-
We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.
Celeron 1.4 + Radeon 9100 250/230 too slow in UT2003. Will Radeon 9500 Pro solve my problems?
- Thread starter Eug
- Start date
Eug
Lifer
OK. It's a 64 MB, and clocked at 250/230 (3.5 ns EtronTech). The 128 MB is clocked at 250/200 (4 ns, can't remember brand).
I tried Rage3D Tweak 3.8 and it actually gives me the wrong speeds for the stock speeds, at 250/250. Even with this speed, I get lots of artifacts. "Underclock" the memory to the 230 MHz Sapphire intended it to be, and the artifacts disappear (although I've seen rare ones even at this speed). The GPU goes up beyond 275 no problem though.
OK then, nobody else has any dissenting opinions about my slow CPU? If not then maybe in a year I'll get a new 3 GHz processor with a $200 Radeon 9700 non-Pro or something. (The retail Radeon 9500 Pro is about US$200 at the moment.)
Keysplayr
Elite Member
Ok, If it's a crippled processor, why can I play UT2003 just fine on a PIII 1GHz box, with 384MB SDRAM and a GeForce2 Ti 64MB card, 13 GB HD.
Certainly a 1.4 GHZ Celeron can perform just as well as a PIII 1.0 GHZ cpu. Dont you think?
Before anyone condemns this persons processor (1400 MHz is nothing to sneeze at) I would try to direct questions about what was tried as far as fixing this problem.
Eug, what have you tried, if anything, to solve your problem?
Keys
Certainly a 1.4 GHZ Celeron can perform just as well as a PIII 1.0 GHZ cpu. Dont you think?
Before anyone condemns this persons processor (1400 MHz is nothing to sneeze at) I would try to direct questions about what was tried as far as fixing this problem.
Eug, what have you tried, if anything, to solve your problem?
Keys
Keysplayr
Elite Member
I have found that some people have poor performance when they do not properly install their sound card drivers. I know it sounds odd, but if your sound card drivers aren't up to snuff, it could severely impact video performance, cause choppiness and I have even seen clicking popping noises to accompany it. And, what kind of vid card did you have before you bought the Radeon 9100? ATI? Nvidia? other?
Keys
Keys
Eug
Lifer
The Celeron 1.4 Tualatin is the same design as the old PIII 1.0 GHz Coppermine but at 0.13 u. Both of these chips have 256 KB cache. So essentially, it's an old PIII running at 1.4 GHz on a 100 MHz bus. (The new PIII Tualatin has a 512 KB cache, and the old Celeron Coppermine has a 128 KB cache.)
Maybe I'll try playing with the settings of the game and the video card some more. I'm running the CAT 3.1 drivers.
The problem is NOT with indoor maps, but only with a few outdoor maps. It's not completely unplayable, but it does sometimes drop below 25 fps.
Previous card was Radeon 7200. I also have a 1 MB Trident ISA card for a few days after my Radeon 7200 was killed by me. I never did install the Catalyst drivers for the 7200 though, since with even the generic MS drivers, it was fine for Quake III.
I did try a Radeon 9000 and Geforce4 MX 440, but I believe I reghosted back to my Radeon 7200 settings after trying those.
Maybe I'll try playing with the settings of the game and the video card some more. I'm running the CAT 3.1 drivers.
The problem is NOT with indoor maps, but only with a few outdoor maps. It's not completely unplayable, but it does sometimes drop below 25 fps.
Previous card was Radeon 7200. I also have a 1 MB Trident ISA card for a few days after my Radeon 7200 was killed by me. I never did install the Catalyst drivers for the 7200 though, since with even the generic MS drivers, it was fine for Quake III.
I did try a Radeon 9000 and Geforce4 MX 440, but I believe I reghosted back to my Radeon 7200 settings after trying those.
Keysplayr
Elite Member
Eug, I think you need to find a way to completely remove any and all video driver remnants from your old vid cards. The trident and the 7200.
