Sign in with
Sign up | Sign in
Your question

My 5870 vs my 3 285 GTX, very interesting results!

Last response: in Graphics & Displays
Share
November 25, 2009 2:08:39 PM

I haven't been this shocked since I was able to max out half life 2 in 2004.

I benchmarked a couple of games, here is what I came up with:

System Specs:

Q9450 @ 3.8 ghz w/big water cooling
MB: EVGA sli 780i
4 gigs Ram Corsair Dominator @ 1:1
VGA: XFX 5870 @ 850 mhz (Nidia 285 GTX tri sli @ 700 mhz)
HDD: 2 xs Raptors in Raid 0
PSU: Corsair 1000HX


Crysis @ 1280x1024 4X AA, Maxed out (yes sorry my old monitor broke, so I have to use this 1 till I buy another):

5870 @ 850 mhz ( avg Framerate: 41 min Framerate: 35)
285 GTX @ 700 mhz (avg Framerate: 34 min Framerate: 21)
285 GTX sli @ 700 mhz (avg Framerate: 54 min Framerate: 24)
285 GTX tri sli @ 700 mhz (avg Framerae: 63 min Framerate: 18)

Alright? For a game that is an Nvidia came, the min frame rates suck!

Although the avg gameplay is high on the sli and tri sli, the min framerate ruins the pace of the game. I would end up at a wide opening and the Nvidia's would struggle to maintain even 30 fps, in a scene with no action, while the 5870 would stay at a stable 42 (Yes 42 fps, weird number, but its always at that!)

The same rules apply with Crysis Warhead as above.

I'm going to benchmark more now and see what I can come up with, I seriously wouldn't play Crysis just because the min frame rates would just ruin the game. Although i did beat it on my laptop with the settings on high/medium, but atleast I could justify a laptop!
a b U Graphics card
November 25, 2009 2:21:04 PM

This is why I find it rediculous that review sites rarely (a few do now) include min FPS values.

Frankly, a FPS vs time chart should be included in all reviews.

It is very missleading for places to focus on the average FPS as if that represents game play, it is one metric for sure, but not the only one.
m
0
l
!