i5 3570K quick test.

Ollie1132 · 27 Apr 2012 at 05:58

Nice. Are you going to attempt past 4.8Ghz?

stulid · 27 Apr 2012 at 06:08

nope, I might have been a bit early to say the temps were similar, I'm in the low 70's

30mins of prime so far, 65,73,72,67

the maximum s are higher, 68,77,75,70

Ollie1132 · 27 Apr 2012 at 06:27

Thats not all that big of an increase however. Your still managing to keep the temps down even though they are approaching somewhat high levels now. So long as you don't break 1.38v and 4.8Ghz I do believe your temps should stay just below 80C across all cores with between something like 75-85 for max temp respectively for each core. Just a prediction however. But lets hope Im somewhat right if not the max temps are lower.

thenewoc · 27 Apr 2012 at 06:32

Good read, this IB vs SB has had me on a knife edge. Thanks stulid for being an early adopter and sharing your findings.

@ollie, cooler performances, http://www.vortez.net/articles_pages/alpenfoehn_k2_cpu_cooler_revisited,10.html

stulid · 27 Apr 2012 at 06:34

one of the cores has touched 78.

vcore in CPU-Z reading 1.392

coming upto one hour prime95

Ollie1132 · 27 Apr 2012 at 06:53

weird your on 4.7Ghz at 1.39v? once prime95 finishes try changing your multiplier and see if your stable with 4.8Ghz, if your not I wouldn't suggest going past 1.4v since you've probably hit your thermal limit. @thenewoc Ive already done a tonne load of research on coolers and IB but thanks anyway

covenantuk · 27 Apr 2012 at 07:41

I know the 5ghz is nice to see in a sig, but what's the real world performance like? If you're getting 4.7ghz most of the reviews suggest that's roughly equal to 5.1ghz on SB?

the_director · 27 Apr 2012 at 08:42

Interesting. I am wanting 4.5 stable so it seems that is achievable. Not sure why you changed your 2500k though - apart from having a new chip to play with.

chriscubed · 27 Apr 2012 at 09:19

I'm hoping to run around 4.5 on a reasonable vcore and temp so I think I can buy with confidence now, based on your findings. Thanks stulid.

Will be interesting to see the final results of what is achievable though, and whether IB at 4.7 is comparable to SB at 5.0, benchmark wise.

stulid · 27 Apr 2012 at 10:49

Unless I have done something wrong with the 2500K score, then the above is correct.

The 2500K was when it was on a GA-Z68XP-UD4, the 3570K is on a UD5H.

the1gooner · 27 Apr 2012 at 10:53

^^^ very small difference then in SuperPi, thanks for posting this is helping a lot of people make up their mind what to go for.

bulldog147 · 27 Apr 2012 at 11:18

stulid - Can you ran 10x at the maximum IBT (Intel Burm Test) please on your 4.7GHz and see if it passed ?

thenewoc · 27 Apr 2012 at 11:30

@ollie, no worries, just that you raised an interesting point about closed loop water coolers of which the link above shows how they fair, was a surprise to me too.

nuttyprot · 27 Apr 2012 at 11:31

just wondering what are the safe temps for ivy?

bulldog147 · 27 Apr 2012 at 11:39

nuttyprot said:
just wondering what are the safe temps for ivy?

got to be below 70C same as sandybridge! Ivy got more heat problem than Sandybridge

Spenty · 27 Apr 2012 at 11:42

bulldog147 said:
stulid - Can you ran 10x at the maximum IBT (Intel Burm Test) please on your 4.7GHz and see if it passed ?

If your overclocking and using IBT on Maximum you will never reach a good OC to be fair. IBT on max creates temps 10-15c above blend on Prime95.

bulldog147 · 27 Apr 2012 at 11:46

Spenty said:
If your overclocking and using IBT on Maximum you will never reach a good OC to be fair. IBT on max creates temps 10-15c above blend on Prime95.

Yes that true but IBT is a quick test to detect early BSD ! Prime 95 take longer to detect BSD like eg: after 15 hours might get BSD! That's why I did 20x maximum IBT (passed) then follow by 35 hours prime 95 and it passed and all in good go!

opethdisciple · 27 Apr 2012 at 11:55

Is 4.5ghz failry easy to achive using a £30-40 air cooler?

hyperst · 27 Apr 2012 at 12:02

Glad to see you are getting better results stulid, at least now you can be confident it was a good move after all,I went ahead with getting a 2500k and 8gb of lp corsair vengence for £194 delivered, kinda just had to buy it as funds were tight anyway and I think IB would have pushed me too far over budget as I would have spent more on memory to get some faster stuff.

Dervious · 27 Apr 2012 at 12:18

hyperst said:
Glad to see you are getting better results stulid, at least now you can be confident it was a good move after all,I went ahead with getting a 2500k and 8gb of lp corsair vengence for £194 delivered, kinda just had to buy it as funds were tight anyway and I think IB would have pushed me too far over budget as I would have spent more on memory to get some faster stuff.

I`ve just done the same thing, same CPU, same RAM (at last, I have Corsair RAM that works first time !).

The big thing for getting IB is that it is new. A slightly less important thing is that it might be a bit faster (although I doubt enough to really notice). SB is proven technology, and until we get a clear picture of what IB is capable of, buying IB at launch is a bit of a gamble. Having said that, I don't think you can really lose with IB or SB, unless you want 5ghz and can only get 4.2.

If I were buying today, I'd still go i5-2500k over the new IB. In fact, that's what I did a few days ago. If I was really convinced and patient enough to go with an IB, I'd be waiting a few more months.

BTW.

opethdisciple said:
Is 4.5ghz failry easy to achive using a £30-40 air cooler?

4.6ghz on a £25 cooler with SB is fairly easy. I wouldn't be so sure with IB.

In the end, it's all about practicality vs price vs desire to "play". If you fancy the challenge of getting the most performance, despite price, then IB is the way to go. If you want VFM and enjoy the task of OCing to a reasonable level, then SB still makes a lot of sense.