Reboot while overclocking?

[known12345]

I have a q6600 and an ABIT IP35-E and was able to oc to 3.6 at 1.5v on prime95 for 10 hours but at high temps between 68-73C. However when I set my fsb to 405 or 410 or 415 (with multiplier 9) after running prime for less then a minute, my comp automatically reboots and i cant seem to figure out why? I set my ram voltage to 2.1 (its max, at 3.6 had timings at 2.0), it timings to 5-5-5-15, my cpu voltage to 1.54 (from 1.5 on 3.6) something, my MCH to 1.37 (from 1.25), my ICH to 1.6 (from 1.6), ive lowered pwm temps from 80C (during 3.6 stress test) to 70C, and ive even lapped my cpu and heatksink (Arctic Freezer 7 Pro).

Ive tried lowering the multiplier to 8 and an fsb of 450 (3.6) and it ran stable on prime95 so I do not think it was a motherboard issue. Ive even been able to get to the windows screen at 430 fsb, 9 multiplier, and 1.56 volts (3.780) so I doubt motherboard is the problem (ive heard of people hitting 4 GHz on the board). When I oc to 3.6 during a moderately hot day and cores reach 77C (testing 3.6, 1.5v) it still doesn't reboot so I do not think it is a temp issue because when I try 405-410 at night, when temperatures are much lower, it still reboots. And ive also tried booting at 420 and 425 just in case there was an fsb gap and it still reboots less then a minute in prime95.

Does anyone know a particular reason why I cannot hit 405-410 or know of something I can try to hit 405-410?

 

[Markfw]

1) What ram do you have ?
2) Is it a G0 stepping Q6600 ?
3) It sure sounds like your chip won;t go over 3.6

 

[known12345]

1) I have a G.SKILL 4GB(2 x 2GB) DDR2 1000:
http://www.newegg.com/Product/...x?Item=N82E16820231145

2) Yup, it is a G0 stepping Q6600.

3) Ahh I think you are right. Was wondering though, if I switch my AF7 Pro for a TRUE 120 or a water cooling set up, is it possible i can clock past 3.6?

 

[krnmastersgt]

It might not be a temp problem, I'm under 55C at 100% load with 3.6, but when I tried for 3.8 I don't even get into windows, it boots, posts(it posts at 4.2 ), runs the windows loading animation and after that just hangs on a blank screen or blue screens, not all chips will go above 3.6, not all will even hit 3.6 so maybe we got stuck with decent chips as opposed to top of the line cherry picked ones ^^ oh well

 

[VirtualLarry]

It could be a problem with the CPU power section. Some people have commented that the IP35-E has a wattage problem, in terms of what it can cleanly supply to the CPU. With a Q6600 at 3.6, you're really taxing it. 405 or 410 FSB may just be pushing it over the edge.

What's wrong with sticking with 3.6Ghz, btw? There's no way you're getting to 4Ghz on that thing.

 

[krnmastersgt]

Originally posted by: VirtualLarry
It could be a problem with the CPU power section. Some people have commented that the IP35-E has a wattage problem, in terms of what it can cleanly supply to the CPU. With a Q6600 at 3.6, you're really taxing it. 405 or 410 FSB may just be pushing it over the edge.

What's wrong with sticking with 3.6Ghz, btw? There's no way you're getting to 4Ghz on that thing.

People have been able to get these babies stable at 4.2 so no reason we can't hit 4 xP but in all seriousness my 3.6 is fine, I just want to know how far I can push it, I'll probably drop it down to 3.6 no matter how high I go, but I am rather curious.

 

[Vixx]

Seems like your voltages are really high.

My Q6600 G0 will do 400*9 or 445*8 @ 1.42v, however if I try to push it to 450*8 I need to up the voltage to 1.5v or more just to get it stable. The additional speed isn't worth the additional heat and wear-n-tear on my Q6600.

 

[known12345]

Oh no, I have no problem with sticking with 3.6. In fact, I usually run my comp at only 3.0 to avoid high temps. I would just like to know why I am unable to get past 400. I mean out of all the numbers that could have been my barrier, such as 385, 405, 435 and numbers not multiple of 5, 400 was my barrier. I mean if its the cpu I would understand but I would to know if there is something ive just havent tried?

Yeah, the voltage required to push my card to 3.6 is 1.5. Actually 1.48 runs stable as well. Anything lower... well I can get to my windows screen at I think 1.4 but if I run prime it reboots automatically? If i raise to 1.42, well a core has an error when running prime95. I mean it depends on what you define "stable." If you mean getting into windows and doing a few programs, I think I can do 1.46 or 1.44 at 3.6 but if you mean running prime95 for 10+ hours with no errors, then I need 1.48.

I am sorry if I go off topic but the talk about cpu voltage has got me thinking, ive heard of water cooling (and other extreme forms of cooling) systems hitting past 4.0 but how are they able to go that far? I know one of the main constraints of overclocking is high temperature but another is how lucky a card you get. I mean do individuals with over 4.0 just have an exceptional card? I mean in order for me to run at 3.6, I need 1.5 volts. If I have a water cooled system, does it necessarily decrease the voltage required to run 3.6 or does it just allow me to run at 1.5 with much lower temperatures? I suspect the later but because it decreases temperatures, does it then make it safe to run at higher voltages or is it still dangerous to run past 1.4/1.5?

 

[spinejam]

looks like you're reaching the max o/c on the cpu / mb combo!

 

[krnmastersgt]

Heat isn't an issue with my TEC, and wear and tear? This baby will last at least 4 years at least at 3.6, I wanna see if I can get 3.9+ though . Can't be a temp problem, probably the mobo, I really hate how the 780i boards are 1333 fsb native, 1600 is already a little pushing, and when I go as high as 1800 it just doesn't seem to like me, although others have been able to get it, even on 780is, further experimentation shall be conducted later and for some strange reason I can't get 3.8 to get into windows but 3.9 can....I get a BSoD rather shortly, but I doubt its a voltage since I upped it to 1.65 for that little experiment.