Q6600 overclocking

[dpk33]

1.65 is way too high in bios. I'm at 1.42 and even that makes me a little nervous, though I really push mine hard with dc 24/7. If you want this thing to last you for a couple of years then keep bios voltage at 1.5 or lower. There is nothing inherently incapable of getting you to 3.6 on an ip 35e, though you need some luck bc the mono doesn't handle super high temps/voltages as well as, say, an ip 35 pro or dq 6 does.

Didn't I give you my settings as well in your other thread?

Even 9x333 doesnt work, let alone 9x356.

 

[lavaheadache]

Even 9x333 doesnt work, let alone 9x356.

get another motherboard.


I just sold my EP45-UD3p last week for $70 bucks. It was a monster motherboard for ocing core 2 quads

 

[dpk33]

get another motherboard.


I just sold my EP45-UD3p last week for $70 bucks. It was a monster motherboard for ocing core 2 quads

It does 400fsb. What makes it so that it can't go higher than 7x multi?

 

[lavaheadache]

dunno... I'm not convinced it is incapable. I just know a better board would be easier.

I'm sure you can go higher than 7x400 but you must be doing something/or not doing something that is keeping you from getting higher.

Vdroop is a factor but I doubt it is THE factor. What is your chip default VID ?

 

[dpk33]

dunno... I'm not convinced it is incapable. I just know a better board would be easier.

I'm sure you can go higher than 7x400 but you must be doing something/or not doing something that is keeping you from getting higher.

Vdroop is a factor but I doubt it is THE factor. What is your chip default VID ?

Forgive me for the noob question, but what's VID and where do I find it?

If it's the stock volts, then it's about 1.2375v

EDIT: I believe it's stable at 8x400 and 1.4 volts in bios and 1.3 volts reported in cpu-z. Maybe it was because i raised volts too high?

 

[lavaheadache]

vid is the default voltage of the chip. Every single chip is different. The program Coretemp will tell you what you VID is.

Too high of volts could definitely be a culprit. Overclocking can be very time consuming. If you are new to it you can run into a lot of troubles and not know what to look for or tweak.

 

[dpk33]

vid is the default voltage of the chip. Every single chip is different. The program Coretemp will tell you what you VID is.

Too high of volts could definitely be a culprit. Overclocking can be very time consuming. If you are new to it you can run into a lot of troubles and not know what to look for or tweak.

Yup, Coretemp states my VID is 1.2375v. So what can I do to get to 9x multi on my rig?

 

[lavaheadache]

9@400 fsb is unlikely, but 8 is doable and more than enough. To hit the 9 multi you need a new motherboard and a little luck with your processor.

Your VID is decent. My old one was a 1.225 and was capable of 9@400 at like 1.45 volts.

what are the voltage options of your board.... verbatim

 

[rbk123]

9*400 is fine with that motherboard, however your CPU may not be capable of hitting it. Only the G0 steppings had much success hitting 3.6G.

However you are going about this backwards - instead of trying to get 3.6G stable, start out much lower and see how far you can get it to OC at stock voltage before it becomes unstable. Then start increasing the voltage until it's stable, once stable, increase the frequency, check for stability and increase voltage until it's stable, lather rinse repeat. You'll get to a point where the next bump in frequency will require a lot more of an increase in volts to make it stable and that's your threshold.

If it's unstable at 3.6 with 1.65V, then I doubt your chip is capable of doing 3.6. That's just luck of the draw. Anything over 1.5V is starting to get into danger territory, so if you can't get it stable at settings over 1.5V, you're going to have to settle for a lower overclock.


Also - make sure you only have 2 sticks of ram in the board and that you have the ram set to the slowest timings as I listed (5-5-5-15; 2T; and the overall memory speed is within your memory's spec). You need to make sure ram isn't impacting your overclock and you mistakenly think it's the CPU.

 

[dpk33]

8x400 is fully stable with 1.4175 volts in bios, 1.3 reported by cpuz under load. I tried 9x400 with 1.4175, 1.45, 1.5, 1.55 and 1.6. I guess my chip just wont do 3.6ghz.

 

[Shmee]

your chip has a good VID, mine was 1.25 and it could do 3.6, but I had a better board, though that was too much for my cooler.

If you are serious about ocing this chip to its potential, I would also recommend a better board. FYI, you will still need to play with NB voltage IMO, for quads.

 

[lavaheadache]

Yes he still will need to up the north bridge a notch or two but I really don't think that motherboard is up to the task at 3.6 ghz. Honestly, a p45 would make light work of this. But also at 3.6 cooling starts to become an issue on 65nm quads.

 

[dpk33]

Yes he still will need to up the north bridge a notch or two but I really don't think that motherboard is up to the task at 3.6 ghz. Honestly, a p45 would make light work of this. But also at 3.6 cooling starts to become an issue on 65nm quads.

Well, I guess I'm happy with my chip now at 8x413. I can't seem to really get any type of fsb over 300 on the 9x multi. Weird board/chip. It sits on 1.45 volts and load temps are around 60-65.

 

[Shmee]

thats a pretty good OC for an old board. Definitely good compared to the 680i series lol. Also, those are good load temps, mine were higher, but then I was on a xigmatech dark knight with only one fan. Your cooler I believe is just better lol.

Glad you found a nice stable OC. grats!

 

[rbk123]

That board will do 3.6G just fine; I've been running my IP35e at 3.6G for almost 4 years now. Not knowing what the memory is set to, it's not clear that that's not causing the fail as something isn't right if the 9x multi isn't working over 2.7G. If you can hit 3.2 at 1.4175, you probably can hit 3.4 and most likely 3.6. Something else is not right, and it's not the board. That board was very popular and many people hit 3.6 and over on various chips.

 

[Blitz KriegeR]

8x400 is fully stable with 1.4175 volts in bios, 1.3 reported by cpuz under load. I tried 9x400 with 1.4175, 1.45, 1.5, 1.55 and 1.6. I guess my chip just wont do 3.6ghz.

Sorry to chime in so late! This reminds me a lot of my own chip. My chip will do 6x400 at stock (board default 1.25), 7x400 at 1.3125, and needs 1.42v to run 8x400. That's with LLC, without I was using 1.5125 for 100% stability @ 3.2Ghz. My chip won't do 3.6, I'm not happy pushing the voltage above 1.5v.

For reference, here are my settings for this speed; (sig settings)
My current spec: 3.2Ghz 400x8
1.5125v VCC (1.325 default VID) // 1.425v VCC /w LLC (results in 1.40v idle, 1.408v load)
1.20v PLL (default)
1.50v FSB Term (default)
1.25v MCH / NB (default)
1.05v SB (default)
1.5v SB2 (default)
2.0v DRAM (board default 1.80v, EPP 2.0v default)

I've similarly strange walls with my setup. For example, if I change the FSB strap to 400Mhz for 1:1 memory it fails to boot 100% of the time. I can't figure it out; I've relaxed the tRD to 16. MCH all the way to 1.4v, PLL to 1.35v, FSB to 1.6v and it still fails to boot at 400Mhz 1:1. Oh by the way 333x9 also fails to boot no matter what I tried.

 

[Shmee]

hmm, sounds like you got one of the later, not so good G0's, as may be indicated by your VID, which is pretty high for a G0. for reference, my chip is 1.25, and the OPs is lower.