Asus P5W way faster than Bad Axe

[gocchin]

At least for me it has been.

I keep saying it and I can't figure out why but the Asus P5W DH is WAY WAY
faster than the Bad Axe 304. Unless I have the Bad Axe setup wrong, with
exact same hardware and settings. The Asus gives the Bad Axe a whipping
in Super Pi 32M. However looking around the forum, my Bad Axe 32m times
are in line with other members. While on the other hand, my Asus times are
similar to Kyosen-san's Asus results at similar clocks.

Both at 3.5ghz with 1.4v and exact same ram and ram timings at 2.2v.
Nearly 1 min 45s difference at 3.5ghz.

Can't quite figure it out....

Asus:


Bad Axe:

[bachus_anonym]

What's the strap on that BadAxe run, gocchin? 1333fsb or 1066fsb?

[makatee]

I thought I noticed a small difference myself, but now Im gonna have to do a side by side comparison.....That is a HUGE difference....I didnt even think to check and see if it was this dramatic....Ill post back later and see if my results are similar. But a quick question....what's your strap on the BadAxe? That can make a world of a difference...check some of the Xtreme bandwith threads to read about the quirks....Not that you wouldnt think to check this, im just trying to help any way i can

[gocchin]

Hmmmm I had it forced at 1333fsb... I better go read.

[mine]

... rather interesting
same is refered here by some of our forum members
but as far as I remember noone has done an exact side by side comparison

and ..hmmm this is huge , isn`t it

will you try a 1066 fsb run ?
think it will be much closer
though I think the Asus will still be faster

[makatee]

... rather interesting
same is refered here by some of our forum members
but as far as I remember noone has done an exact side by side comparison

..hmmm this is quite huge , isn`t it

Though its only a glimpse right now of one benchmark, the difference is huge relatively speaking.

[bachus_anonym]

Hmmmm I had it forced at 1333fsb... I better go read.

That's exacty why It seems that forcing 1333fsb strap on BadAxe does "something" to this board that allows for supposedly HUGE memory OCs. Now, there are two options:

1. either 1333fsb strap looses some MCH settings to the point that memory performance is just crap but allows for some unreal frequecies of up to DDR-1100 or DDR2-1200 8-3-3-3.0 or 8-2-2-3.0 or
2. none of the monitoring software can detect the real speed and/or latencies the RAM is actually running at

Your finding seems to prove that at forced 1333fsb, BadAxe seems to have some horrible memory performace. Thanks for posting that

[theteamaqua]

lol maybe that extra 4 Mhz on Asus might made the difference :P j/k, but possible

also is it possible for this kyeosen to run other test like 3d mark, pc mark ... and such??

[gocchin]

lol maybe that extra 4 Mhz on Asus might made the difference :P j/k, but possible

also is it possible for this kyeosen to run other test like 3d mark, pc mark ... and such??

Those are my results above I was just mentioning that Kyosen had similar results. I already posted my prelimnary 3D results in the 3D section .

[gocchin]

That's exacty why It seems that forcing 1333fsb strap on BadAxe does "something" to this board that allows for supposedly HUGE memory OCs. Now, there are either two options:

1. either 1333fsb strap looses some MCH settings to the point that memory performance is just crap but allows for some unreal frequecies of up to DDR-1100 or DDR2-1200 8-3-3-3.0 or 8-2-2-3.0 or
2. none of the monitoring software can detect the real speed and/or latencies the RAM is actually running at

Your finding seems to prove that at forced 1333fsb, BadAxe seems to have some horrible memory performace. Thanks for posting that

Anytime bachus... we're all here to learn . Well I'm going to run 32m again this time back at 1067 and see what the time is.

[makatee]

That's exacty why It seems that forcing 1333fsb strap on BadAxe does "something" to this board that allows for supposedly HUGE memory OCs. Now, there are either two options:

1. either 1333fsb strap looses some MCH settings to the point that memory performance is just crap but allows for some unreal frequecies of up to DDR-1100 or DDR2-1200 8-3-3-3.0 or 8-2-2-3.0 or
2. none of the monitoring software can detect the real speed or latencies the RAM is actually running at

Your finding seems to prove that at forced 1333fsb, BadAxe seems to have some horrible memory performace. Thanks for posting that

This certainly does seem to be the missing variable. Though I think we do need to try a side by side comparison of some of these boards (I have the BadAxe and the Asus, but I wouldnt mind putting some numbers together) Anyways, I await to hear if Gocchin can pull the BadAxe numbers closer or if his Asus is smokin it

[eva2000]

gocchin, try badaxe at 1067fsb and 350fsb 4:5

this was done at 2.2v vdimm 4-4-3-15 with 2x256MB Centon PC4300 Advance LOL with 1067fsb strap.. not sure how much a penalty 2x256mb would be heh



for conroe what's the diff between 2x256mb vs 2x512mb vs 2x1GB for 32m ?

also timings for tRAS differ it seems for your runs ?

[gocchin]

I'm running 1066 with the same settings now and after a few loops its already faster... give me 10 more minutes to post a result.

[gocchin]

Interesting... with everything the same except using 1066 I was UNABLE to complete a 32M run. I had to go back in the bios and raise VMCH a notch to run it again.

[mine]

while waiting .... what a nice girl you are goccin and such a good oc er :-)

[i||uSi0n^]

while waiting .... what a nice girl you are goccin and such a good oc er :-)

lol if gocchin is a girl really...im in love with her!

i||uSi0n^

[Metroid]

while waiting .... what a nice girl you are goccin and such a good oc er :-)

For all what I have seen so far it depends of the memory, options and software.

Really for you Asus is better in any way.

I wonder if she is a girl.

[mine]

hee... I was first

[ZeroX]

So gocchin, by ur experience, is the Asus better than the Badaxe?

A girl that likes OCing... i want one! j/k

[gocchin]

For all what I have seen so far it depends of the memory, options and software.

Really for you Asus is better in any way.

Well every factor is the same in the above screen shots except for the mobos.

Anyway back to the testing.

1067 is indeed way faster than 1333 bootstrap but still lagging behind
the asus by quite a bit:

[hicookie]

maybe should check memset?
afraid some mem timing different

[zicl]

i have been looking at both these Motherboards, to be honest i think i have made my mind up on the Asus.

As for the chip i am looking at a 6700, and hopefully the G Skill 6400 2gb set.

together with the X1900 in crossfire...i cant wait!!

[mine]

now if andré with the P5 de luxe and the 965 chipset could chime in

we would have a nice comparison 975 vs. 965 @ 32 M

and we could see whats up with the vastly improved new intel 965 memory controler ............

[eva2000]

gocchin is your asus p5w dh volt modded ?

[ted3]

Thats bad, hope intel can fix this 1333 strap error with a Bios update. Good to know anyway that Bad Axe doesnt have terrible performance overall, thanx for the research guys!

OT: Gocchin in that avatar is some kind of Japanese pop star, my guess is that Gocchin@xtremesystems is one of her biggest fans