Sign in with
Sign up | Sign in
Your question

[ASUS P5N-E SLI] - Cannot get past 300 FSB with Q6600

Last response: in Overclocking
Share
January 31, 2008 2:38:02 PM

All -
Long time reader (reason why I purchased what I have) first time poster. Ive searched the forums and saw a few others having this problem, but no solution.

I have tried to overclock my P5N-E SLI Q6600 past an FSB of 300 (QDR =1200). Ive increased the voltage to Vcore (1.40) North (1.59), and memory (2.01).
All other bios options are their defaults.

On reboot it freezes at the startup screen and I cannot get it to pass POST. CPR does not kick in and i have to clear CMOS every time.

Whats interesting is if I use the P5N-E AI OC and select 20% i can boot into windows, but my memory is overclocked and thus the system becomes unstable, so I know my system can get past the 300 FSB.

Any ideas are welcome, as I have spent extra on fans and watercooling because of the ability for this chip+mobo combo for OC.

Thanks

P.S Memory is 2x1Gb Corsair XMS2 PC6400 DDR2 Sticks in the Black slots.
January 31, 2008 2:52:33 PM

that north bridge voltage seems to high to me....but maybe it's not for your board....mine runs @ 1.4v

How are you running your ram? 2 gigs or 4?
And what FSBs have you tried...sometimes going up in small increments doesn't work...try 333, or 350.

Could also bump your ram voltage to 2.1v....are you getting good power from your PSU.

You'll probably have to do some research to figure out the best manual memory timings to set in the BIOS as well as which settings to disable...there are quite a few processor functions that don't really add performance but can easily make an OC fail. (at least that was the case with my 680i board).

January 31, 2008 2:58:54 PM

Thank you for the reply.

Im running 2 x 1Gb sticks with stock timings (I believe 5-5-5-12) and voltage. Ive tried 325 (QDR of 1300), and 333 (QDR of 1333). Maybe go higher?

I can bump up my voltage to my memory. Would loosening the memory timings help on a POST? I havent even been able to boot to windows for stability testing. If so, what timings are used as a safe bet for OC before people try to tighten the timings?

Related resources
January 31, 2008 3:10:31 PM

You could try higher, I mean you can always check your temps once you're booted.

try 350....

Memory speed and timings seem to give me the most trouble....

I'd run your ram unlinked and try a mild oc...but loosen the timings a little..5-5-5-15 or 5-5-5-18....with a little more voltage. Is it DDR2 800 or 1066....try your 800 @ like 900mhz (or the closest you can come to that for a good ratio if you want)

have the latest BIOS?
January 31, 2008 3:15:27 PM

If you have access to these setting then

Set Limit CPUID Maxval to Disable

Set CPU Thermal Control to Disable

Set C1E Enhanced Halt State to Disable

Set Intel Speedstep to Disable

Set any CPU, SATA, PCIe or any other "Spread Spectrum" variable to disable

(If you want you can try to re-enable some of these later after you have a stable OC to see if they cause issues)

January 31, 2008 3:15:56 PM

again thank you for your reply Plastic.

I could try 350 and check the temps. I will also loosen timings. Could I REALLY loosen them? Like 6-6-6-15 or 7-7-7-18? Just to make sure?

I have DDR2 800. If I OC the memory to 900, how do i get it to be stable? A combination of more memory voltage/looser timings?

and no, I am using 0703 (latest is 0803).
January 31, 2008 3:32:04 PM

You shouldn't really have to loosen the timings that much....especially if the timings they suggest to use are rather tight.

You'll know your memory is stable from stress testing your system later.

I've had my DDR2 800 running:

960mhz @ 5-5-5-15
and 900mhz @ 4-4-4-12

still trying to determine which is better...

the ram spec was 1.9v but I found 2.075 to be necessary for stability....there should be some kind of heat spreader on your ram...if there is and you have semi-decent airflow in your case I wouldn't worry about a mild OC on it.

I've seen people run DDR2 800 at 1000-1100mhz without problems.

And that 1.59v to your northbridge should be re-examined...your NB just might not be able to handle the temps over 300 @ that voltage....come down to 1.4v and see if that helps (too much voltage is as much of a killer as too little)

an good OC takes patience..hang in there
January 31, 2008 5:09:11 PM

Thank you again Plastic for your reply.

Supposedly there are heat spreaders on my sticks. For my info, what do they look like?

Ill drop my Northbridge back down (I read that could be an issue for stability).

So do you think 5-5-5-15 is loose enough for OC stability at this point (i.e. trying to get past the 300 FSB)?

January 31, 2008 6:19:10 PM

no problem on the replies, I enjoy talking about hardware a bit too much I think...

If you don't just see the Silicon with the ram chips on it...then they've got heat spreaders....just a metal clasp that goes over the whole stick to bring the heat from the ram chips to the surface and distribute it evenly so that air flow can take it out of the case.

and yes I think 5-5-5-15 will be enough....just remember you have to unlink your ram from the FSB so that when you're pushing 1333mhz or more on the FSB your ram isn't going with it and find it's limit...

or if it is linked make sure it's in a ratio that won't make your RAM go past say 900mhz or so.

