• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Hypothetically -- Building the fastest Gaming PC

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Pinky

Member
Joined
Apr 21, 2001
Location
Las Vegas, NV
http://service.futuremark.com/compare?3dm05=466765

Looks like 12634 is the 3dMark 2005 score to beat. Damn that's high!! :p

All of the top scores are with VERY overclocked A64 processors and dual (SLI) 6800 ultras. Also appears a promethia or better is needed for cooling of the proc, and a good watercooler on the video card.

Someone from this forum has been nice enough to give my friend an original promethia and he has existing watercooling for the video card. Now we need funds and the correct components.

Well, let's say we have the funding (which we do). We're looking at which motherboards right now? Only a handful have SLI... Then securing an A64 at a reasonable price will be the next hurdle. While expensive, dual 6800 ultras shouldn't be hard to find. Mail order prices of course. ;)

The idea is to mod the gear (for use with the promethia and H2O cooling), do the suicide run, then test the components at stock before reselling them. This way we wet our whistle but only eat maybe a few hundred. If it all blew up in our faces it's not too big a deal. Losing any amount of money sux, but it's found money anyway...

I'm guessing everything needed (proc, mb, dual 6800 vid cards, ram) should cost less than $2,000. Could shave some of the initial cost if we find some of this used. We would obviously be documenting the experiment.
 
Go get em pink...I will be loking closely when the time to tear down arises :)
 
are these 3 "Detailed Test Results

Game Tests
GT1 - Return To Proxycon 50.8 fps
GT2 - Firefly Forest 37.3 fps
GT3 - Canyon Flight 68.1 fps" enough to get a accurate 3dmark ? if all were run do you think it'd be exactly the same ? or a bit more or less ?
 
DDR-PIII said:
are these 3 "Detailed Test Results

Game Tests
GT1 - Return To Proxycon 50.8 fps
GT2 - Firefly Forest 37.3 fps
GT3 - Canyon Flight 68.1 fps" enough to get a accurate 3dmark ? if all were run do you think it'd be exactly the same ? or a bit more or less ?

I thought 3dmark only uses the first tests, the rest are just there for testing sake and don't factor into the score?
 
cetoole said:
Looks like you would need a couple cascades to set any records these days.

yea.. if your looking for top 3dmark, youll need atleast a cascade, maybe a double, or even a triple :eek:

a single prommy isnt gonna get ya near cold enough to get the cpu and vid card speeds high enough to break top scores. (esp unmodded!! r404a atleast)

I'd look for the new DFI mobo... it seems to be really tearing it up with the NFIII, and teh NF4 is gonna be even better.
 
techun said:
only 512 of ram..odd

Fewer individual ram chips means less to overclock/go wrong/hold you back. ;)

I believe 512MB is the minimum for 3dMark to run at it's best performance with all the textures loaded.
 
threeme2189 said:
i say do it!!!!
man thats gonna be one he\_\_ of a kickass pc...

While it will kick some serious butt, it probably won't stomp the existing high score. Maybe in the top ten at best (depending on how well the cpu clocks, etc). Not worth it just for a brief top 10 shot.

Was thinking liquid nitrogen for the suicide run, but on the other hand I'm not feeling THAT lucky! ;)
 
Back