Asus 680i, P5N32-E Sli Results.

[charlie]

hmmmmm... wonder how much of this is fixable with a BIOS update?

[situman]

I highly doubt that is vdroop being that much of a difference. It has to be some type of bug.

for 300 bucks, it better be only a bug.

[ewitte]

for 300 bucks, it better be only a bug.

You might want to specifiy that as a BIOS bug Because they could have screwed something up witht he design too.

[charlie]

What's up with these $300-$400 top drawer boards?? The Stryker is on pre-order at $400 at zzf.

A couple years ago... Abit's top shelf 875p was the IC7-MAX3, the IC7-G and the IC7...

The vanilla IC7 was around $120 with the MAX3 around $175.

Have capacitors, plastic and pcb DOUBLED in price in the last 3 years?? We certainly know the BIOS programmers are no better than they were 3 years ago...

[ineedaname]

\Have capacitors, plastic and pcb DOUBLED in price in the last 3 years?? We certainly know the BIOS programmers are no better than they were 3 years ago...

Well as a side note the striker uses a cap-less design.

It looks like the P5N32-E needs a bios update or a new revision. That "Droop" problem is pretty serious. I agree that it has to be some kinda bug.

[Kunaak]

I think Vcore Aside... people should be looking at the actual overclocking results of this board so far...

1. hits well above 500, with some trial and error. (still testing)
2. runs 1T exceptionally well, easily into 900 mhz, for my cheap ram.
3. runs 1200 mhz extremely easily. ram thats always been "Kinda" stable at 1200 on my P5B-Deluxe, easily does 1200 stable on this board.

only real problem with overclocking on this board so far... I have no way of showing results from ram overclocking.

CPUZ, CPUID, Sisoftsandra, Everest, NTune, crystalmark, AI Boost...

nothing works with this board to show ram speed, or cas, cause this boards not using the standard Nvidia Reference bios, so even Nvidias own programs dont work on this board.

thats my biggest problem so far.

[pentium777]

I can run 390x9 for 3.5G on my E6600, would I benefit much doing 439x8 instead? Or even lowering to 7x multi?

Not interested in very small benching scores but if I can make a fairly decent difference in 3dm06 or real world gaming then I'd be interested.

This is on a BFG 680i BTW.

Thanks

[Kunaak]

What's up with these $300-$400 top drawer boards?? The Stryker is on pre-order at $400 at zzf.

A couple years ago... Abit's top shelf 875p was the IC7-MAX3, the IC7-G and the IC7...

The vanilla IC7 was around $120 with the MAX3 around $175.

Have capacitors, plastic and pcb DOUBLED in price in the last 3 years?? We certainly know the BIOS programmers are no better than they were 3 years ago...

not that I am advocating these boards prices, I certainly dont think they should be in the $300-400 range.
I got it, cause I am just ready for Anything other then another 975 or 965 board.
the lack of a working 4:5 divider for my P5B deluxe, and the complete and total lack of any ability to recover from a failed overclock, without having to completly clear the CMOS, is really really annoying.
but the P5B Deluxe, is still a damn solid board and a damn fine competitor for this board, so its hard to say 1 of these is better then the other.
the P5B Deluxe will reach 550+ Easier then the 680i will.
but the Intel chipsets are really really limited in thier ram divider options, which the 680i definatly has no problems in this area.

anyways, I am losing track here...

the IC7 days were the best for back then.
but boards these days have definatly greatly increased in complexity and design, like heat transfering PCB, three 16x PCIE lanes, digital PWM's.

like I said, I am not condoning the price of these boards, but I can definatly see why they cost more, cause the complexity of these new high end boards is far higher then boards a few years ago...

[Senater_Cache]

Well as a side note the striker uses a cap-less design.

It looks like the P5N32-E needs a bios update or a new revision. That "Droop" problem is pretty serious. I agree that it has to be some kinda bug.

actually the stryker is littered with caps all over the board.

although they are solid-caps, they should not drive the price up THAT high.


BTW, th stryker does not have Digital PWM AFAIK. It looks like it still has mosfets and coils and all. are they using single inductors vs DFI's and ABIT's 5 stage chips?
can anyone confirm?

[lowfat]

What's up with these $300-$400 top drawer boards?? The Stryker is on pre-order at $400 at zzf.

A couple years ago... Abit's top shelf 875p was the IC7-MAX3, the IC7-G and the IC7...

you americans are getting ripped. The p5n32-e is for sale for $280 canadian

[Millyons]

kunnak can you compare cpu OC at same voltage between the p5bdlx and this one......thanks

[ToXqc]

I bought mine yesterday for 299$ canadien

[Kunaak]

kunnak can you compare cpu OC at same voltage between the p5bdlx and this one......thanks

theres no difference.

it took me 1.35 volts (real volts, not whats set in bios) with the P5B Deluxe to get 7x500 stable.
its the same on the 680i, so theres no real difference for that for me.

but I have found one hell of a new.... well... not sure what you could call it...



on my P5B Deluxe, I use to need 2.8 volts to run 1200 mhz stable with my Mushkin PC6400, cas 5-4-4-12.

on the Asus 680i, I am doing the same thing... at 2.45 volts now.
I dont know why...
I was getting weird results, cause I was bumping the volts to 2.65 volts and it was almost stable, then 2.75 and it became even more unstable... then 2.8 and it got even worse.

so I, for the hell of went the opposite way... and just set 2.4 volts in bios (which cause of the overvolt, gets 2.45 volts real)

and now its stable...
I will have a pic of this soon.

so far, this board has a huge learning curve, but the results are very good.

