someone please help me overclock

[ByteJuggler]

Cheers Bytejuggler. Just upping the host frequency, will that mean the temps should in effect remain exactly the same?

Temps will be slightly higher purely due to increased clock rate. If you increase the voltage(s) then temps will go up slightly more still. Up to 2.8Ghz from 1.8GHZ is more than 50% increase in clock rate, so I'd expect a non-trivial increase in temps (but it should not be too excessive.)

 

[Supercow]

Temps will be slightly higher purely due to increased clock rate. If you increase the voltage(s) then temps will go up slightly more still. Up to 2.8Ghz from 1.8GHZ is more than 50% increase in clock rate, so I'd expect a non-trivial increase in temps (but it should not be too excessive.)

Thanks a bunch for your help mate - im off to do some stability testing
Think I'm finally kinda getting to grips with the basics
What kind of temps - on relatively crappy air should be ok , or moreso, when should I be concerned?

 

[Gaijin]

Does the same apply to quad overclocking?

 

[ByteJuggler]

Thanks a bunch for your help mate - im off to do some stability testing
Think I'm finally kinda getting to grips with the basics
What kind of temps - on relatively crappy air should be ok , or moreso, when should I be concerned?

Hey mate, that's a pretty good overclock by the looks of things. It looks like you've not even upped the CPU voltage much if at all (?) Very very good. It looks like that chip might have some more headroom to go further given that you've apparently got as far as you have on stock voltage. But for now, it's best to confirm that what you have is actually stable. I'd expect temps on crappy air for you to perhaps be high 30s, under load perhaps 40-50C. Max temperature is probably around 60C. Tom's hardware has a page on temperatures that you might want to look at. (http://www.tomshardware.co.uk/forum/221745-11-core-temperature-guide) Someone else can perhaps confirm or correct my termperature estimates as appropriate. For proper temperature testing you might also want to download Intel's TAT tool. (Temperature Analysis Tool.)

Anyway, I would suggest you now stress test with a couple of stress testing apps. Orthos is one which I personally like. It is quite good at highlighting marginal instabilities under stress, both in the CPU and RAM. Fortunately, your RAM is running at its rated speed, so for your currently overclock you don't have to worry too much about the RAM stability. That said, you probably might still have to up your RAM voltage some (by 0.2v or 0.3V) because the motherboard may default to a too low voltage for the RAM you've got, as one of the other posters have mentioned. Although then again you may be lucky and it may not actually be neccessary (as the RAM may run stably despite being undervolted slightly.) To be quite sure, you'll need to test your RAM with e.g. Memtest86 to prove it's all good, then if any errors are reported, you can up the RAM voltage a little to bring it to the spec required by your RAM, which should eliminate any instability.

Anyway, as I say, run the machine a couple of hours or so on several of the Orthos modes, to make sure no obvious instabilities are present and/or run some memtest tests to make sure the RAM is fine. Once both of those report a clean bill of health you can conclude with a successful first overclock attempt.

 

[ByteJuggler]

Does the same apply to quad overclocking?

Yes, the same techniques etc etc apply to quad's as well. However, of course you need to make sure you test all the cores for stability etc.

 

[Supercow]

Some experimentation, and got it to 3.2!
Well happy about that

Orthos only ran for 2 hours - will do a propper session at some point this weekend.
Arctic silver only applied a day ago, so want that to settle a bit first, might clock back to 3GHZ while the silver settles

 

[CyZMyAssBaby]

3.2 on cud well chewed Sup3rc0w m8 now get from behind that ant hill its an exclusion zone

Nice post and friendly advice ByteJuggler m8