Intel has solved the temperature problem in Haswell ?

[hersounds]

An early sign of the upcoming fourth generation core i5, was received and tested by chinadiy site.

The exhibition runs at 2.8 Ghz sent and turbo mode to 3.0 Ghz ... the final version will be higher 3.5 / 3.9 Turbo ( 4770k ) but still allows us to have an idea.

Test temperature 18 degrees

Idle 5 minutes



Intel burn test 2.54 - full load



The maximum TDP of this sample is 83 watts, the final version is 84 watts.

Finally bear the memory controller (via overclocking) DDR3 3200 mhz.

Lack and little by little come out reviews and tests premature Haswell suffice your response in OC to see if Intel has finally solved the problem that both suffered Ivybridge.

[Gunslinger]

Solder FTW?

[Computurd]

low power goodness

[eXa]

Uhm, that looks really hot actually.
10+c above room temp idle and 35+c above room temp during load. For a 3ghz cpu...

[Toolius]

I hope they do solder it.. Ivy sucked pretty much !

Just noticed.. cpu fan is at 900 odd rpm ?? If that is the silly stock intel heatsink with its fan at 900 rpm then it's not bad.. !

[Movieman]

I hope they do solder it.. Ivy sucked pretty much !

Just noticed.. cpu fan is at 900 odd rpm ?? If that is the silly stock intel heatsink with its fan at 900 rpm then it's not bad.. !

Exactly, if that is one of those POS factory Intel HS then IMO all this is good for is a reference
100% load on a quad with one of those HS in the high 50'sC isn't that bad especially with the load from Intel Burn test WHICH is much more than anyone will ever toss on a home PC..

[Splave]

Solder FTW?

Nope tim :P but thats a good thing this time

[Dumo]

This...http://kingpincooling.com/forum/show...54&postcount=1

[Movieman]

This...http://kingpincooling.com/forum/show...54&postcount=1

We may need Vince's pots just to get 4GHz out of these!

[OldChap]

No, the heatsink used in the above test is this one:

http://www.smalltao.com/product/1713...155-2011%20B81



Fan is at a constant 900 rpm

http://translate.google.co.uk/transl...584%25E7%2594%

[paulbagz]

Wasn't exactly a hard fix

-PB

[OldChap]

Interesting memory performance options....

http://translate.google.co.uk/transl...%2Fn-8458.html

[owikh84]

The Gflops looks rather low. Should be getting 125+ for i5 then only it is valid.
Did I see a 2800MHz north bridge frequency?

[dasa]

be nice if they used something better than a ihs that makes bad contact on the k models at least we would be getting something other than just a unlocked multi for the extra paid

[AMDforME]

Uhm, that looks really hot actually.
10+c above room temp idle and 35+c above room temp during load. For a 3ghz cpu...

It's an Intel CPU. They tend to run hotter than other CPUs, going back decades. They actually fried an egg a on PIII 600 and video taped it... Ivy Bridge has heat issues due to the poor FinFET design in addition to the TIM used. Intel has officially abandoned that FinFET design as a result.

[Olivon]

Ivy Bridge has heat issues due to the poor FinFET design in addition to the TIM used. Intel has officially abandoned that FinFET design as a result.

[eXa]

It's an Intel CPU. They tend to run hotter than other CPUs, going back decades. They actually fried an egg a on PIII 600 and video taped it... Ivy Bridge has heat issues due to the poor FinFET design in addition to the TIM used. Intel has officially abandoned that FinFET design as a result.

I have a 3570k. Its not delidded, it idles cooler than this haswell sample.

I still dont understand how people can scream halleluja form this result. Its a 3ghz sample, it will consume less power and produce less heat just becasue its at only 3ghz. It might even have less core voltage at load for the same reason for all that we know.
I see no evidence that the TIM is improved or that theyve gone back to solder.


Edit: And i cant remember they doing that on a PIII, but i do rememeber they frying a egg on a Athlon tho. Also, the tim issue isnt only the tim used, its the gap between the core and the heatspreader. Get your facts straight.

[eXa]

Look at how ivy bridge power consumption scales with frequency alone. If haswell is similar, that 3ghz is significant.

[kromosto]

vaow %20 increase in 1ghz

[Rollo]

It's an Intel CPU. They tend to run hotter than other CPUs, going back decades.

That would be a good trick as they always consume less power than "other CPUs", at least lately. What CPUs are you referring to AMDforME?

[repman]

Just a quick reminder: Heat ≠ Temperature

[Kallenator]

That would be a good trick as they always consume less power than "other CPUs", at least lately. What CPUs are you referring to AMDforME?

Is that a trick question?

Let's not feed the trolls

Just a quick reminder: Heat ≠ Temperature

Thanks, was about to say this myself.

[Manicdan]

amd numbers are useless because of the way they read out. my trinity htpc run at 0c in most cases

[Mumak]

Recent AMD CPUs feature an extremely inaccurate thermal diode on CPU, especially at lower temperatures.
Later Intel CPU DTS sensors should have an uniform accuracy of +/-5 C, but it seems there are some exceptions too.

http://www.hwinfo.com/forum/Thread-C...Facts-Fictions

Many users don't realize this...

Moreover, that CPU in OP is not even a Qualification Sample...

[mattkosem]

Isn't the on-package VRM likely contributing extra heat that wasn't previously there?

--Matt