512 DDR400 or 1gb DDR266

Hi all,
O.K., here is the question:
What do you think will give better performance; 1 stick of 512mb Corsair Extreme DDR400 cas2, or 2 sticks of 512mb Centon DDR266 cas2 for a total of 1gb RAM.
I am using the Asus P4S8X mobo.
I know that some boards will perform better with 2 sticks of RAM instead of just 1 stick. Something about the double bandwidth.
Now before you answer you need to know that on my board the Corsair will not run at any higher than a setting of 166 fbs at 2-3-3-7-2T.
I have already spoken with tech support at both Corsair & Asus, and the only way to fix this problem is to send the board back to Asus and let them do something to it. I can’t go without a computer for that long.
I will be benching both configurations just to see the difference, but I thought I would post this just to see what you guys thought before I post the results.
For more detail on my system see my sig.

Only a dual channel motherboard will benefit from multiple sticks of memory. Otherwise multiple sticks just makes it that much harder to overclock.

So is my board dual channel???
I looked at the website but didn't see any reference to "dual channel". Would it say "dual channel" or something else???
http://usa.asus.com/mb/socket478/p4s8x/overview.htm

i dont think ur mobo has dual channel (i also think only nforce mobos have dual channel memory)

im not sure about this info

Well, the results are in.
The 1gb of DDR266 out-performed the 512 of DDR400 on my system.
Running the system at stock speeds, the DDR400 got me 13800 on 3DMark2001, while the DDR266 got me (drumroll please) 14104!!!!
HELLO 14K!!!! And all at STOCK SPEEDS!!!!
http://service.futuremark.com/compare?2k1=5546532

good job....you better update your sig

Why don't you test it overclocked? If you are planning on running them at 166fsb, the Corsair might be stable while the Centon could cause problems, being rated for 133fsb. And even if you don't reach 166fsb (which would give you 3.5GHz), the Centon still might not be stable at speeds higher than it's rated for, but you should test it. This is for if you running overclocked. If not, then you might as well go for the gig since it's giving you better results.

do you run your memory at spd settings or do you specify them yourself? I don't think swapping out 2 kinds of memory would make any difference at all but if you run spd settings then its possible your 266 memory had more aggressive timings, thus increasing your 3dmark scores. Try forcing timings to the same number for both tests and see how the scores are.

Oh and as for dual channel I think there will be some Intel dual channel mobos coming out but currently the nforce(not so good performance) and nforce2 chipsets are the only ones that have it.

First to Tipycol:
I will very soon be building a new system (promie/water cooled 3.0 P4) that I will O/C the $HIT out of. This system will then be given to my girlfriend. So, I don't want to o/c it and void any warrenties or worse yet, completely fry something.

As for Deathknight:
Yes, I was running both the DDR400 & the DDR266 at SPD settings for the above reason.

I honestly think the only reason that I got such better performance out of the DDR266 is because my board is having some troubles running the DDR400 at it's maxed settings. The Asus Tech guy said that I would have to send my board in for them to fix the problem. He also said that some of the P4S8X boards work fine w/DDR400 while some just freak out. I am sure that if my board was able to run the DDR400 the way it is supposed to, The results of this test would be much different.