my new update to the 12H contest
ran about 45...yikes...wellll u lower machines might as well kill yourselfs

[Cobra]

I get 70 -71 everything on max, 20 players all the time t dosenz drop. 1024x768

[Doofus]

Originally posted by Cobra
I get 70 -71 everything on max, 20 players all the time t dosenz drop. 1024x768

What's your rig?

[Cobra]

It is a Alton xp 2000+ GF4 512MB is it good for the test 70-71 because I d'ont know.

[Doofus]

Originally posted by Cobra
It is a Alton xp 2000+ GF4 512MB is it good for the test 70-71 because I d'ont know.

Sounds good to me. That's a nice rig.

If you want to compare overall video performance, you can go to MadOnion.com and download their 3dmark benchmark. Run the test, then post and compare your results with others online.

[Cobra]

Thanks I will do it.

[Cobra]

Hi Doofus, the benchmark is CPU= 4931, memory= 3124, HDD= 445, know tell me what it is saying please, because I d'ont understand a thing, thanks Doofus.

[Cobra]

Sorry Doofus, rong test I did it again and it is, 5700 Dmarks my husban tol me that it was not to good, he toll me to by a GF4 Ti 4600 128Mb and put 512MB more, I will do it this month, thanks again Doofus.

[Zinia]

Cobra

You should be getting marks in the 10000 point range.

My AMD 2000 with GF4 4600 and 512mb Corsair is about 10800

My AMD 1900 with GF3 and 256mb Mushkin gets 9900

My Pentium 2.56 with Gf4 4200 and 512mb mem is 10300.

These are all stock numbers with no tweaking.

[Cobra]

The Gf4 that I have, 440mx my husban told me that in fact, it is simply a GF2 upgraded is that right.

[Ghost_Recon]

oh.......... the MX series. well in that case yes. I guess that sorta explains that score. We all thought you were using the Titanium series.

[Susume]

Vsync is on, on my GPU so I get 70fps as my monitor refresh rate is 70mhz.

[HisDivineShadow]

I am jealous of you all......I need new equipment!:angst:

[Nimda]

Haven't looked at FPS in UT. How do you set it again?

My 3DMark2001 SE score is around 11,700 non overclocked anywhere (CPU, Video, or mem).

[Nimda]

Nevermind... I'm lazy... Found it in first page "timedemo 1"

Will post this weekend. I run at 1280x1024 32 bit color an a 21" monitor. I would suspect it would be over 100, but I am sure the high detail and resolution will not allow it to get 170 like someone else posted...