I also noticed that your vcore on the CPU was 1.4v....that's a good number for testing in my opinion because on a G0 it will run most any clock you can achieve with air cooling....but

you may find that your temps get too hot when you stress all 4 cores @ 100%...so when you get some OC you're happy with.....say 3.4 ghz....try coming down on this number....

I wouldn't be surprised if a G0 q6600 would run @ 3.4 ghz with of vcore of 1.3 or even lower....
January 31, 2008 6:28:41 PM

I was thinking the same thing. Thats why I am a little confused on why I cant push past 300 FSB. I read that there is a frequency whole at around 400, but I am not there yet.

Should I push the memory voltage up to over 2? what about the northbridge?

As of now i am just trying to get it to boot to windows, let alone stress test and temp test the system.
January 31, 2008 6:40:41 PM

as I've suggested try your ram voltage @ a little over 2

I don't know what your increments are....try 2.1v

and your NB at 1.4v or 1.45v......1.5 should be about the max you want to go. Unless you have some kickin' active cooling on your NB your temps will be pretty warm at 1.4v already

let us know how it goes...

just FYI, my work buddy has his Asus Maximus board runnings @ 1800 FSB rock solid...crazy
February 1, 2008 1:59:59 PM

Update:
Tried to OC again after work with setting the LDT to 3x, disabling the execute bit, and setting the northbridge voltage to 1.39, vcore to 1.3625, and vdimm to 2.01. I also manually set my memory to 5-5-5-15-2T.

After all of this, I was able to get past the POST screen (yay!)...BUT... I could not boot to windows. The only good thing about this is now when I reboot I dont have to reset the CMOS.

I dont know what to try next. The last thing I did before bed was to flash my bios to 608 on this board (I read that 703 was difficult to OC). I havent tried to OC the new bios.

Any ideas are welcome.

Thanks.
February 1, 2008 3:10:42 PM

This is a tough nut to crack...the new BIOS may help....anyone else wanna take a stab at this...

do you mean 2.1v or 2.01v on your Memory?
and what is the clock on your memory?
a b K Overclocking
February 1, 2008 3:27:49 PM

just try a vcore of 1.5v nb of 1.6 mem 2.0 and see how fast it can go then when you hav reached a speed thats good then lower the voltages, when it doesnt get into windown usuwally means not enought voltage. i got a q6600 at 3.6 on wc but i still need 1.5v in bios and 1.46 by cpu-z, dont be supprised all chips need diffrent voltages, if you have good cooling then you can easly push up to 1.5v.
February 1, 2008 3:32:45 PM

richardscott said:
just try a vcore of 1.5v nb of 1.6 mem 2.0 and see how fast it can go then when you hav reached a speed thats good then lower the voltages, when it doesnt get into windown usuwally means not enought voltage. i got a q6600 at 3.6 on wc but i still need 1.5v in bios and 1.46 by cpu-z, dont be supprised all chips need diffrent voltages, if you have good cooling then you can easly push up to 1.5v.


I was just about to come pack and say something about his CPU voltage....

I had a complete brainfart and realized that BIOS voltage is not a great indicator....

he said he's @ 1.3625v (I assume that's in BIOS)

SO yeah raise your CPU voltage to at least say 1.45v and see what is reported by CPUz later.

Your inability to load windows suggests vdroop, so give it more volts that you think you need...get stable and then come back down incrementally until you have problems again...then go back up a step or two
February 1, 2008 3:37:29 PM

thank you for your replies.

I have a screen cap and it says in cpu-z my vcore is 1.296 when i set in bios 1.3625... (.06 vdroop?).
February 1, 2008 3:39:41 PM

PlasticSashimi said:
This is a tough nut to crack...the new BIOS may help....anyone else wanna take a stab at this...

do you mean 2.1v or 2.01v on your Memory?
and what is the clock on your memory?



vdimm is set to about 2.01, the bios only let me set it at discrete values. I think the next is 2.09ish...

memory clock is at ddr2 800... ive locked that.
a b K Overclocking
February 1, 2008 3:48:32 PM

ok you want to like ure mem to ure fsb its more stable that way, second up ure vcore to 1.5 nb to 1.6, if you have good cooling then you can jump a big step an go for say 350 fsb and see if its stable or posts then get back if it doesnt and ill see what else you can do.
February 2, 2008 1:18:28 AM

I am having the same problem with my P5N-E SLI and Q6600. I can put in 1199 for my FSB (which is 299) and be rock solid stable but if I put in 1200 or more, I can get past POST but never into windows. I've been spending hours upping the FSB by one at a time to get past any hole and the only other spot I got into windows was at 1531 (not 1532 or 1530, but only 1531) but that would crash out in stress testing.

By the way, to avoid CMOS reset on this board, I've found you can hit your reset button about 10 times in 10 seconds and then turn it off by holding power switch, it will reset BIOS to last spot it was able to POST at.

Anyway, I'm beginning to think that it's something beyond changing my numbers cause I've been fiddling with VCore settings and FSB settings one tick at a time and still nothing. It almost seems as if this board is killing the voltage when a power drain starts. Maybe that's what people are talking about with the VDrop or VDroop... whatever that is.