I dont know where you can get the "cleaner" that people on this forum use. Maybe one of you guys know of the link and of the program I am talking about?
Keys.
EDIT: Definately install the Creative Drivers for that SB Live card. Try that first above all.
I dont know where you can get the "cleaner" that people on this forum use. Maybe one of you guys know of the link and of the program I am talking about?
Keys.
EDIT: Definately install the Creative Drivers for that SB Live card. Try that first above all.
We're not talking about playing at 640X480 Medium settings. The 8500 Should be able to run 2K3 at 1024*768 with default settings. Yes a card that came clocked 275/250, or 275/275 would be better at this. I had a gf2 gts, and going from a 700 Celeron, (which i know is not tultarian) to the 1600 was night and day difference with the same video card.
Makes sure your settings are right in the bios, such as 4x agp, and 64MB Agp aparture size. Even with the Celery Stick you should be able to play 2k3 decently. You don't need a 3Ghz cpu to play 2k3.
Keysplayr
Elite Member
Quote: By CurtCold
We're not talking about playing at 640X480 Medium settings. The 8500 Should be able to run 2K3 at 1024*768 with default settings. Yes a card that came clocked 275/250, or 275/275 would be better at this. I had a gf2 gts, and going from a 700 Celeron, (which i know is not tultarian) to the 1600 was night and day difference with the same video card.
Makes sure your settings are right in the bios, such as 4x agp, and 64MB Agp aparture size. Even with the Celery Stick you should be able to play 2k3 decently. You don't need a 3Ghz cpu to play 2k3.[/quote]
End Quote:
His processor is more than enough to run UT2003 and so is his 9100 card. Definately not the fastest system you would want to play UT2k3 on but surely fast enough. I think Eug should just clean up his system. Maybe even a total reinstall like so many ATI users on this forum suggests. Get all the proper drivers for his Mobo, audio, video, etc. Not the Microsoft drivers PNP stuff.
Keys
We're not talking about playing at 640X480 Medium settings. The 8500 Should be able to run 2K3 at 1024*768 with default settings. Yes a card that came clocked 275/250, or 275/275 would be better at this. I had a gf2 gts, and going from a 700 Celeron, (which i know is not tultarian) to the 1600 was night and day difference with the same video card.
Makes sure your settings are right in the bios, such as 4x agp, and 64MB Agp aparture size. Even with the Celery Stick you should be able to play 2k3 decently. You don't need a 3Ghz cpu to play 2k3.[/quote]
End Quote:
His processor is more than enough to run UT2003 and so is his 9100 card. Definately not the fastest system you would want to play UT2k3 on but surely fast enough. I think Eug should just clean up his system. Maybe even a total reinstall like so many ATI users on this forum suggests. Get all the proper drivers for his Mobo, audio, video, etc. Not the Microsoft drivers PNP stuff.
Keys
End Quote:
His processor is more than enough to run UT2003 and so is his 9100 card. Definately not the fastest system you would want to play UT2k3 on but surely fast enough. I think Eug should just clean up his system. Maybe even a total reinstall like so many ATI users on this forum suggests. Get all the proper drivers for his Mobo, audio, video, etc. Not the Microsoft drivers PNP stuff.
Keys[/quote]
You my friend are most certinly right. Most of these hardware/software problems come down to just having the right drivers, installation procedure...etc...
I def think the Celeron will run it. I know the 66Mhz fsb Coppermines were crap, I know from experience. 🙂
Eug
Lifer
The reason I ask about the 9500 Pro is because of Tom's Review.
With UT2003 running on an Athlon 1000, the 9500 Pro still has an enormous advantage over the 8500, even though the Radeon 9500 Pro is severely CPU limited. Athlon XP2700+ vs. Athlon 1000: UT2003 1024x768x32. 9500 Pro: 122 vs. 90 fps. 8500: 65 vs. 65 fps ie. With a fast CPU a 9500 Pro means almost a 2X increase in speed over the 8500, but with a slow Athlon 1000, it's still almost a 40% increase in speed. So are some of you saying that all this speed increase in occuring in stuff like indoor maps, where the CPU is OK? ie. Even the bad outdoor maps I'll still be very CPU-crippled? In the meantime I'll try tweaking the settings. The one thing I don't like is turning the world textures too low, because then it looks crappy. I wonder if Rage3D's Tweak for UT2003 will give me extra speed.
