• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

need help overclocking qx9770 please!!!!

swale

Limp Gawd
Joined
Feb 3, 2008
Messages
274
Hi have just recently brought a qx9770 have manged to get a stable 4.05ghz@1.5v (1.44 under load i think see bios) out of it but thats about all i can get heard many are getting 4,275mhz and even 4.4ghz took some photos of my poor attempt at 4,275, obviosley i dont wanna go to mad with the voltages as i don't wanna fry anything ;-)


Please see sig for specs below.

here are some pics of my 4.05 ghz overclock posting this to see if anyone can see any flaws at all???
SNV30867.jpg

SNV30868.jpg

SNV30869.jpg


heres what i did for an attempt at 4.275gz wih resulted in crash a little after booting up vista...
SNV30862.jpg

SNV30863.jpg

SNV30864.jpg

SNV30865.jpg

SNV30866.jpg


Any help/advice greatly appreciated, Thanks!!!!
 
Use your unlocked multi. You might be a able to use less vcore that way as opposed to using 1900 fsb.
 
All chips are not created equal. It's very possible your CPU simply can't go any further. I highly suggest that you try tweaking your setup to make it stable with less voltage though, as 1.5v is extremely high for a 45nm Core 2. Perhaps using a higher multiplier like the poster above me suggested could help you accomplish this, and it might even allow you to overclock further since you'll have more headroom with the FSB.
 
try a better cooler and mess around with the multi's. But like he said no two chips are the same. So you might have to stay where you are. My chip only got me to 3.2.
 
Am already watercooled see below, surely theres not much more i can do is there, so you guys recommend may lowering fsb and upping multi, so should i unlink the ram from link and sync with fsb???
 
Unlink your ram. First find your OC "goal" then once you have that stable you can begin messing with the ram. Is also usually best to run unlinked as reports show it to be more stable.
 
Ok will give the above a try what vcore would you recommend and what fsb??? as i said above i'm currently @1.5v (setting in bios) what should i try??

just to clarify when were talking about vcore volts are we supposed to give bios setting voltage or the under load voltage read from elsewhere, cause i thought that different boards are a little different whith there vcores under load, can anyone clarify this for me please???

thanks for all the help so far guys :)
 
Am already watercooled see below, surely theres not much more i can do is there, so you guys recommend may lowering fsb and upping multi, so should i unlink the ram from link and sync with fsb???

It would be a good idea to try unlinking your RAM and keeping it at a low speed, although it's possible that it might prevent you from reaching higher FSBs(I've heard that running unlinked on nVidia chipsets can sometimes cause instability at higher bus speeds). It's still worth a shot though. Also, you should definitely try to raise your multiplier and lower the FSB; that puts less stress on the chipset and helps to make the CPU the only factor limiting your overclock instead of having to worry about the chipset and RAM as well.

Edit:
Ok will give the above a try what vcore would you recommend and what fsb??? as i said above i'm currently @1.5v (setting in bios) what should i try??

just to clarify when were talking about vcore volts are we supposed to give bios setting voltage or the under load voltage read from elsewhere, cause i thought that different boards are a little different whith there vcores under load, can anyone clarify this for me please???

thanks for all the help so far guys :)

Try setting it to 12x333 to start, and see if it can handle the high multiplier at that speed. If it works out, gradually decrease the voltage until it isn't stable any more. Once you find the minimum voltage, start raising the FSB and voltage to get it higher, and play around with 10x and 11x multipliers if necessary to see if that'll get you any further. I highly recommend that you don't raise the voltage past 1.5v, as that's already dangerously high for a 45nm quad(Intel's maximum recommended voltage for Penryns is around 1.38v IIRC).
 
in an ideal world roughly what sort of overclcked voltage is safe on my quad core for 247 use, i just seem unable to get it pas 4.05ghz but on my striker extreme i got it to 4.2ghz without much agg at all, any ideas, am i giving fsb and southbridge etc enough volts?
 
sorry and when we talk about vcore do we measure it from something like cpuz or the bios setting itself????

i just can get it stable above 4.05ghz even with stock mboard frequinces and just upping multi i get bsod with 1.5vcore (bios setting not a reading from cpuz)
 
ok the best i can seem to do is 4.05 with it @ 1.4750 vcore (which am still testing now but hasn't crashed as yet... my core temp is around 55 idle and 72 max ambient is around 22c i guess running prime95 quad test small fft's

p.s core teemp reads my vid @ 1.3625v (this is my core volts?)
cpuz sees it at 1.368v this reading is take whilst running stress test...
 
ok the best i can seem to do is 4.05 with it @ 1.4750 vcore (which am still testing now but hasn't crashed as yet... my core temp is around 55 idle and 72 max ambient is around 22c i guess running prime95 quad test small fft's

p.s core teemp reads my vid @ 1.3625v (this is my core volts?)
cpuz sees it at 1.368v this reading is take whilst running stress test...

Sounds like 4.05GHz is your chip limit if you RAM stock FSB and upped the multiplier to that point and it failed at well. Grats on breaking 4GHz with a Quad.
 
Sounds like 4.05GHz is your chip limit if you RAM stock FSB and upped the multiplier to that point and it failed at well. Grats on breaking 4GHz with a Quad.

LOL Thanks :) Yes i think i agree anyone have any idea on my vcore volts question?
 
There is no accurate way to measure VCore from within Windows. Generally when people talk about raising voltage to a certain level, they're referring to the level stated in the BIOS.

It's too bad that you weren't able to get any higher, but 4.05GHz is still a very respectable overclock, especially for a quad :).
 
There is no accurate way to measure VCore from within Windows. Generally when people talk about raising voltage to a certain level, they're referring to the level stated in the BIOS.

