Gigabyte GA-P35-DS3L Thread

[BrotherTheodore]

Hey everyone, not sure if this is the correct place to post this question, as I'm not sure if it's a motherboard problem, but here it is anyway:

First of all I seem to have an FSB hole from around 375-410FSB, and anything after about 425FSB will not boot even on 8x multi. I typically run at 3ghz (333x9), no problems at all. Over the past few days I've wanting to see if I could get a better OC, so the only real option for me is to go up slightly to 3.2/3.3 which I'm not too interested in, so I've been trying at 3.7, and 3.8.

I set the voltage in the bios for 3.7/3.8 to anywhere from 1.48 to 1.52. The temps seem to be ok, using Everest to read my temps, the cores are mid 30s to low 40s and cpu is low 30s. At stress, the cores go up to upper 50's.

The problem I am having, is that after about 30 seconds of stress testing (Prime), something kicks in and drops the multi down to 6. I have C1E, EIST, and TM2 all disabled in the bios. If stop stressing the chip then about a minute or two later it will return to 9x Multi.

An interesting thing to note: If I have TM2 enabled, then with CPUZ it will show the multiplier dropping. If TM2 is DISABLED, then it will NOT show the reduced multiplier and clock speed, however I have tested my computer with SuperPi, and can confirm that it does indeed still drop the multi, it simply doesn't show up in CPUZ. When I first boot up with 3.7/3.8 and run SuperPi I get about 13sec or 14sec 1M test. If I then proceed to prime for a minute, turn off prime go back to SuperPi, it will then be anywhere from 19sec to 22sec 1M test indicating that my CPU is indeed being limited.

Any help would be greatly appreciated.

[other bios settings in case people may ask: 5-5-5-15 timings, G.Skill 800 Ram at 2.0V, PCI Freq 100, all other voltages I have left default, although I have tested them at .1 up to .3 seemingly with the same effect I am having now.

[technogeek]

Originally Posted by MZ4Adam

Hey guys, quick question.. can someone tell me what manual voltage I should run with this board and an e8400 3.0Ghz chip to run at both 400 and 450mhz on the stock 9x multiplier? With the board set on auto the voltage looks way too high and is raising idle temps above 40-44c which isn't too good as far as I can tell.

Any information is GREATLY appreciated!!!

I have my voltage set in BIOS @ 1.3 and CPU-Z showes it @ 1.26. Both my core temps are @ 40c-45c at idle. My FSB is at 445, the max voltage is 1.3625 and the max temp is 72c according to Intel.

Edit.. Intel spec sheet-->http://processorfinder.intel.com/det...px?sspec=slapl

[MosIncredible]

Originally Posted by BrotherTheodore

Hey everyone, not sure if this is the correct place to post this question, as I'm not sure if it's a motherboard problem, but here it is anyway:

First of all I seem to have an FSB hole from around 375-410FSB, and anything after about 425FSB will not boot even on 8x multi. I typically run at 3ghz (333x9), no problems at all. Over the past few days I've wanting to see if I could get a better OC, so the only real option for me is to go up slightly to 3.2/3.3 which I'm not too interested in, so I've been trying at 3.7, and 3.8.

I set the voltage in the bios for 3.7/3.8 to anywhere from 1.48 to 1.52. The temps seem to be ok, using Everest to read my temps, the cores are mid 30s to low 40s and cpu is low 30s. At stress, the cores go up to upper 50's.

The problem I am having, is that after about 30 seconds of stress testing (Prime), something kicks in and drops the multi down to 6. I have C1E, EIST, and TM2 all disabled in the bios. If stop stressing the chip then about a minute or two later it will return to 9x Multi.

