Bottleneck

Can anyone give me a general idea of how much FPS i will loose if i pair an X1900XT with my P4 prescott @ 3.4Ghz, against a better CPU?
I keep hearing that it would be a HUGE bottleneck, but to make my decision i need a number.. at least a general one.

Thank You!

People blow "Bottlenecking" way out of proportion. Yes, a 3.4GHz Prescott will hold back some framerates, but not anything that will kill all the performance. Especially in gaming.

Yup, the crappy OCZ [/couldn't resist]

That review says that there IS a huge difference beetween CPUs!
2.8Ghz Single-core = 43FPS
3.4Ghz Dual-core = 50FPS

Such a small difference in speeds and such a big difference in FPS. Unless Dual-core matters that much, wich is unlikely since most games dont use more than one thread.

Yea, at lower settings, and resolution, you will see a noticeable difference, but with A64s, and C2Ds, no one will play at such low settings.

Your 4000+ will be fine until you need a dual core.

Anyway, I was wondering, how's the OCZ treatin' ya?

Nice.

I wish I could say that about the OCZ kit I had. Piece of crap to say the least.

Yeah, but those are useless lower res/detail benchmarks aimed at showing the power of the CPU. Look at [H]'s C2D review and they show the FX-62 completely keeping up at max playable settings despite getting cremed in the low res test settings in those other reviews.
http://enthusiast.hardocp.com/arti [...] VzaWFzdA==

Yes the 2.8GHz does take a backseat causing them to lower settings a little bit. And they didn't show dual core as making a difference in those games.

But still, your 3.4GHz, should do a pretty decent job.


So ur saying that if i run at LOW settings, the CPU DOES matter?? How can that be? If it dosent matter in the most INTENSIVE settings, how can it matter in the lowest settings?

So ur saying the 2.8Ghz DID take a back seat. So there IS a huge difference??? I thought u said it was still playable. Damn it!

And BTW, the review did not compare a Core 2 Duo and a P4, so i dont know what kind of FPS i could be getting with an E6300 compared to a P4!

Click Here!
The CPU Charts have some game benchmarks you can look at.

You do see quite big differences between FX62, and a 3.4 Prescott, however, you can't tell the difference while playing. They seem to use 1024x768 or 1280x920 with maxed out settings.

Gives you the idea.

Also, at lower setting the GPU doesn't work as much, so the CPU does the work. Kind of like the CPU tests in 3DMark, the resolution, and quality are very low.

You will see a big difference on paper, but your eyes won't notice any change, the framerates are high, thus you cannot notice the difference while playing.

You don't notice a difference between 100fps and 70fps.

IMO I think you're making a bigger deal out of this than it really is. If I had the same option as you, I'd have no qualms about pairing an X1900XT with that CPU. Unfortunately for the time being, I'm stuck with a 2600+, and an X800XL AGP, which BTW I'm still quite content with...for now. 8)

Playing a low setting is a way to remove the GPu from the equation. Since it's not under stress, only the CPU is understress so its a good way to show a CPU strenght even though nobody sane will buy a Top of the line card to run in 640X480

That makes NO sense, but whatever.

What abt at high settings? A Pentium 4 would score 10 FPS less than a Core 2 Duo at a high resolution according to the benchmarks.

That makes NO sense, but whatever.

Huh? Makes perfect sense. Just like this does...

They're just trying to help, so at least TRY to understand...I still think you're making too big a deal of this.

Like he said... it's not about whether it will or will not, but how much. In your case, not enough for you to really notice. That CPU will pair just fine with that card. That being said... where do live!? $600 is WAY too much to pay for an X1900XT. If you're willing to pay that, I've got a 9800pro I'll sell you. I bought it for $300, I'll give to you for $250...you gotta pay shipping though...

Ouch, that's expensive. If you're really that concerned about not bottlenecking your card, you could go for a 7600GT, but you're not going to get close to the performance you'd get with an X1900XT. That being said, you might want to look at getting an X1950pro instead. Either way, I see buyers remorse in your future.

I was joking...

You most certainly will not get a slide show from either the x1950pro, or the x1900xt, so take your pick. I say "buyers remorse," because you're worried about loosing a few FPS with arguably one of the most powerful cards you can currently put in your system. So, if you feel you are losing some FPS because of your CPU, you won't be happy, but if you don't buy it, you'll wish you had, because you know you could have gotten more (FPS) from it...See where I'm going with this?

I'd have no problem with pairing that card with that CPU, but then again, X1900XT's don't cost $600 here in the U.S.. You could always consider something cheaper to hold you off until you can afford a new rig.