I tried removing remnants of the 7200 drivers, but I had a hard time finding anything manually. Like I said, I hadn't even installed ATI's Catalyst drivers for it. I was running just really old generic ATI ones I think that Win XP installed automatically. I did run a couple cleaner programs but they didn't seem to do anything. I do know that there must be 7200 remnants around though, since when I ran Rage3D Tweak 3.8, it asked me if I was running a 9100 or a 7200. Hmm.... I'm just trying to avoid a reinstall, but if I must, I must. In fact, one of the reason I got my Celly 1.4 in the first place was so I didn't have to reinstall. 😛
As for the AGP, I dunno if I can even support 4X AGP. 2X AGP yes, but my BX mobo is 4 years old. :Q
P.S. Running UT2003's benchie app (in the system folder), at 1024x768 I get about 80/34 fps. Yet in real life gaming with 5 bots on an outdoor map, I've even experienced sub 20 fps rates. 🙁
With UT2003 running on an Athlon 1000, the 9500 Pro still has an enormous advantage over the 8500, even though the Radeon 9500 Pro is severely CPU limited. Athlon XP2700+ vs. Athlon 1000: UT2003 1024x768x32. 9500 Pro: 122 vs. 90 fps. 8500: 65 vs. 65 fps ie. With a fast CPU a 9500 Pro means almost a 2X increase in speed over the 8500, but with a slow Athlon 1000, it's still almost a 40% increase in speed. So are some of you saying that all this speed increase in occuring in stuff like indoor maps, where the CPU is OK? ie. Even the bad outdoor maps I'll still be very CPU-crippled? In the meantime I'll try tweaking the settings. The one thing I don't like is turning the world textures too low, because then it looks crappy. I wonder if Rage3D's Tweak for UT2003 will give me extra speed.
I tried removing remnants of the 7200 drivers, but I had a hard time finding anything manually. Like I said, I hadn't even installed ATI's Catalyst drivers for it. I was running just really old generic ATI ones I think that Win XP installed automatically. I did run a couple cleaner programs but they didn't seem to do anything. I do know that there must be 7200 remnants around though, since when I ran Rage3D Tweak 3.8, it asked me if I was running a 9100 or a 7200. Hmm.... I'm just trying to avoid a reinstall, but if I must, I must. In fact, one of the reason I got my Celly 1.4 in the first place was so I didn't have to reinstall. 😛
As for the AGP, I dunno if I can even support 4X AGP. 2X AGP yes, but my BX mobo is 4 years old. :Q
P.S. Running UT2003's benchie app (in the system folder), at 1024x768 I get about 80/34 fps. Yet in real life gaming with 5 bots on an outdoor map, I've even experienced sub 20 fps rates. 🙁
AtomicDude512
Golden Member
No duh. The 128KB of cache not to mention the lower speeds are murder. Get a better CPU first, how about a dirt cheap Athlon XP 2100 for $87 (or so)?
Keysplayr
Elite Member
Yes Curt, those 66MHz FSB celeries were glorified word processors. 🙂
Eug, all of these "GPU based" cards are supposed to relieve the CPU's by taking on a lot of the functions that only the CPU's used to handle.
However, the newest and future games are very polygon intensive demanding even more geometry processing from the CPU. Not to mention the AI (artificial intelligence) these new games have and will have. So the vid cards are still "very CPU limited. I have a Radeon 9500 Pro on order and I cant wait to try it out. I have a P4 2.8 GHz rig 512DDR. Dying to see the frames flyby...
