why cant i hit 3 ghz?

[Fayd]

e4400... MSI P35 Neo2-FR

i can get up to 2.8 ghz no problem, just upping fsb.

temps look okay at 2.8. idles at 38-42, loads at high 50's.

but i cant get beyond that. i'm not upping the memory with it (well, i did up to 2.8, but when i went to 2.9, i dropped the memory back to like 725 mhz)

upping voltage on cpu has no effect.

the comp posts okay, gets to loading windows, then either reboots or BSOD's. safe mode has no effect.

any ideas?

 

[yh125d]

Raise NB voltage and CPU both

 

[error8]

Originally posted by: yh125d
Raise NB voltage and CPU both

This

 

[AzN]

Originally posted by: yh125d
Raise NB voltage and CPU both

It's a e4400 with a fsb of 200. No need to raise voltage. I've gone high as 480fsb and increase to NB wasn't needed until 500fsb.

To OP you might just have hit your limitations or not.

If it doesn't post you are hitting the barrier on this particular CPU or memory. Make sure your memory is set to 1/1 divider with spec timings. Memory is the biggest limitation when it comes to overclocking if you don't set it up correctly. I can go 3.5ghz or more on my e6300 revision b2. I'm limited by memory.

 

[Fayd]

Originally posted by: Azn

Originally posted by: yh125d
Raise NB voltage and CPU both

It's a e4400 with a fsb of 200. No need to raise voltage. I've gone high as 480fsb and increase to NB wasn't needed until 500fsb.

To OP you might just have hit your limitations or not.

If it doesn't post you are hitting the barrier on this particular CPU or memory. Make sure your memory is set to 1/1 divider with spec timings. Memory is the biggest limitation when it comes to overclocking if you don't set it up correctly. I can go 3.5ghz or more on my e6300 revision b2. I'm limited by memory.

the memory is corsair xms2, 800mhz. i've taken it to 840 at spec timings, 1.90v (but i bumped to 1.95v anyways) this is 1.5:1

when i went to 290 fsb, i dropped the mem multi back to 1.25:1, so 725 mhz.

i'm fairly certain it's not memory holding me back.

 

[error8]

Then the only thing holding you back it's the vcore.

 

[skillyho]

Originally posted by: error8
Then the only thing holding you back it's the vcore.

It's probably not vCore....they're good for more than that on Stock voltage, unless his board has really bad vDroop.

OP...you probably just have a bum chip, or a FSB wall on your mobo. Check a different CPU (if you can) to see if the higher FSB you're attempting works on other CPU's, or just drop the multiplier of the chip and raise the FSB until you hit an unstable point. You could always try that the other way around if you have another 775 board, too.

 

[demiurge3141]

This makes no sense at all. It probably is vcore, if OP has a bum chip then a raise in vcore is all the more neccessary.

Originally posted by: skillyho

Originally posted by: error8
Then the only thing holding you back it's the vcore.

It's probably not vCore....they're good for more than that on Stock voltage, unless his board has really bad vDroop.

OP...you probably just have a bum chip, or a FSB wall on your mobo. Check a different CPU (if you can) to see if the higher FSB you're attempting works on other CPU's, or just drop the multiplier of the chip and raise the FSB until you hit an unstable point. You could always try that the other way around if you have another 775 board, too.

 

[Fayd]

vcore is 1.35 normally. i run with eist enabled, but even when i disabled it and pushed the vcore out to 1.45, it didnt make any difference.

also tried upping the nb voltage until it showed red in bios (max reasonably allowable, i guess. fwiw, red vdimm = 2.3, red vcore = 1.5, etc) still no dice.

only thing i havent done now is attempting in between 2.8 and 2.9. but it wont go 2.9, and wont go 3.0.

do the straps make any difference? like should i be setting it to the 266 or 333 strap?

 

[skillyho]

Originally posted by: demiurge3141
This makes no sense at all. It probably is vcore, if OP has a bum chip then a raise in vcore is all the more neccessary.

Originally posted by: skillyho

Originally posted by: error8
Then the only thing holding you back it's the vcore.

It's probably not vCore....they're good for more than that on Stock voltage, unless his board has really bad vDroop.

OP...you probably just have a bum chip, or a FSB wall on your mobo. Check a different CPU (if you can) to see if the higher FSB you're attempting works on other CPU's, or just drop the multiplier of the chip and raise the FSB until you hit an unstable point. You could always try that the other way around if you have another 775 board, too.

What doesn't make sense? If he has a bum chip with a FSB wall, then the chip will not POST after a certain figure. You can raise the vCore to 1.5V but some CPU's simply will not do a higher FSB (which is why chips are binned...no overclock is guaranteed). It's the same thing with a FSB wall on certain motherboards, even though the CPU is capable the mobo just can't handle the high clocks...which could also be the case.

I had an e4300 and GA-965P-S3 that couldn't do 2.7GHz stable, even with 1.5V....I swapped out CPU's with a different e4300 on the same Mobo/RAM/etc. and I could do 3.4 with 1.44V.

OP....did you try lowering the multiplier of the CPU down from 10 to something like 8? If you do this and then try to achieve the same FSB for 3GHz (300FSB) you can see where your weak spot is.

 

[Fayd]

Originally posted by: skillyho

Originally posted by: demiurge3141
This makes no sense at all. It probably is vcore, if OP has a bum chip then a raise in vcore is all the more neccessary.

Originally posted by: skillyho

Originally posted by: error8
Then the only thing holding you back it's the vcore.

It's probably not vCore....they're good for more than that on Stock voltage, unless his board has really bad vDroop.

OP...you probably just have a bum chip, or a FSB wall on your mobo. Check a different CPU (if you can) to see if the higher FSB you're attempting works on other CPU's, or just drop the multiplier of the chip and raise the FSB until you hit an unstable point. You could always try that the other way around if you have another 775 board, too.

What doesn't make sense? If he has a bum chip with a FSB wall, then the chip will not POST after a certain figure. You can raise the vCore to 1.5V but some CPU's simply will not do a higher FSB (which is why chips are binned...no overclock is guaranteed). It's the same thing with a FSB wall on certain motherboards, even though the CPU is capable the mobo just can't handle the high clocks...which could also be the case.

I had an e4300 and GA-965P-S3 that couldn't do 2.7GHz stable, even with 1.5V....I swapped out CPU's with a different e4300 on the same Mobo/RAM/etc. and I could do 3.4 with 1.44V.

OP....did you try lowering the multiplier of the CPU down from 10 to something like 8? If you do this and then try to achieve the same FSB for 3GHz (300FSB) you can see where your weak spot is.

yeah... it wouldnt even post when i lowered the multi.


edit:

when i open the case next (gonna do to pull the vid card and apply my AC accel s1r2) i'm gonna see if i can change the strap to the 266 one. if what i'm reading online is correct, it'll give me some looser timings within the NB that might allow me to clock a little higher on the fsb.

 

[alcoholbob]

What about dramatically decreasing RAM speed? Just as a test.

 

[cubeless]

i feel your pain... i have a e4500 that needs 1.45 to get virtually any oc, and just stops at 3ghz... there are some 'bad' chips out there...

 

[kmmatney]

My E4400 stops at 3.0 Ghz - won't go any higher. I have it back down to 2.9 Ghz, but sometimes overclock it in windows to 3.0 Ghz for encoding. It won't go any higher though, and voltage doesn't help.

 

[bailw]

what are you using to monitor temperature. Also, what cooling method?