Get your overclocks out for the lads!!!! (The Return)

Status
Not open for further replies.
Reduce the speed of the memory as well, best to to find what your CPU is capable of without factoring in the memory. You could find its your memory causing issues rather than CPU and you can achieve 4ghz on much lower vcore. You should really seperate the two, CPU then memory
 
Last edited:
OK, when I called it a night I had dropped from an apparently (needed to be properly left to test) 4.0Ghz @ 1.390 down to 4.0Ghz @ 1.387 (this was dropping two increments) which failed IBT. Temps were looking better but its irrelevant as it wasn't stable.
As advised Memory has been dropped down from 716Mhz down to 5xxMhz so probably clocked at 1000-1100Mhz instead of 1600Mhz that I was at with my stable 3.8Ghz OC.
 
mine is at 3.8ghz mate. good enough. 1.25v, 71c max using intelBurn test which destroys/caines your cpu far more then prime does.

am happy.
 
IBT is superb, works the Cpu much harder than prime i find.

Its only thing that makes my i5 2500k at 4.5ghz go over 60degC.
 
Think you might have found the limit if a voltage drop that small causes a crash. Might be worth trying for 3.9 @ 1.387 and call it a day? otherwise looking at a better cooling solution WC etc. Great funs but a lot more cost.
 
AJBek said:
Think you might have found the limit if a voltage drop that small causes a crash. Might be worth trying for 3.9 @ 1.387 and call it a day? otherwise looking at a better cooling solution WC etc. Great funs but a lot more cost.

Uhh?
 
OK, when I called it a night I had dropped from an apparently (needed to be properly left to test) 4.0Ghz @ 1.390 down to 4.0Ghz @ 1.387 (this was dropping two increments) which failed IBT. Temps were looking better but its irrelevant as it wasn't stable.
As advised Memory has been dropped down from 716Mhz down to 5xxMhz so probably clocked at 1000-1100Mhz instead of 1600Mhz that I was at with my stable 3.8Ghz OC.

What timings you run on the memory?
 
Going to be playing about with my 2600k tonight. At the minute I am running at 102 x 45 but need to reign in the voltage a bit. I am hoping to get close to 5ghz tonight. Will be interesting.
 
I never really bothered with IBT as used prime95 but you guys made me curious. What setting is best? And how many runs do you use?
 
I'll normally start with prime but I don't tend to use blend. If I look stable over an hour I used to then try LinX but these last few weeks I haven't bothered. Instead I hopped to the IBT and let it run for 10 cycles. If that passes I'll leave prime running for a good few hours.
 
I'll normally start with prime but I don't tend to use blend. If I look stable over an hour I used to then try LinX but these last few weeks I haven't bothered. Instead I hopped to the IBT and let it run for 10 cycles. If that passes I'll leave prime running for a good few hours.

a lot of people advise 24 hours of prime on blend.

That's what I did to check stability on here and then hit it with LinX. IBT is far to extreme for testing day to day gaming and p.c use.

I'm only at 4Ghz on my 1500k but its plenty for anything I do :rotfl:
 
Just used to hours of prime and 10 cycles on extreme just seems short :). Is it the toughest stress prog out there then?
 
Just used to hours of prime and 10 cycles on extreme just seems short :). Is it the toughest stress prog out there then?
Personally I stick to LinX as I've seen overclocks fail under LinX that had passed Prime but were unstable during some real-world uses too. I guess if you have some time then it's best to give them all a blast.
 
Personally I stick to LinX as I've seen overclocks fail under LinX that had passed Prime but were unstable during some real-world uses too. I guess if you have some time then it's best to give them all a blast.

Doesn't IBT use the LinX but with a 'user friendly' GUI?
 
Doesn't IBT use the LinX but with a 'user friendly' GUI?

Not sure, but I doubt it. IBT was written by/for Intel to weed out faulty processors, and fail those not good enough for going to retail - it was essentially designed to break processors. As such, many people see it as being too strenuous for most peoples needs - most people in everyday gaming / use will never put their processor through the stress even prime 95 will show it.
 
a lot of people advise 24 hours of prime on blend.

That's what I did to check stability on here and then hit it with LinX. IBT is far to extreme for testing day to day gaming and p.c use.

I'm only at 4Ghz on my 1500k but its plenty for anything I do :rotfl:

temps are key for me as im on air. IBT is obviously going to show your system under the most extreme temps so every OC i do gets tested with it. Blend in Prime tests memory aswell which im not so interested in. Large SFTs are normally what I use.

Id love to get my hand on a SB build and go for 5Ghz :p
 
Yea i highly recommend people tets on IBT and nothing else. get IBT tests at below 80c and your golden
 
Not sure, but I doubt it. IBT was written by/for Intel to weed out faulty processors, and fail those not good enough for going to retail - it was essentially designed to break processors. As such, many people see it as being too strenuous for most peoples needs - most people in everyday gaming / use will never put their processor through the stress even prime 95 will show it.

They both use linpack apperntly. I got me lin's a bit wrong!
 
temps are key for me as im on air. IBT is obviously going to show your system under the most extreme temps so every OC i do gets tested with it. Blend in Prime tests memory aswell which im not so interested in. Large SFTs are normally what I use.

Id love to get my hand on a SB build and go for 5Ghz :p

Don't think mine goes over 65 degrees and voltage is at 1.3v :D

That's on air as well with a Corsair A70
 
Status
Not open for further replies.

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom