• 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.

CPU performance differences for gaming

I understand that if processor "A" is slower overall than processor "B", then you will get a lower framerate in games using A at low resolutions.

But would you see the same lower framerate using A at higher resolutions (ie. 1280x1024)? (ie.)With everything else being equal, would the slower processor give a slower framerate at higher resolutions or would the video card become the bottleneck and show negligible performance differences from processor to processor?
 
Depends on just HOW low you talking about. For instance My comps have a 2.533Ghz 533FSB P4 and a 2.8Ghz 800FSB P4. But their cards are only a Geforce 4 MX420 and 440 (8x). When I try the 420 in both systems there is a 1 FPS frame difference in UTK2004 at 1024x768 all high settings (18FPS vs. 17FPS.) So when the graphics card is that low, the processor is not the bootleneck at all, but rather the video card.
 
Okay, say an X800XL.
I was just looking at the Anandtech HalfLife2 CPU comparison and wow, even at 1280x1024 the AMD's have such a huge advantage:
click

 
Bottlenecking
well, benchmark tests in general run top of the line optimal peripheal equipment to what they're testing. This means if they're testing cpus, they'll have really great motherboard, ram, powersupplys, etc. They want to make sure the CPU is the bottleneck. When HL2 came out I also looked similar charts at TomsHardware. I ended up by an amd 64 1800+ but I never got framerates even close to what they got even on the slowest intel. The bottleneck was other things in my system (probably memory). I got like 60-15fps. So it would make neglibile difference which processor you chose as long as it wasn't the bottleneck.

If you're talking about resolution, from my understanding of video card reviews, that usually is bottlenecked by the graphics card. I notice that my fps on HL2 doesn?t drop too too badly between large resolution changes (1024 -> 1600) as I have a 6800gt. But my memory is really poor quality so certain parts of game run laggy no matter what resolution. I believe anti-analyzing pulls on the cpu harder than resolution changes.

I?m pretty much a newbie but figured this might help you so why not post it. If anyone can correct me please do.

Specs
AMD Athlon 64 2800+ (1.8ghz default)
512 DDR MB 2700 Kingston value ram
512 DDR MB 2700 Generic ram
MSI k8t Neo Motherboard
antec sonata Case, 380 tru power
250GB HD
160 GB HD
120 GB HD
40 GB HD
Nec 1300a dvd-r drive ? packaged as mad-dog
BFG 6800 GT graphics card (default speeds)
windows xp

 
Originally posted by: rhk0327
Bottlenecking
well, benchmark tests in general run top of the line optimal peripheal equipment to what they're testing. This means if they're testing cpus, they'll have really great motherboard, ram, powersupplys, etc. They want to make sure the CPU is the bottleneck. When HL2 came out I also looked similar charts at TomsHardware. I ended up by an amd 64 1800+ but I never got framerates even close to what they got even on the slowest intel. The bottleneck was other things in my system (probably memory). I got like 60-15fps. So it would make neglibile difference which processor you chose as long as it wasn't the bottleneck.

If you're talking about resolution, from my understanding of video card reviews, that usually is bottlenecked by the graphics card. I notice that my fps on HL2 doesn?t drop too too badly between large resolution changes (1024 -> 1600) as I have a 6800gt. But my memory is really poor quality so certain parts of game run laggy no matter what resolution. I believe anti-analyzing pulls on the cpu harder than resolution changes.

I?m pretty much a newbie but figured this might help you so why not post it. If anyone can correct me please do.

Specs
AMD Athlon 64 2800+ (1.8ghz default)
512 DDR MB 2700 Kingston value ram
512 DDR MB 2700 Generic ram
MSI k8t Neo Motherboard
antec sonata Case, 380 tru power
250GB HD
160 GB HD
120 GB HD
40 GB HD
Nec 1300a dvd-r drive ? packaged as mad-dog
BFG 6800 GT graphics card (default speeds)
windows xp


Your vid card is probably the reason, not bad, just the x850xt pe beats it.

Thilan-
It has a huge advantage in the cpu limited maps (alot of triggers and objects+AI), and not so much in the graphics intense maps (less objects more scenery.) (I have the game) It all makes sence.
 
I think HL2 runs best on 2GB RAM. It chops up a bit on my setup in areas playing 1280x1024 4x/16x AA/AF
 
Yeah I was thinking that Half Life 2 is a fairly CPU intensive game (probably more so than most) so that could account for those charts.

I'm gonna be running a dual XEON at 3.0GHz with an X800XL, hopefully it won't be too much of a bottleneck.
 
Back
Top