An interesting thing to note: If I have TM2 enabled, then with CPUZ it will show the multiplier dropping. If TM2 is DISABLED, then it will NOT show the reduced multiplier and clock speed, however I have tested my computer with SuperPi, and can confirm that it does indeed still drop the multi, it simply doesn't show up in CPUZ. When I first boot up with 3.7/3.8 and run SuperPi I get about 13sec or 14sec 1M test. If I then proceed to prime for a minute, turn off prime go back to SuperPi, it will then be anywhere from 19sec to 22sec 1M test indicating that my CPU is indeed being limited.

Any help would be greatly appreciated.

[other bios settings in case people may ask: 5-5-5-15 timings, G.Skill 800 Ram at 2.0V, PCI Freq 100, all other voltages I have left default, although I have tested them at .1 up to .3 seemingly with the same effect I am having now.

3.7/3.8 is kind of high to shoot for on a Q6600 so quickly. I'm pretty sure you're going to need some extra FSB and MCH voltage to run a FSB of 400+ on a quad.

[BrotherTheodore]

Originally Posted by MosIncredible

3.7/3.8 is kind of high to shoot for on a Q6600 so quickly. I'm pretty sure you're going to need some extra FSB and MCH voltage to run a FSB of 400+ on a quad.

I have tried the FSB, MCH, and PCI voltage all from .1 to .3 and it doesn't change my problem, my processor still drops after about a minute, it seems as if TM2 isn't disabling or something..

[MosIncredible]

Originally Posted by BrotherTheodore

I have tried the FSB, MCH, and PCI voltage all from .1 to .3 and it doesn't change my problem, my processor still drops after about a minute, it seems as if TM2 isn't disabling or something..

Ah, I read your last statement wrong. You tried clearing the CMOS?

[BrotherTheodore]

Originally Posted by MosIncredible

3.7/3.8 is kind of high to shoot for on a Q6600 so quickly. I'm pretty sure you're going to need some extra FSB and MCH voltage to run a FSB of 400+ on a quad.

Originally Posted by MosIncredible

Ah, I read your last statement wrong. You tried clearing the CMOS?

No haven't ever cleared the CMOS since I got this board, does clearing CMOS help with a problem of this type or FSB hole type problems? Also, I have been updating my bios whenever a new one comes out, hoping I can OC better, but to no avail...I have F8b right now I think.

[MosIncredible]

Originally Posted by BrotherTheodore

No haven't ever cleared the CMOS since I got this board, does clearing CMOS help with a problem of this type or FSB hole type problems? Also, I have been updating my bios whenever a new one comes out, hoping I can OC better, but to no avail...I have F8b right now I think.

I was trying to find out the reasoning behind the multi dropping after a minute of priming. Don't expect it to change a FSB hole problem.

[Rpg2]

Originally Posted by BrotherTheodore

No haven't ever cleared the CMOS since I got this board, does clearing CMOS help with a problem of this type or FSB hole type problems? Also, I have been updating my bios whenever a new one comes out, hoping I can OC better, but to no avail...I have F8b right now I think.

I'm having the same multiplier problem as you. There is F8f BIOS out now on the Gigabyte site. PM me if you find a solution to the problem.

[BrotherTheodore]

Originally Posted by Rpg2

I'm having the same multiplier problem as you. There is F8f BIOS out now on the Gigabyte site. PM me if you find a solution to the problem.

Hey, I actually mistyped/misremembered my bios revision, it is in fact the F version. I actually googled my symptoms yesterday and found some of your posts about it. Thats kind of what prompted me to come back to overclock.net

[Indignity]

Originally Posted by SgtSpike

Well, I'd certainly love to do that, but I don't have a Q9450, and I haven't seen anyone here with the DS3L that does have a Q9450... If you've got one, by all means share your results!

I have a QX9650 on it right now & it's stable as hell at 3.8GHz... Since I haven't played with a Gigabyte board since my A64-2500+, i'm trying to get used to it.

I did have it clocked as high as 4.3GHz, but froze up after about 2 minutes of stress. I know I need more volts, but I really don't wanna fry any of this equipment.

I did update the BIOS to F8f & this is a revision 2 board I got from Newegg on the $50 Open Box deal.