[mikeynavy1976]

Ok...now I'm confused and frustrated. I can't get it to overclock over 3.2GHz (I've tried repeatedly to get it to 3.4GHz and it won't boot). It won't post at all and it won't accept any manual voltages (except DRAM). I encountered this with a previous Gigabyte P35 motherboard trying to overclock an E8400 past 3.4GHz and the same thing. Are these Gigabyte boards that bad at overclocking and adjusting settings? The CPU voltage doesn't change no matter what settings I put. Right now I'm running the Q9450 on an EP45-DS3R and the following BIOS settings:

CPU Vcore - 1.2875V (manually picked)
CPU Termination - 1.30V (I don't know what this means but I tried to pick a higher value hoping my voltage would show an increase).
CPU PLL - 1.5 V (Auto) - I don't know what this means and I haven't changed anything.
CPU Reference - 0.76V (Auto) - I don't know what this is

MCH Core - 1.1V (Auto)
MCH Reference - 0.76V (Auto)
MCH/DRAM - 0.9V (Auto)
ICH I/O - 1.5V (Auto

DRAM - 1.8V (manually changed to 2.16V but shows lower in BIOS and Windows)
DRAM Termination - 0.9V (Auto)
Channel A Reference - 0.9V (Auto)
Channel B Reference - 0.9V (Auto)

When I open up the PC Health Tab in the BIOS to view the actual voltages it shows CPU VCore as 1.252V and DRAM as 2.128V (there is approximately 0.03V difference from BIOS settings). When I log into Windows and view with Everest, HW Monitor, and CPU-Z the CPU voltage is at 1.23V and Memory is at 2.13V. The VID on my processor is supposed to be 1.25V. When I change the CPU Voltage to Auto in the BIOS it jumps to 1.3V and my CPU temperatures are 3C cooler. Can someone please explain what the previous voltage settings mean and why everything seems so messed up, settings won't hold, and why I can't get 3.4GHz??

[Ric0h]

Not all chips will overclock well. Just be happy you got it to 3ghz

[mikeynavy1976]

Ok...I did some research on the web and found someone with the same setup who posted what he did. I was able to get my computer to boot at 3.4GHz and went up to 1.3V (overall). Unfortunately, Prime 95 had an error relatively quickly. Temps looked ok, though...highest was 60C...with ambient room temp being 27C. For those that have overclocked the Q9450 to 3.4GHz and got it stable. What voltage did you have?

[Nagasama]

here are my settings on a p5q pro FOR 3.5ghz:

FSB frequency : 435
fsb to nb strap : 333
DRAM @ 1045 5-5-5-15
performance level : 10
Vcore : 1.225
GTL reference : auto
PLL voltage : 1.6
FSB termination : 1.4
DRAM : 2.1
NB : 1.3
cpu clock skew : 400ps
NB clock skew : 100ps

something is wrong if the voltage isnt changing when you input it into the bios. arent gigabyte boards the ones where you hit "F1" to get into the good settings?
the difference in bios vs. program readings is vdroop. you will notice it more when the processor is being stressed. my mobo has a setting called "loadline calibration" that keeps the voltage regular from idle to load. i dont know if yours has a similar feature.
"auto" setting will overvolt sometimes. thats why its higher. as for why its cooler, i have no idea.

[thirdshiftdj]

be careful w/ the PLL voltage. I wouldn't go above 1.6v for 24/7

[Nagasama]

yepper thats the next one on the list to drop.
i just started P95 with a NB of 1.25 and a performance level of 8.
if that passes tonight, i'll drop the PLL to 1.5 so on and so forth.
Vcore is as low as it will go stable i think.

[thirdshiftdj]

thats pretty freaking low Naga. How are you liking it so far?

[Nagasama]

it passed flying colors 10 hours p95. wont boot with a performance level of 7.
ima set the PLL down now and let 'er rip while im at work. if that passes, i'll lower the NB another notch.
its fast, dude. fast. i was thinking of getting watercooling, but i just cant do it with temps like this chip has. kinda irritates me that the chip could go so much higher (IMO) on air, but the FSB wall is made of bricks and mortar, and it AINT doin no more.
i'll keep jackin with it, i may be able to tweak it up some, but i think i've milked it to all it can do.
i may sell my pro and get an x48 mobo and see if it will hit 3.8 or so. lol thats a lot of cash for 300mhz. and i dont think (pretty positively) that it is the mobo stopping, its the chip. tried 6 x 450 as well and nothing. lol 8 x 348, 7 x 348, 6 x 348, 8 x 460, 7 x 465 on and on and on it will boot, but hang when any kind of load is applied. or i suppose the FSB may be at its absolute limit for this mobo. right now it runs around 1750mhz, my mobo "unofficially" supports 1600mhz FSB. so, i guess that could be the reason...maybe...lol, a $300 gamble i would be taking....
i'll need to see some p5q delux or rampage formulas or lanparty boards hitting those speeds with this chip before i am willing to concede that it is not the chip. mainly, because i could run right up to 500mhz with my e8400

[thirdshiftdj]

From what I've seen the P45 chipsets seem to do better w/ 45nm quads. I've said it once and I'll say it again 450-475 fsb on a quad is no easy feat and if it is doable your going to be running some voltage not suitable for 24/7


edit: I shouldn't say all P45's do better

[Drizzt]

Originally Posted by nagasama

it passed flying colors 10 hours p95. wont boot with a performance level of 7.
ima set the PLL down now and let 'er rip while im at work. if that passes, i'll lower the NB another notch.
its fast, dude. fast. i was thinking of getting watercooling, but i just cant do it with temps like this chip has. kinda irritates me that the chip could go so much higher (IMO) on air, but the FSB wall is made of bricks and mortar, and it AINT doin no more.
i'll keep jackin with it, i may be able to tweak it up some, but i think i've milked it to all it can do.
i may sell my pro and get an x48 mobo and see if it will hit 3.8 or so. lol thats a lot of cash for 300mhz. and i dont think (pretty positively) that it is the mobo stopping, its the chip. tried 6 x 450 as well and nothing. lol 8 x 348, 7 x 348, 6 x 348, 8 x 460, 7 x 465 on and on and on it will boot, but hang when any kind of load is applied. or i suppose the FSB may be at its absolute limit for this mobo. right now it runs around 1750mhz, my mobo "unofficially" supports 1600mhz FSB. so, i guess that could be the reason...maybe...lol, a $300 gamble i would be taking....
i'll need to see some p5q delux or rampage formulas or lanparty boards hitting those speeds with this chip before i am willing to concede that it is not the chip. mainly, because i could run right up to 500mhz with my e8400

Wern't you the one who said water cooling + 45nm = meh?