Overclocking Madness

[dklingen]

Hi all,
I have been through this before on this forum. But this X-bit Lab's article brings into question the actual practicality of cpu performance vs game play - indirectly translated in my mind to, "Do I want to overclock my cpu?" (and yes I understand the hobby/fun rational). That is not the point. The majority of us play games at resolutions over 1024x768 with AA/AF, so the cpu becomes a non-issue unless you are pre P4 or pre Athlon64.

Here is the quote:
"So, the best gaming system configuration from the price-to-performance point of view should definitely include a powerful graphics card, which will determine not only the image quality, but the gaming comfort in general. As for the CPU, you shouldn?t spend tons of money for a high-end model. You will be able to get almost the same performance from a CPU like Athlon 64 3200+ or Pentium 4 650."

Here is the link for the whole article:
http://www.xbitlabs.com/articles/cpu/display/cpu-games2.html

 

[coomar]

hl2 though is the exception to the rule

 

[lifeguard1999]

Should have tested Far Cry...

 

[sodcha0s]

I have an A64 3000+. I started playing Q4 online the day it was available, with CPU at stock speeds. A couple of weeks ago I O/C'd my chip to 2.5G and immediately noticed the gameplay was much smoother. BTW that game is capped at 60fps as is Doom3.... HL2 single player was definitely smoother as well, not to mention windows performance. But yeah, I agree that a vid card is a more important factor in determining the level of gaming performance. However, CPU speed should not be discounted.

 

[almach1]

when the next generation games start coming out and you're cpu is not enough is it time to upgrade? No, just OC the one you have and you're good for another year before upgrading. Other do it as cost savings. Save money by buying cheap and OC'ing.

 

[Furen]

There's also the fact that some of us do things besides play games (I'd say I play games around 4-5 hours a week... kind of makes me regret buying a $300 video card but hey new hardware is nice to have). Also, looking at average framerates does not mean that the gameplay will be smooth, it's the sudden drops in framerate that are frustrating and affect gameplay, not the "average" after a 10 minute run.

EDIT: Like many people have said: "you cannot represent smoothness in a graph."

 

[Elfear]

In HL2 at 1680x1050 4AA/16AF I saw a 15-20 fps jump going from stock clocks (2.2GHz) to 2.85GHz. UT2004 is the same way. That isn't just seat of the pants either. I did some extensive benching to see whether ocing the cpu or the gpu made more of a difference. I concluded that it came down to which game I was playing.

 

[Madellga]

Normally reviews and benchmarks have the OS barebones + drivers.
They should also benchmark MINIMUM framerates, not only the maximum.
Normally the MINIMUM framerates is where we have problems.

In a normal setup, you have OS + Antivirus + Firewall + Skype + Control Panels + Atitool/Rivaturner + other stuff etc etc.

I've decided also to go dual core, me thinks for single threaded games that the second core will help taking care of the other stuff - so the first one will be able to do a better job.