[Millyons]

wow thats really interesting

[Kunaak]

oh my god...

you guys arent gonna believe this.
its now running 1300 mhz cas 5-4-4-12, TRC 6, at 2.45 volts.
32M is now running.

(my Test PC has the 680i, my Main PC, this one, is my P5B Deluxe... since I still use Crossfire X1950's)

aaaahhhhhhhh.... 1300 mhz....

why isnt there anything that will work with this board like CPUZ or Everest....... damn it, damn it, damn it...

[althes]

Hopefully a couple people can get the board and we can see what happens to others.

[DeToNaToR.cl]

This board is based on the nvidia reference desing or it has other desing by asus?

[[XC] gomeler]

What sticks are you using to hit these amazing speeds? It sounds like the 680i just needs to have a few bugs worked out and it'll be an amazing chipset, hopefully the prices will drop to somewhat acceptable levels also. It really needs more than 1.47 volts to be competitive in the xtreme section, my crappy DS3 laughs at the pathetic 1.6 bios volt.

[cpulloverclock]

I bought mine yesterday for 299$ canadien

ncix the best

[msgclb]

why isnt there anything that will work with this board like CPUZ or Everest....... damn it, damn it, damn it...

My suggestion is that you contact the developers for these tools and see if you can find out if they are going to support the 680i. Your experiences and needs might get the ball rolling.

[Millyons]

oh my god...

you guys arent gonna believe this.
its now running 1300 mhz cas 5-4-4-12, TRC 6, at 2.45 volts.
32M is now running.

(my Test PC has the 680i, my Main PC, this one, is my P5B Deluxe... since I still use Crossfire X1950's)

aaaahhhhhhhh.... 1300 mhz....

why isnt there anything that will work with this board like CPUZ or Everest....... damn it, damn it, damn it...

man hopefully its not kicking in some devider and running the ram at less than what you think, its pretty strange for ram to need that much voltage less than on another board

[Kunaak]

man hopefully its not kicking in some devider and running the ram at less than what you think, its pretty strange for ram to need that much voltage less than on another board

Bios and Ntune match up.
what I set in bios, is what Ntune shows.

Ntune doesnt show much... like voltages or cas.
but it does show MHz.
and the speeds match what I set in bios, and the results are linear.
1300 is faster then 1200 and such. I've already confirmed it.
getting pics ready for a post about it...

by the way...

1333 mhz is the top I got.
unstable.
cas 5-4-4-12, 2.45 volts.
gonna try cas 5-5-5-15 see if 1333 mhz can get stabilized.

[shogan191]

Kunaak is the main difference between this boad and the Stryker just the LEDs and switches?

[charlie]

So let me get this straight...

Years ago I wished for a motherboard that allowed independant FSB/Memory OC'ing... if your 6600 tops out at 3.6gHZ (9 x 400) and your D9 tops out at 557 you can set the FSB and mhz independantly??

And no I'm not talking about dividers, but separate PLL's to adjust both? hmmm....

[Kunaak]

So let me get this straight...

Years ago I wished for a motherboard that allowed independant FSB/Memory OC'ing... if your 6600 tops out at 3.6gHZ (9 x 400) and your D9 tops out at 557 you can set the FSB and mhz independantly??

And no I'm not talking about dividers, but separate PLL's to adjust both? hmmm....

yes, you can.

you just select "Unlinked" then you have a whole range of new dividers to use.

you set in any ram speed you want, reguardless of the FSB, and it will set the closest divider possible.

it's usually really really close to the speed you want.

even numbers are always perfect.
strange numbers, always results in "Close enough".

like set 1224 mhz, and you might get 1233 mhz, or 1214 mhz.

really odd mhz, results in not exact MHz, but its alot more useful then the usual 3-4 dividers available.
which, are still there if you want to use those too.
those are the dividers you can set manually.
the truely odd dividers, that come from the use of "unlinked" those aren't selectable, and are determined by the ram speed you try to use.
which works just fine for me, cause all I want is ram speeds that are close to the speed I ask for.

right now, I set a 500 FSB, and I can select any range of ram speeds from 667, to 1300 mhz, and it all works perfectly in Unlinked mode, and I dont have to make any changes to the FSB to get the speeds I want.

by the way, the multiplier on these boards seems to have zero effect on how well they OC.

like with a E6700... try setting 7 x 500 and your running over 2980 mhz on the NBCC, which is garaunteed to fail on those boards (965/975).

on this board... doesnt seem to matter one bit.
any multiplier downward, just works.
so 7 x 500 on a E6700... should work on these boards just fine, without having to worry about the NBCC.

atleast... I think so, I mean, it works....
what speed the NBCC is on these boards... I have no idea.

I have no tool available that works for reading anything from these boards, other then voltage from Asus probe, and Ntune to read the actual ram MHz.... nothing else.

definatly need alot of work in the program support, before these boards can be used by reviewers and such.
with no way to show Cas, or MHz (other then Ntune) its just really hard to do much more then say what your results are....