• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Q9550 worth £160? ([email protected] > Q9550)

Have you set memory settings correctly?
Those will easily keep trashing stability regardless how good CPU is if they're wrong.

lowrider007 said:
As you can see the vid for the chip is 1.1375v and he has it set @ 1.192v to get 3.4ghz, does that seem reasonable?
Click to expand...
Running 1.25VID chip at 3.7GHz and 1.2V in here...
3.4GHz might be tweakable for 1.15V considering how I accidentally forgot Vcore around 1.13V while increasing FSB and how Prime started giving errors faster only near that clock.

P5Q deluxe, NB 1.24, VTT 1.24 and PLL 1.56 (GTLs and FSB strap at auto)
 
I usually run my 9550 at 3.2 with 1.27vcore (1.30 in bios), but to go to 3.4 I need the vcore up at 1.36. only gets to 60ºc at max load in Prime95, and is stable in Prime.
 
My most resent settings,

CPU Ratio = 8.5
FSB Frequency = 376
PCI-E Frequency = 100
FSB Strap to North Bridge = 333
DRAM Frequency =
DRAM Timing Control = 6-5-5-15
DRAM Static Read Control = disabled
DRAM Read Training = disabled
Memory OC Charger = disabled
AI Clock Twister = Stronger
AI Transaction Booster = Manual Perf 12
CPU Voltage = 1.375
CPU GTL Voltage Reference = .63x
CPU PLL Voltage = 1.50
FSB Termination Voltage = 1.30
DRAM voltage = Stock Voltage value
Northbridge Voltage = 1.28
Southbridge voltage = Auto
PCI-E SATA voltage = Auto
Load Line Calibration = Enabled
CPU Spread Spectrum = Auto
PCI-E Spread Spectrum = Auto
CPU Clock Skew = Auto
NB Clock Skew = Auto
CPU Margin Enhancement = Performance Mode
C1E = Disabled
Max Cupid Value Limit = Disabled
Intel Virtualization Mode = Enabled
CPUTM Function = Disabled
Execute Disable Bit = Enabled

Failed prime and this is only @ 3.2ghz and I was hitting 80c, I can;t really push the vcore up any more it's already over recommended.

What's the max I can push my nb volts to?

Perhaps it's the high temps causing the instability?, I've got the chip now @ 3ghz under the vid @ 1.192 and seems to be priming fine atm, it's odd that I can under volt @ 3gig and it's fine but @ 3.2gig @ max volts and it fails but then my temps are now staying under 65c so perhaps this chip just doesn't like high temps?
 
Last edited:
Why do you assume it's the volts that is the problem?

You're at 375fsb on a p5q ...what makes you think you need more than 1.28v? :) 1.2v is overkill for that speed.
 
marscay said:
Why do you assume it's the volts that is the problem?

You're at 375fsb on a p5q ...what makes you think you need more than 1.28v? :) 1.2v is overkill for that speed.
Click to expand...

Because people tell me that is the main reason why workers fail in prime, low vcore, I'm not getting any bsod's or anything, what else can it be?, I've tried lowering the voltage and it still fails so I thought max out the voltage and try and get the system stable first and then lower the voltage in increments but that is not working either, I've tweaked practically every setting on my board, it has to be the high temps that are causing it to fail I can't think of anything else.

I'm going to purchase a premium air cooler next week when I get paid and I'll know then whether it's that or not.
 
Last edited:
What voltage are you using for the ram, when i had my p5q deluxe, the ram sets i used were rated for 2.1 vdimm, but cerain bios revisions for the p5q's were known to seriously overvolt ram, upto +0.08v, so with my ram i set the voltage to 2.02, ram sets were ocz reaper pc6400 and reaper pc8500.
 
setter said:
What voltage are you using for the ram, when i had my p5q deluxe, the ram sets i used were rated for 2.1 vdimm, but cerain bios revisions for the p5q's were known to seriously overvolt ram, upto +0.08v, so with my ram i set the voltage to 2.02, ram sets were ocz reaper pc6400 and reaper pc8500.
Click to expand...


I'm using 1.8v, that's what the ram is rated to use, TWIN2X4096-6400C5, also I've had this ram overclocked @ 920mhz with tighter timings with my last cpu and it was fine.

I'll give it a go though, also I'll try a different bios.
 
Try 2.1 volts for ram and 6-6-6-18 also look on corsair forums for some good settings for this ram.

My ram cant do it's rated timings with certain fsb I have to run 5-5-5-18-42 if I am going above 400mhz.
 
Last edited:
Right, I think my problems could be to do with my memory after all,

With my FSB @ 355, memory @ 1.9v 800mhz 6-6-6-18 : Passes memtest

With my FSB @ 400, Memory @ 1.9v 800mhz 6-6-6-18 : (and tried 7-7-7-24) Fails memtest

Why would the memory fail with a higher fsb even though the chips are running at the same speed ?

(TWIN2X4096-6400C5)

Any suggestions ?
 
Last edited:
dave_beast said:
Unstable memory divider?
Click to expand...

But he is using a 1:1 ratio to get 800mhz from 400fsb so I fail to see that being a reason.

Is a strange one, at least you now its not your chip

Maybe the north-bridge is getting stressed too much / not getting enough juice, its the memory controller right?
 
I was going to say that maybe your mobo is undervolting but that seems unlikely when you're already putting an extra 0.1v through them already & seeing as the P45 ASUS boards have a tendancy to overvolt anyway :confused:
 
Thanks for your continued input on this guys, I'm determined to crack this.

Anyway what does this tell you,

FSB @ 400, CPU ratio @ 6 (2.4ghz) : Memtest fail

Even with the ratio dropped down to 6 from 8.5 but the FSB kept @ 400 memtest still fails.

Yet if I lower my FSB to 355 with the CPU ratio @ 8.5 so the CPU is now running @ 3ghz, 6oomhz faster than the above test then memtest passes fine.

I've tried this a few times and the result is the same, so the increase in CPU speed isn't the problem here, it seems to be solely the increase in FSB.

The higher you raise the FSB the more stress it puts on the memory controller right?, and you normally remedy that by adding volts to the nb yes?
 
Maybe try turning off Load Line Calibration, worth a shot, not sure why heard some problems of it causing instability, it won't hurt anyway.

The thing is you should have to raise the nb voltage at the speed your going for, you could try upping it, but 1.28 is already quite high, maybe even try lowing it.

Also what BIOS you using get the latest version, 45nm can be a pain.
 
I updated my bios to 1406, it's not the latest one but it's is one that is reccomened by everyone for overclocking my CPU, I may try updating to the latest one if all else fails.

I'm also thinking that maybe my PSU could be an issue, I'm gonna try taking out my gtx260 replacing it with an old school 6200 and see if that makes a difference.
 
What psu have you got?

When I was doing mine I hit 3.2 with a lot of the settings on auto apart from vcore which I set to its vid.

Maybe its worth trying again but try hitting 3 with some of the auto settings, its a small overclock should have no problems.
 
Good luck buddy, Im trying to find the lowest voltages I need to run at stock, at 1.0v at the moment and seems fairly stable.

http://valid.canardpc.com/show_oc.php?id=1169110

Your chip can't be that thirsty, feel we are missing a setting some where or a factor, how successful have you found people with your board to be?

The chip is good just keep pushing :)
 
I've got the poorer C1 version of that chip and I'm fairly stable at 3.4Ghz on relatively low volts as well. Good luck in getting your OC sorted out though. :)
 
You must log in or register to reply here.
Back
Top Bottom