I'd sure be interested in hearing from anyone who's actually figured it out. I've been checking overclocking forums and it seems this is a common problem with the P5N-E board.
a b K Overclocking
February 2, 2008 10:01:33 AM

well i have had 2 asus boards and the easy was to reset the overclocking settings when you dont post is swithc on then turn off at plug after about 10 secs then turn back on and it will disable ure overclock settings, second if you cannot get into windown after a sucessful post is either due not not enuf cooling or vdroop ie u need more vcore, dont think you can reach the same overclocks as other people useing the same settings it just wont happen. the best q6600 to overclock were the first ones that came out. g0 that is
February 4, 2008 8:55:56 PM

Hey guys, I'm having the same problem!!! I used to have a E6300, and I was able to overclock to 3GHZ without any problems! Now I got the Q6700 and the max Im getting is 1110 on fsb. I flash the BIOS to the latest 0801. I have no clue!!!

Thanks for the help in advance!
February 5, 2008 4:43:10 AM

Any update on this front? Unfortunately I am TDY until next week so I cant try out any settings...

Any other insight would be greatly appreciated.
a b K Overclocking
February 5, 2008 7:29:05 AM

VDroop is the diffrence in voltage in the core between full load and idle, vdroop will cause a crash on your computer useually when booting into windows because it posts fine with no cpu load, then as soon as it boots windows it crashes cos vdroop drops the voltage and it becomes unstable.

Link your memory to your fsb in the link mode this will be more stable than unlinked mode.

Disable all cpu options like speed step, etc aslong as you dont go crazy on the voltages or running your cpu to hot then you wont fry your computer.

Relax the timeings on your memory eg if they are ment to be 4,4,4,12 then 5,5,5,15 will probally do.

input a vcore voltage of say 1.4,1.45 or even 1.5 if you fell sefe going that high, then overclock and see how fast you can go, then once your stable you can reduce the voltage, and heat output.
February 13, 2008 11:08:30 AM

Hehe, I have the same problems, except mine is stable until about 318 FSB then all hell breaks lose. No posting up to 1485 FSB! Then it fails to boot into windows.
Oh, its a b3 q6600, and believe it or not, I accidentally booted all the way into windows at 2870ish Mhz @ Vcore setting of 1.81250!!! Was supposed to be 1.31250. Booted fine, loaded all the suaul progies, CPUZ, Core temp, HWMonitor... and in HW Monotor my VCore was at 1.78, and CPUz was 1.813! Wow, I thought the maximum was 1.5 volts?? Haha, guess not, temps were at 89C per core! Wonder if I should try and OC at that Vcore? hehehe, nothing else works!

--Lupi!
a b K Overclocking
February 13, 2008 3:18:10 PM

errr no because you will fry your cpu. stick to 1.5 and below
February 14, 2008 8:27:28 AM

Hehe, yeah, was being sarcastic, but its still funny. I didnt think it would load vista at those voltages, let alone happily idle like nothing was wrong. Anyways, I am thinking that it is more a hard ware issue when overclocked. Dunno why. My BSOD is usually the IRQL Not Less Or Equal. I have always had trouble with my xtreme n card, and the thing is on the same IRQ as my Ultra. And you know when windows re sizes your monitor to suit your vista resolution, and just after you enter your password and log in, it kinda fades to black then loads yer icons and junk. It seems to freeze at either one of these two actions and fork over that IRQL error. Dunno if thats the cause, but it sounds better than anything else I have tried! And I get tired of the wireless card bombing out when vista discreetly updates its driver. it only works with the one from 06, hehe.
February 27, 2008 11:58:28 PM

Seems alot of people with that mobo and CPU got the same problems. I had very similar problems. I was unable to find any answers to my problems and I got so frustrated I ditched that mobo and got a gigabyte ep35c ds3r. Now I'm OC'ed at 3.0 ghz with temps under 50c(43+- by 2c) running prime95 stable.
January 12, 2010 3:02:08 PM

Anyone figured out this issue in the last couple of years with going above 300mhz on the cpu FSB on a Q6600 on an nforce 680i SLI mobo.

I've hit the same brick wall myself and its driving me crazy. Everything stable and running cool (under 60*C during 100% load) at 1199mhz then won't even post at 1200mhz.

Any ideas at all are welcome...no matter how unlikely you may think they are.
a c 156 V Motherboard
a c 197 K Overclocking
January 17, 2010 12:32:09 PM

shane, I couldn't push my eVGA 680i that high with a Q6600. Was puzzling because my 680i's FSB worked up past 440 MHz with an E6600.

Got a Gigabyte GA-EP35-DS3P and I was 24 hour Prime95 stable at 3.6 GHz with the Q6600.
January 25, 2010 10:59:08 AM

Thanks for the reply.

Wow, 3.6ghz is fantastic. I imagine that must be with some good water cooling.

Anyway, I finally managed (admittedly by pure luck) to get past the 300mhz wall by upping the CPU VVT voltage from 1.2v to 1.55v and it has been pretty stable up to 3.0ghz. With my tiny stock air cooler I can't really go much higher without the temps getting a bit worrying.
!