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

what kind of 3dmark2001 scores do dualie's get?

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

dustybyrd

Senior Member
Joined
Jan 31, 2003
Location
San Francisco, CA
edit---i have converted it into a Quadro4 700 XGL and now get 11,300 3dmark2001...as well as all those goodies that comes with the Quadro...here are links to the directions:

courtesy of cruc1fy, here are the instructions:
http://forum.oc-forums.com/vb/showt...threadid=159603

http://nvworld.ru/docs/sq4e.html

i just got a fast ti4200...an asus 8420 deluxe with 128mb 3.3ns BGA RAM....

and it overclocks great to 300/650 3dmark stable...

but the 3dmark scores are about 10,500?

seems kinda low to me...as my 8500le got 8,000 at 290/525

is this about right for the ti4200 at these speeds (also dual 2.1ghz processor speed 14x150mhz)?



so far the card has been great, no artifacts at all even o/c'd

and i've noticed the winfast tv tuner card is "more stable" ...before with the ati 8500 sometimes when i changed the channel the picture would shrink to a tiny little box inside of the winfast window....now with the ti4200 it doesn't do it anymore....

not sure why, maybe the 128mb ram or faster or more stable video card?
 
Last edited:
Hrm that sounds about right. Only 1 processor is going to get used and 3dmark loves fsb (and 150 isn't exactly a high fsb).
 
That score is about right.

Don't worry though. I bet you didn't build a dually to be a professional benchmarker;)

Seriously, duallies rock for gaming because your other apps and OS don't hog CPU cycles from the game, so in many instances, games run faster, or at least smoother, on a duallie.
 
That score is about right.

Don't worry though. I bet you didn't build a dually to be a professional benchmarker;)

Seriously, duallies rock for gaming because your other apps and OS don't hog CPU cycles from the game, so in many instances, games run faster, or at least smoother, on a duallie.



yeah i don't really care about 3dmarks, it's just that i thought there would be a greater difference between the 8500 and the 4200...but there's still a 30-35% increase in speed....and hopefully the tv card and my recordings will be more stable....

cmc...i was wondering if you have used a 5x00 series card yet....and if there is a difference in using AA and other dx9 features versus your ti4600?

also, will any of the newer games require AA or other dx9 features to be turned on?

AA seems like an "eye candy" feature to me...
 
I haven't personally used a 5x00 series card, yet, but I just ordered 3 FX5600's (non-ultras), today, so I'll find out, soon how good they are.

*EDIT* I just got the FX-5600 non-Ultras. These cards seem to be a little slower than my Ti4600 at DX8, but they are DX9 capable, so that will give them a small edge.

Some new games will require DX9 features, but these are very few, right now.

AA is pretty nice, if you've got the horsepower to spare.
 
Last edited:
I haven't personally used a 5x00 series card, yet, but I just ordered 3 FX5600's (non-ultras), today, so I'll find out, soon how good they are.

Some new games will require DX9 features, but these are very few, right now.

AA is pretty nice, if you've got the horsepower to spare.


i think your ti4600 is about the same speed (or slightly faster) as the 5600 ultra or a little slower than 5900 non-ultra on dx8 games from all the benchmarks i have seen...

http://www.tomshardware.com/graphic/20030311/geforcefx-5600-5200-03.html

however, performance suffers by about 50% with AA turned on...

but with dx9 games won't they allow running them without the AA type features?
 
I'm only getting about 11,500 with my 9800pro at 420/720.
That seem right for 1.7GHz at 148fsb?

Hopefully I'll be able to slap in some 2500+ bartons in a few weeks
 
I'm only getting about 11,500 with my 9800pro at 420/720.
That seem right for 1.7GHz at 148fsb?

thanks for the info,

wow, that tells me that you can't expect to get very high 3dmark2001 scores at all with a dual AMD...and NOT to upgrade my ti4200 at ALL

probably for three very big reasons:

one: FSB is limited to 150 or less in most cases

two: top CPU speed is in general way less than single CPU ~2-2.2ghz vs. ~2.4-2.5ghz

three: the sharing of the bus speeds with a dual system...
 
yea exactly, looks like the 9800pro will be going back in the other rig once its back up and going hehe.

Although when the FPS do start to take a dive, the games/programs do seem to be a little more smooth with duals.
 
Although when the FPS do start to take a dive, the games/programs do seem to be a little more smooth with duals. /B]


if your games are smoother and run as well as in a single then
maybe you should still use the dual as the gaming rig...

3dmark2001 is still just a artificial benchmark...not real world...

and who can tell the difference between 60-70fps and 100fps anyway?
 
I love my dual for gaming because when some other app hogs some CPU time, the game doesn't flinch.

In a way, it comes down to these choices (to me): Do I want the highest FPS gaming, or the smoothest FPS gaming?
 
yea thats a good point, after playing around with it somemore it deffinately seems smoother. Can't beat that and this probably will remain my gaming rig.
I think I might just convert my 2nd rig to a dually instead of buying another NF2 lol :D
 
And btw, here is a thread that I posted.
I was interested in seeing some reference points with others with rigs close to mine.

And I have to go with cmc too. Id rather loose a few fps or even minor details to get it to run smoother without any sort of hiccuping or jumps/lags in a game.
 
Just wondering if you guys have tried halo?I was going to start messing with one of my daullys and compare with my single procs system I use for gaming.I was hoping to get a smoother game play.The only thing that holding me back is the fact I have a 9500 pro and a 9700pro,Ive read about the trouble's people have had with K7D masters and ATI.I was hoping they would resovle this soon.
 
If you guys start running 3dmark 2k3 (only do this with dx9 cards, otherwise scores can't be compared to dx8) you'll start to see that video cards really do make a huge difference and cpu's don't. In gaming the gpu is the most important part of a system, if you have 2ghz or more there is definetly dimishing returns in terms of FPS, especialy with dualies since the game can be run soly on one cpu, while other stuff (OS, background stuff etc) gets dumped on the other.
 
especialy with dualies since the game can be run soly on one cpu, while other stuff (OS, background stuff etc) gets dumped on the other. [/B]


in practice this is kind of true...

but actually a game process uses both CPU's...although not at 100% and usually not SMP enabled...~25% of each processor is used to play a non-SMP enabled game (end result is 50% or both processors is used)
 
dustybyrd said:



in practice this is kind of true...

but actually a game process uses both CPU's...although not at 100% and usually not SMP enabled...~25% of each processor is used to play a non-SMP enabled game (end result is 50% or both processors is used)

Not exactly. In a single-threaded application, each CPU will contribute about 50% of its cycles. The end result is 50% of both.
 
Back