• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

can IBT be used to compare cpu throughput across platforms?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

caddi daddi

Godzilla to ant hills
Joined
Jan 10, 2012
like the title asks, can intel burn test be used to compare, let's say my 4790k rigs throughput to my 965 rigs throughput and such?
in ibt my 965 rig makes 51-52 GFlops @3.8, my 4790K rig makes 129 GFlops @4.4, is this accurate?
I also use super pi and wprime.
what would you use?
 

Silver_Pharaoh

Likes the big ones n00b Member
Joined
Sep 7, 2013
I'd like to think we can caddi!
Just a front end for Lin pack, which every computer can run so I guess you can look at it like a benchmark! :thup:
 

mackerel

Member
Joined
Mar 7, 2008
Haven't looked at IBT in a while. Unless I'm looking in the wrong places, it hasn't been updated since 2012, which pre-dates FMA. Reason for mentioning that is I understand it is big on FPU performance. For example, in Prime95 AVX gives about 2x IPC compared to without. If you factor that difference between the 965 and 4790k, allowing for clock, it is roughly in that ball park. The remaining difference might be due to ram performance. If IBT supported FMA, quite likely Haswell and later would show even higher results.

So like any benchmark, it only shows how fast your computer is at running that benchmark, and caution needs to be applied to interpreting that.
 
OP
caddi daddi

caddi daddi

Godzilla to ant hills
Joined
Jan 10, 2012
that's why I use more than one and tend to think that super pi is better to compare throughput.