AMD X2 AM2 4600+ clocking

I have this chip, along with the Asus M2N32-SLi Deluxe motherboard and some Corsair 4-4-4-12 timing memory. CPUz info below

http://img421.imageshack.us/img421/3...4600oq0.th.jpg


This is basicaly the highest stable clock it seems to want to give out and I'm sure it has much more in it, I have gotton 2.8GHz iffy and 2.9GHz to post but not getting into Windows. I haven't done much towards the volts etc. as I'm unsure as to what I should change to enable it to get higher clocks stable.
Temps aren't much of an issue, I have a Zalman AM2 cooler (9500 or something) and for the current clock it's running at a low 1800rpm but has about 2600rpm capable.

What settings should I change to get a higher stable clock?

I'm sure this chip can push 3.0GHz (unsure if it can on aircooling admitidly).

I have the multiplier at x11 and the HT thing at x4 and the RAM is at 667mhz instead of it's rated 800mhz as recommended by an OcUK forum member.

With a little bit of tweaking I have gotton it stable at 2760mhz at 41C idle with a vcore of 1.5v, FSB of 230mhz, 12xMulti. If people know how I can get it further please tell me.

Cheers,

FinalD

2.8 seems to be about the limit for these. My highest benchable. http://valid.x86-secret.com/show_oc?id=125623

484 bus speed? that on x2 HT ?

That's a shame really, I was hoping I'd be able to hit the 3.0 mark, ah well, roll-on 65nm.

Cheers chucksull

FinalD

came by this on another forum i use,

http://www.xbitlabs.com/articles/cpu...x2-3600_3.html


they used 3600 upto 2.6ghz, it makes interesting reading.

Quote:

---

Originally Posted by FInalD

484 bus speed? that on x2 HT ?

That's a shame really, I was hoping I'd be able to hit the 3.0 mark, ah well, roll-on 65nm.

Cheers chucksull

FinalD

---

Yeah I dropped right down so I could just set the FSB from within windows without worrying. It wasn't stable enough to run 24/7, though. I suspect it'd hit 3.0ghz with better cooling, but 200mhz isn't worth installing water for me :P

More voltage, more cooling.

I've been pumping 1.65v through my venice for over a year, on mostly silent air cooling, with no problem.

A dual core chip obviously would put out more heat at equavalent voltage, but still...

One of my problems is that I don't know the max Vcore for the chip, I don't exactly want to overload it. At 1.5v it doesn't change it's stability all it does is increase the temps and if anything make it less stable.

After reading some of that article about the 3600+ I may have to do some tweaking and see just what the max is, as 2760mhz from a 2400mhz isn't that great a clock, if they can get 30% out of their's so should I (simpleton's logic :P), and this would mean I can in theory get 3120mhz out of the chip.

FinalD

The max vcore is whatever you can use that stays stable and under the max tcase. If you are idling at 41C that is a problem. Even at 1.65v, with a 911MHz OC my chip idles at 6-8C above ambient. Load temps are rather high(within a few degrees of it's rated max), but it is stable.

Your theory doesn't really hold much water. A lot of these chips just dead end around 2.7-2.8GHz, no matter what they are ratted for. The process has it's limits. My chip will do a 200Mhz OC with a .25v undervolt, remains stable at stock voltage to 2.6GHz, then needs another quarter of a volt for the last 110MHz where it hits a wall and will not remain stable even 5Mhz higher with any reasonable cooling. Your chip seems to be at the point where it needs a rather great increase of voltage for the last bit of speed you may be able to get out of it.

Checking the rated TDP and tcase max can help you guess at how well your chip should be OCing. I've found this program useful: http://www.thecoolest.zerobrains.com...topic.php?t=83

2760mhz seems to be the speed these chips like, it's actuall what mine runs at now I running grid.org. not worth the extra temp and volts(more importantly to me, fanspeed) to get the last 100mhz. Thats just me though. I like things to stay cool and very quiet.

@ oralpain - I was more just saying in theory if a lower chip gets 30% so should all other chips, not that it is true - I'm not stupid. The max Vcore is 1.5v or just above as I've read on the AMD site, along with the info that the max temp is 50C-70C, so wherever it idles it's fine, plus I have my cooler running about 1800rpm and it can easily do 2800rpm, but I like it to be quieter, this is the reason for the 40C (max) idle temp. Also I live in a hot flat which is well above normal ambient room temp and I have no control over the temp of the room (window open and heating off).


@ chuckskull - it does seem to be the top end of the chip, which is a bit of a shame really that AMD didn't make it highly overclockable, but then when the 65nm chips come out I plan on getting one anyways.

FinalD

http://arstechnica.com/news.ars/post/20061011-7961.html :D