It's too bad that you weren't able to get any higher, but 4.05GHz is still a very respectable overclock, especially for a quad :).

CPU-z measures vCore.... If your temps are fine then I suggest raising volts more. Also lower FSB with a 10-12 multi is the best idea, 790i is not known for high FSB.

edit: just read your post. Vid is what volts the CPU is supposed to have to operate at stock speeds. The number CPU-z is showing is your actual vCore, taking into account vDroop/etc. It sounds like you have a massive amount of vDroop.
 
CPU-z measures vCore.... If your temps are fine then I suggest raising volts more. Also lower FSB with a 10-12 multi is the best idea, 790i is not known for high FSB.

edit: just read your post. Vid is what volts the CPU is supposed to have to operate at stock speeds. The number CPU-z is showing is your actual vCore, taking into account vDroop/etc. It sounds like you have a massive amount of vDroop.

CPU-Z measures VCore through the motherboard, which is neither designed for nor capable of providing an accurate readout. This has been proven many times over, and the fact is that short of using a voltmeter on the motherboard, there is no accurate way to determine the actual VCore. This is not a matter for debate, it's simply a fact.

Also, raising the voltage would be a bad idea even if his temps are good, simply because he's already basically at the maximum safe voltage for a 45nm Intel CPU, and higher voltages are likely to cause damage to the processor.
 
CPU-Z measures VCore through the motherboard, which is neither designed for nor capable of providing an accurate readout. This has been proven many times over, and the fact is that short of using a voltmeter on the motherboard, there is no accurate way to determine the actual VCore. This is not a matter for debate, it's simply a fact.

Also, raising the voltage would be a bad idea even if his temps are good, simply because he's already basically at the maximum safe voltage for a 45nm Intel CPU, and higher voltages are likely to cause damage to the processor.

Ok it might not be 100% accurate, but it's pretty damn close. Of course you can have your own opinion about that but I for one trust CPU-z. The OP is only at 1.36 volts under load, that is nowhere near the max safe voltage.

Edit: just to clarify, of couse CPU-z isn't the greatest when it comes to voltage, but it's the best we've got (other than a multimeter). I for one would up the volts when CPU-z is only reading 1.36 under load ;)
 
I for one would up the volts when CPU-z is only reading 1.36 under load ;)

For the umpteenth time, the maximum safe voltage for a Penryn is around 1.4v. Assuming 100% accuracy on the part of CPU-Z(which is obviously nowhere near the reality), 1.37v is already pushing it, so anything other than very minor voltage changes would be a bad idea. This is a $1000 chip we're talking about here, so I really wouldn't risk causing damage to the CPU just to potentially increase the clock speed by a small amount. Obviously the OP can do what he wants, but you're giving him advice that could damage his CPU, which is why I'm trying to set the record straight.
 
I've seen 1.45 quoted as max safe based on intel's white sheet, it is possible my number is wrong ;)

I'd be happy with 4.05 :D
 
For the umpteenth time, the maximum safe voltage for a Penryn is around 1.4v. Assuming 100% accuracy on the part of CPU-Z(which is obviously nowhere near the reality), 1.37v is already pushing it, so anything other than very minor voltage changes would be a bad idea. This is a $1000 chip we're talking about here, so I really wouldn't risk causing damage to the CPU just to potentially increase the clock speed by a small amount. Obviously the OP can do what he wants, but you're giving him advice that could damage his CPU, which is why I'm trying to set the record straight.

I think your about right there its an expensive chip and i don't wanna risk burn it out for the sake of another 100mhz or something, think i could go further but its at the point were it would take alot of voltage for little mhz i reckon, this may be a umb statement, but isn't a qx9770 a Yorkfield chip no a Penryn???
 
I've seen 1.45 quoted as max safe based on intel's white sheet, it is possible my number is wrong ;)

I'd be happy with 4.05 :D

Is that bios setting or voltage under load (ex cpu-z) do you have the link by chance???
 
Is that bios setting or voltage under load (ex cpu-z) do you have the link by chance???

Usually they mean BIOS voltage, since that is what could be getting applied at any point in time. I think you are really over thinking this though. You are already at or just past what more people would consider the safe zone for Penryns. No matter what any white paper or design doc is going to say, the fact is you are already pushing your CPU, and if you want to go for more, that is entirely up to you how far you go.

At some point, it really does not matter what any of us say, or what anyone else on the net says. You need to determine if the potential reward outweighs the risk to you at this point. If you are unhappy with your setup the way it is, then by all means keep pushing and see how far you get, then you can come tell us how you got your QX9770 @ 4.2GHz with 1.6vcore or whatever the result is.
 
this may be a umb statement, but isn't a qx9770 a Yorkfield chip no a Penryn???

Penryn is kind of the blanket codename for the entire generation of Intel 45nm CPUs. People use it because it's a bit of a pain to remember the individual codenames for every single variation of the core(dual, quad, different amounts of cache, etc.).
 
Usually they mean BIOS voltage, since that is what could be getting applied at any point in time. I think you are really over thinking this though. You are already at or just past what more people would consider the safe zone for Penryns. No matter what any white paper or design doc is going to say, the fact is you are already pushing your CPU, and if you want to go for more, that is entirely up to you how far you go.

At some point, it really does not matter what any of us say, or what anyone else on the net says. You need to determine if the potential reward outweighs the risk to you at this point. If you are unhappy with your setup the way it is, then by all means keep pushing and see how far you get, then you can come tell us how you got your QX9770 @ 4.2GHz with 1.6vcore or whatever the result is.

Totaly agree Im happy with 4.05ghz, don't wanna over do for a couple of mhz extra as i said before, but do you consider 1.4750v(bios) 1.3625v (coretemp) to high and unsafe for my core?
 
Back
Top