i3-550 questions ...

[Gravewyrm]

Wondering if anyone has managed to get an i3-550 to overclock stably?
Got one given to me * still NIB even * and thinking of tinkering with OC'n it with my next build.

Cheers ..
~~Grave

 

[frkino]

I just recently got an i3-530 and managed to clock it @4.4GHz.
I ran it through burn tests in Windows environment to make sure I got it stable.

Ram is running @1600MHz and QPI @4.8GT/s

Hope this helps

 

[Gravewyrm]

I just recently got an i3-530 and managed to clock it @4.4GHz.
I ran it through burn tests in Windows environment to make sure I got it stable.

Ram is running @1600MHz and QPI @4.8GT/s

Hope this helps

Interesting. I'll have to test it out once i get the rest of the build together. PSU, Case, and DVD drive is all thats left

 

[frkino]

It's funny, I would have expect this built to perform very closely to some of the lower end Sandy Bridge iMacs. But it was a whole lot of a different story when I benched it with GeekBench. I was nowhere close to 5 figures scores.

At this stage I am more convinced that there must still be something that weren't running in their native mode and fell back to some software compatibility mode. It is also particularly worth mentioning that GeekBench identified my 4.4GHz as 159MHz

So anyway, sometimes you just can't rely on GeekBench to brag about things, it is just so blatantly obvious that my current built is more responsive than some of the previous, lower-end built MacPro, yet Geek Score will suggest the otherwise.

Good Luck to you too!

EDIT: An update, and I was right!
There were a couple of things that weren't running the way they were expected to. I have now switched all those Auto options in BIOS to Enable. My Geek Bench Score improved massively to around 8700 despite it is still recognizing a 105MHz CPU. The current score proved my pre-built hypothesis that the chip should firmly establish its stand in the realm of existing quad-core Core i5 machines once its potentials is exploited.

 

[Gravewyrm]

It's funny, I would have expect this built to perform very closely to some of the lower end Sandy Bridge iMacs. But it was a whole lot of a different story when I benched it with GeekBench. I was nowhere close to 5 figures scores.

At this stage I am more convinced that there must still be something that weren't running in their native mode and fell back to some software compatibility mode. It is also particularly worth mentioning that GeekBench identified my 4.4GHz as 159MHz

So anyway, sometimes you just can't rely on GeekBench to brag about things, it is just so blatantly obvious that my current built is more responsive than some of the previous, lower-end built MacPro, yet Geek Score will suggest the otherwise.

Good Luck to you too!

EDIT: An update, and I was right!
There were a couple of things that weren't running the way they were expected to. I have now switched all those Auto options in BIOS to Enable. My Geek Bench Score improved massively to around 8700 despite it is still recognizing a 105MHz CPU. The current score proved my pre-built hypothesis that the chip should firmly establish its stand in the realm of existing quad-core Core i5 machines once its potentials is exploited.

Ya, ive been toying with the idea of OC'n a "Low End" CustoMac for a while, seen some impressive numbers from i3's and low i5's. I should be starting my build in the next cpl weeks, just waiting on parts. I'll have a thread about it when i get it all goin

~~Grave

 

[Gravewyrm]

currently have a rock stable 3.6Ghz OC on my i3-550. gonna see if i can take it a bit higher.

*EDIT*

I did a burn in with Prime95 today, did what i call a "Wave Form" test.

10 minute burns, with 2 minutes cool down. did 10 burns total. System seems to be stable in windows, with one odd glitch. UAC makes the sound go corrupt, and thus locking the system up with white noise comming from the speakers.

Managed to get UAC disabled, and so far, so good.

Temps atm are hovering around the 30*C mark idle, and with prime burnin, using both cores, all 4 threads, temps maxed at 72*C(!!) changed the test layout from massive heat / CPU intenseive stuff, to the default method. Temps never went over 65*C, with all cores burning.

This OC is sorta buggy in Win7, but in OSX, ive not encountered any issues, to date.