Without any of the AA/FSAA features enabled, the Radeon 9500 Pro mostly falls between the GF4Ti 4400 and 4600. But. When you max out the AA/FSAA settings on any game that supports it, the Radeon quite literally crushes the Ti 4600.
And the comparison of a Radeon 9500pro and 8500 "hell yes" and "hell no". (If you were to compare the two against each other) the 8500 is still a good card.
Keys
Eug, all of these "GPU based" cards are supposed to relieve the CPU's by taking on a lot of the functions that only the CPU's used to handle.
However, the newest and future games are very polygon intensive demanding even more geometry processing from the CPU. Not to mention the AI (artificial intelligence) these new games have and will have. So the vid cards are still "very CPU limited. I have a Radeon 9500 Pro on order and I cant wait to try it out. I have a P4 2.8 GHz rig 512DDR. Dying to see the frames flyby...
Without any of the AA/FSAA features enabled, the Radeon 9500 Pro mostly falls between the GF4Ti 4400 and 4600. But. When you max out the AA/FSAA settings on any game that supports it, the Radeon quite literally crushes the Ti 4600.
And the comparison of a Radeon 9500pro and 8500 "hell yes" and "hell no". (If you were to compare the two against each other) the 8500 is still a good card.
Keys
I run a C1.1 @ 1.5 and on 3DM01 my hacked 9500 (near 9700P) can outperform a high-end 8500 by up to 37%. The difference should be similar between a 9500P and 9100 (low-end 8500). Although my 133 FSB provides more oomph than your 100. If your ol' BX mobo has the necessary dividers to run at 133 you may want to consider "downgrading" to a C1.1 or 1.2.
I find that my system is more than adequate for the latest games at 1024x768. I have not played UT2003 but have dabbled in Raven Shield which uses the same ginny. BF1942 is what I play most and while it is quite demanding even with all A/V maxed out plus 8x AF it flirts around the default limit of 100 FPS.
The PIII Tualatin was available with either 256K or 512K L2 cache (the latter designated S for server). On my informal tests a C Tualatin outperforms a PIII Coppermine by 12% where both are 1.1GHz/100FSB/256K.
The 7200 thing may just be a detection of the secondary RAMDAC. If you think your system is a bit messy but do not want to re-setup from scratch then a repair setup is relatively painless and maintains most of your settings and all of your programs.
I find that my system is more than adequate for the latest games at 1024x768. I have not played UT2003 but have dabbled in Raven Shield which uses the same ginny. BF1942 is what I play most and while it is quite demanding even with all A/V maxed out plus 8x AF it flirts around the default limit of 100 FPS.
The PIII Tualatin was available with either 256K or 512K L2 cache (the latter designated S for server). On my informal tests a C Tualatin outperforms a PIII Coppermine by 12% where both are 1.1GHz/100FSB/256K.
The 7200 thing may just be a detection of the secondary RAMDAC. If you think your system is a bit messy but do not want to re-setup from scratch then a repair setup is relatively painless and maintains most of your settings and all of your programs.
Eug
Lifer
This benchmark is interesting.
It seems to suggest that the Radeon 9100 is similar in speed to the Radeon 9500 (non-pro), when using a slower CPU (Athlon 2000+), with the UT2003 demo botmatch benches.
Judging by this and Auric's comments, it may be that a 9500 Pro (not the non-pro) might help a bit, but not by a huge amt, on a slow CPU.
It's true that there would be tons of polys in those outdoor maps. Back in the old days, the UT map coders used to criticize the hell out of anyone doing this. Now that's the norm I guess, with the new CPUs and video cards out.
It seems to suggest that the Radeon 9100 is similar in speed to the Radeon 9500 (non-pro), when using a slower CPU (Athlon 2000+), with the UT2003 demo botmatch benches.
Judging by this and Auric's comments, it may be that a 9500 Pro (not the non-pro) might help a bit, but not by a huge amt, on a slow CPU.
