Right from Intel Jim.

I'm with Jim Roseberry here. Think about this for a second..........

I OC my system, I also stress test before going live with an OC'd profile. If you stress test a system overnight. i.e. run the CPU at 100% for several hours, and it withstands that then it isn't going to have any problems especially running Sonar.

If you think Sonar is going to push your system anywhere near as hard as a stress test then you are going to need a complete rebuild with a new much, much quicker CPU & MB, not a bit of overclocking......

Just my 2p's worth....

Overclocking then....

I've been reading up about this on and off for a long time, but until a few days ago, had only done tentative and temporary experiments.

I'm a big fan of the economic benefits of keeping a well maintained system that's somewhat behind the bleeding edge; you can buy killer components for next to nothing when the real speed freaks sell off last year's stuff. So while my d975xbx motherboard and e6700 processor are fairly old news, my system overall is fairly nice, for a fraction of what it would have cost were I more finger on the pulse about it all.

So anyway, I recently managed to establish that the motherboard I have can (unofficially) take a quad core Q6700, and I found one on ebay for 90 quid, so I went for it.

As the old processor is both long out of warranty and soon to be retired, I though I'd have a go at applying what I've managed to learn about overclocking from it. Much to my surprise, its turned out to be fairly easy. I've now had a totally stable 18% overclock running since Monday, taking the e6700 from 2.6ghz to 3.15ghz. I'm sure there's more to be had, but I've not got time to experiment further right now, and 18% is not to be sniffed at.

More significantly - or I think so, anyway - is that I used the FSB method rather than just the multiplier method, so the whole system is faster, not just the processor. Mainboard overall is going at around 1250mhz, rather than the stock 1066.

Placebo effect may well account for some of this, but the difference seems entirely tangible to me. More pertinent to the thread above, stability is not at all issue. I've done both an overnight Prime 95 test and three days real work, all without problems.

I now think I'm going to OC the Q6700 from the outset, though I gather more cores means a bit more heat, so possibly a lower overclock. But in a general sense, it seems there's not much to fear from an OC, as long as you think it through carefully and do the maths in advance.

You do have to do your homework though. I've pieced it together from a LOT of googling. I never came across any one site that had a neat all in one guide to overclocking.