UT2004, Far Cry - CPU and VPU (nv40, R420)

[breez]

I agree that ONS runs like crap but DM runs very well unless it's one of those stupid servers with 15 players in 1on1 map. Running on Athlon xp 2.1GHz, 512MB, R9800Pro oc'ed.

 

[hovz]

i dont think dm runs well. i mean i dont think it runs BAD, just not good. godo to me is staying close to 60 fps consistently. i only get that in 1v1. when im playing in factory or such i frequently drop below 40. makes it veyr hard to keep steady with aim.

 

[Diplo]

Here's a few DM screenshots I took.

Settings:
1024x768x32
Textures: Highest
Physics: Highest
World Detail: Highest
Decals, coronas, dynamic lighting etc. : On
Shadows: Blob
Renderer: D3D
Filtering: Trilinear
Note: Firewall, virus scanner, browser, webserver all running in background 'cos I couldn't be bothered to shut them down

System:
Athlon XP 2800
1GB PC-3200 RAM
GeForce4 ti4400
56.72 Dets

First a 6 bot DM match on Asbestos, an average sized map. I tried to take screenshots in action, but it's quite hard to play and do so. So, these tend to be a fraction of a second after killing a couple of bots. Figures in brakets are average FPS

http://www.diplo.co.uk/images/ut2004/Asbestos.jpg (72)
http://www.diplo.co.uk/images/ut2004/asbestos1.jpg (69)

Next is a 10 bot DM game in Metal Deity, one of the largest DM maps:

http://www.diplo.co.uk/images/ut2004/Deity.jpg (49)
http://www.diplo.co.uk/images/ut2004/Deity1.jpg (45)

Last shot has lowest FPS I ever got, and that is in the middle of a mega-kill with full gore and highest physics on a 2+ year old card. Turning physics detail to low and gore off will make a large difference in those circumstances.

 

[hovz]

no actually it wont, if it did i would turn it off

 

[hovz]

also that low in the 30s wa sina very tiny room. imagien if u wer eon a map like antalus, goliath, metallurgy and other more open maps. constantly drops to the 30s. thats not my idea of running good.

 

[Ostsol]

no actually it wont, if it did i would turn it off

It might if there was a situation with lots of people being gibbed at once, but usually not since those are not entirely constant effects.

also that low in the 30s wa sina very tiny room. imagien if u wer eon a map like antalus, goliath, metallurgy and other more open maps. constantly drops to the 30s. thats not my idea of running good.

Actually, I think it'd be worse in a small room. In a large open area it's more likely that you'd be shooting at targets much further away, meaning generally smaller mouse movements for aiming. As such, lower framerates would be much less noticable -- if at all. Swinging the mouse cursor around real fast in close combat, however. . . that's when the difference could become noticable. Indeed, the only time when people ever notice the difference between framerates in the 30s and framerates in the 60s is when the movements are large and you start seeing larger gaps between the start and end positions.

 

[hovz]

i can feel the dsifference in just walking, its very easy to feel.

 

[Zvekan]

I just wanted to say that UT2K4 suprised me as it ran flawlessly in around 30-40 fps with a old P4 2 GHz with all details set to high (Radeon 9800 Pro) without hichups, at least that I can remember.

I was very pleased that I was able to play it unlike PainKiller that works flawlessly for first chapter but with introduction of fog or smoke starts to stutter.

Zvekan

 

[Nayr]

Ok,
FarCry=No internet connection
UT2k4=Internet Connection

Even if you had a 5Ghz Machine with Duel Geforce 79000 and 10 Gigs of RAM, any game played over the internet will lag at some time or another. The ping time from server to server to server(trust me, NET+ course A+ course also) will get your computer to "stall" waiting for data. A stall in data=Lag. It's that simple buddy.

Runs perfectly fine on my rig(1024x768,8XAF, Max detail), and I have an AXP2500+, Epox8RDA3+, 64MB Geforce 4 Ti 4200, PC2700, and an OLD 20GB hard drive. Stop b***hn and runnin your mouth off, cause you really don't know what your talkin about. Have you EVER programed a game? Are you even A+ cert? Are you going to SHOW us how vastly superiour your knowledge of game programing and it's effects on Hardware is? Oh please o mighty one, enlighten us

 

[London Geezer]

Even if you had a 5Ghz Machine with Duel Geforce 79000 and 10 Gigs of RAM,

I actually gasped when i read those numbers. Might be the coffee, but imagine when computers will have those kind of specs... All of them to run Longhorn '25 (as in 2025) at decent speeds...

 

[Nayr]

Even if you had a 5Ghz Machine with Duel Geforce 79000 and 10 Gigs of RAM,

I actually gasped when i read those numbers. Might be the coffee, but imagine when computers will have those kind of specs... All of them to run Longhorn '25 (as in 2025) at decent speeds...

LOL, thats great. It would be nice.

 

[jb]

hovz,

in the screen shots you posted is that with or with out bots?

Did you try to run one of the flyby demos?

 

[breez]

Also have you tried running with the D3D9 renderer? Open ut2004.ini and change some semi-colons there.

 

[grecco_julio]

Interesting, in UT2003 (2003) the framerates aren't CPU Limited and they act like the Far Cry framerates... Isn't that strange and doubtfull?

 

[kyleb]

try br-disclosure, i remember when ut2003 first came out i had a 1900+ which was pretty high end at the time and i saw 5fps with heavy action in the bases on that map. now that i have a barton runnigng at 3000+ it actaully stays above single digits.

 

[KimB]

Interesting, in UT2003 (2003) the framerates aren't CPU Limited and they act like the Far Cry framerates... Isn't that strange and doubtfull?

No.

 

[Diplo]

Interesting, in UT2003 (2003) the framerates aren't CPU Limited and they act like the Far Cry framerates

Both UT2004 and UT2003 use the same engine - Unreal Engine 2. There are some minor differences, bug fixes and optimisations, but basically 2004 is the same engine as 2003.

 

[hovz]

hovz,

in the screen shots you posted is that with or with out bots?

Did you try to run one of the flyby demos?

that was an online game. in a botmatch its even worse.

 

[jb]

that was an online game. in a botmatch its even worse.

Bot match uses much more of your CPU. Try removeing your ut2004.ini (make a back up of it) so every thing is set to defaults. Then play a game with out changing any settings and see if that helps...

 

[grecco_julio]

If they use the same engine, UT2004 would not be CPU limited, because UT2003 isn't! BUT IT IS! THAT'S WHY THAT'S STRANGE SH!T!