It's true that there would be tons of polys in those outdoor maps. Back in the old days, the UT map coders used to criticize the hell out of anyone doing this. Now that's the norm I guess, with the new CPUs and video cards out.
The bot match would seem to be a test for single player with AI as opposed to MP where more of the CPU's work can be allocated to graphics. At 1024 the 9700P is besting the 9100 by 40% at bot but 70% at flyby.
P.S. I assume MP would be somewhere in between if flyby is typical of gameplay and accounting for some CPU time dedicated to network.
P.S. I assume MP would be somewhere in between if flyby is typical of gameplay and accounting for some CPU time dedicated to network.
BFG10K
Lifer
It's definitely your processor holding you back, especially if your problems increase when you add more bots to the game. A GPU upgrade might allow you to run at higher resolutions and detail levels with less of a performance hit but your actual framerate can only be raised by upgrading your CPU and your platform.
1 GHz is far too slow for UT2003 (heck, it's far too slow for most Quake 3 engined games too) and I wouldn't touch UT2003 on anything less than a 2 GHz (or equivalent PR rating) processor.
How about you run the individual botmatch benchmarks and let us now what your average and minimum framerate is. I guarantee that system of yours will get an utter pasting.
1 GHz is far too slow for UT2003 (heck, it's far too slow for most Quake 3 engined games too) and I wouldn't touch UT2003 on anything less than a 2 GHz (or equivalent PR rating) processor.
Eug
Lifer
So using Rage3D Tweaker with the UT2003 settings, and turning off quite a bit of eye candy, it feels smoother. That's including DM-Plunge, which wasn't quite as good before. Running 275/230. Usually stays above 25 fps, with most of it in the 40 fps range. Inside maps are in the 40-90 fps range.
However, DM-Inferno is a complete slide show with some angles. Down to 5 fps just standing there when at the top looking at the citadel and the fire in the background. :frown: Like watching some of the 3DMark2003 benches. LOL 😛 Could DM-2003 be considered a not-so-efficiently coded map?
BTW, so far I feel the maps in UT are nicer than UT2003. UT2003 still has time to improve though.
Bench results:
DM-Asbestos (bot): 2.216833 / 43.259750 / 91.198372 fps -- Score = 43.284378
DM-Asbestos (flyby): 58.420170 / 116.639801 / 317.484406 fps -- Score = 116.736824
DM-Antalus (bot): 16.186348 / 38.783768 / 80.806076 fps -- Score = 38.805805
DM-Antalus (flyby): 47.760868 / 96.251801 / 392.219147 fps -- Score = 96.303078
DM-Inferno (bot): 2.267646 / 40.243225 / 160.775635 fps -- Score = 40.233208
DM-Inferno (flyby): 46.598614 / 95.578873 / 264.641663 fps -- Score = 95.687866
I'm assuming the numbers are: min / average / max fps?
However, DM-Inferno is a complete slide show with some angles. Down to 5 fps just standing there when at the top looking at the citadel and the fire in the background. :frown: Like watching some of the 3DMark2003 benches. LOL 😛 Could DM-2003 be considered a not-so-efficiently coded map?
BTW, so far I feel the maps in UT are nicer than UT2003. UT2003 still has time to improve though.
Bench results:
DM-Asbestos (bot): 2.216833 / 43.259750 / 91.198372 fps -- Score = 43.284378
DM-Asbestos (flyby): 58.420170 / 116.639801 / 317.484406 fps -- Score = 116.736824
DM-Antalus (bot): 16.186348 / 38.783768 / 80.806076 fps -- Score = 38.805805
DM-Antalus (flyby): 47.760868 / 96.251801 / 392.219147 fps -- Score = 96.303078
DM-Inferno (bot): 2.267646 / 40.243225 / 160.775635 fps -- Score = 40.233208
DM-Inferno (flyby): 46.598614 / 95.578873 / 264.641663 fps -- Score = 95.687866
I'm assuming the numbers are: min / average / max fps?
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 25K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 24K
-
-
