4.4gig Phenom 2 validation done on Air

[chew*]

It's dirty, it's ugly and I downclocked quite a few things but the important thing is it was done on air albeit cold air, it was done with all 4 cores and it was done with 1.55v

Temps were ballpark 25C at the core.

http://valid.canardpc.com:80/show_oc.php?id=509917

[EnJoY]

You are the man!

[Oats]

Nice clocks for air

[Marco André]

Hi chew*,

You reach that speed from Bios or by software? Nice chip

[chew*]

Hi chew*,

You reach that speed from Bios or by software? Nice chip

K10 stat, boot into windows at 3.0 1.55vcore.

[cky2k6]

Is it me, or do these chips seem to clock pretty poorly on air compared to how beastly they are on ln2? Seems a chip that does 6.7ghz shouldn't have much issue with a validation well into the high 4ghz range at least. Do they get unusually hot, or is there another reason for this diparity?

[chew*]

Is it me, or do these chips seem to clock pretty poorly on air compared to how beastly they are on ln2? Seems a chip that does 6.7ghz shouldn't have much issue with a validation well into the high 4ghz range at least. Do they get unusually hot, or is there another reason for this diparity?

They don't like heat as in 60C, and scale much better cold. This is just a guess, but a side effect of getting rid of the coldbug makes them less tolerant to heat. I personally will take a heat bug over cold bug anyday

[x-powerx800pro]

Very GOOD ,I have to 0852GPFW air-cooled only 1.55V under 4.2G been under 1m. Interior temperature of 25 C

[Stefan Payne]

What about the internal Northbridge inside the processor?
How was it clocked??

[[XC] gomeler]

Very impressive! I can barely get 3.8GHz out of mine, hoping it'll be a nice LN2 chip though.

Is it me, or do these chips seem to clock pretty poorly on air compared to how beastly they are on ln2? Seems a chip that does 6.7ghz shouldn't have much issue with a validation well into the high 4ghz range at least. Do they get unusually hot, or is there another reason for this diparity?

They just don't like getting hot. They scale less with voltage and more with temperature. Whatever AMD changed in the manufacturing process, it made them clock about equal on air with Phenom + 200MHz but a hell of a lot of fun on sub-zero cooling.

[chris.y2k.r1]

They don't like heat as in 60C, and scale much better cold. This is just a guess, but a side effect of getting rid of the coldbug makes them less tolerant to heat. I personally will take a heat bug over cold bug anyday

I've noticed the same thing, if you can keep the cores under 40c with water or air you can get some pretty amazing clocks...

[chew*]

I had dropped everthing down, cpu-z validations are all about cpu frequency.....its not about how preety your settings are to get there you don't get a cookie for having the 10th fastest cpu with a very high NB, you do get a cookie for the fastest cpu mhz though

Cold air helps alot, having a heatsink with some mass to it that can retain the cold air better helps as well, I did that on the OCZ vindicator, I now have a true with quite a bit more weight to it.

A TRUE copper would more than likely hold the max air cpu-z throne with some cold air over anything out there due to its mass.

Increasing voltage just ticks them off and adds more heat, there is a fine line between voltage/ temps.

[Stefan Payne]

you don't get a cookie for having the 10th fastest cpu with a very high NB, you do get a cookie for the fastest cpu mhz though

That's what I don't understand 'cause the impact on performance isn't that little, although depending on the application.

[FlanK3r]

i found 2 guys better on AIR and with 1.68 on AIR, the best is now (any guy from Japan)

http://valid.canardpc.com/show_oc.php?id=484849

[chew*]

I'd be curious to see that guys setup, 4.7 is a little suspect. 4.5 is probably doable, 4.7 maybe if you covered the board and heatsink in DI or poured ln2 on it or was testing on MT everest. got a link?

Actually that is Dumo from NY city lol........he's on phase was at my house saturday

[chew*]

That's what I don't understand 'cause the impact on performance isn't that little, although depending on the application.

thats why there are other benchmarks that measure performance.......

[Lightman]

Congratz chew*

Nice to see 4.42GHz on all 4 cores AIR!
You've beat me on that one

I can try 1 Core 4.5GHz if it get cold enough outside, but 4 cores 4.3GHz for my chip would be awesome!!

You have a gem! You need to sent flowers to Simon

I can agree 4.5GHz is doable, anything above is possible on AIR only if you live in Siberia!

[chew*]

Congratz chew*

Nice to see 4.42GHz on all 4 cores AIR!
You've beat me on that one

I can try 1 Core 4.5GHz if it get cold enough outside, but 4 cores 4.3GHz for my chip would be awesome!!

You have a gem! You need to sent flowers to Simon

I can agree 4.5GHz is doable, anything above is possible on AIR only if you live in Siberia!

This is the "runt" of the litter I need to send a kidney at the very least

[Lastviking]

i found 2 guys better on AIR and with 1.68 on AIR, the best is now (any guy from Japan)

http://valid.canardpc.com/show_oc.php?id=484849

Thats not with air... its someone thats have miss write it to air trust me , you can find the score in Phenom II List of overclocks(and when Dumo did write it he did write it was not air )

chew* nice with 4419mhz air valid but come back when you dont use a ES chip it will be more fair then

But its still a very great chip.

[chew*]

Thats not with air... its someone thats have miss write it to air trust me , you can find the score in Phenom II List of overclocks(and when Dumo did write it he did write it was not air )

chew* nice with 4419mhz air valid but come back when you dont use a ES chip it will be more fair then

But its still a very great chip.

Hmmm yah well I figured the board I'm on and an OCZ vindicator with a bench setup in a room and not hanging out a window makes it break even, however I will give the m4a79 deluxe a run with the 940BE already booted it up on stock cooler at 4.0 with only 1.475V, IIRC I needed 1.55v on the gigabyte also the gigabyte refused to boot over 1.6v

In the fwiw department I have not submitted it other than to validate it for proof, really have no interest in air records to be truthfull as they are not records, just showing you guys what you can expect to see in the near future.....

Cheers

[FlanK3r]

Thats not with air... its someone thats have miss write it to air trust me , you can find the score in Phenom II List of overclocks(and when Dumo did write it he did write it was not air )
LASTVIKING:hm, maybe its true :-).
But 4.4GHz on AIR is ofc impressive overclock :-). Im waiting on x4 945 and il try in classic case get 4GHz with air cooler (and ASUS M4A79 Deluxe).

[Lastviking]

Hmmm yah well I figured the board I'm on and an OCZ vindicator with a bench setup in a room and not hanging out a window makes it break even, however I will give the m4a79 deluxe a run with the 940BE already booted it up on stock cooler at 4.0 with only 1.475V, IIRC I needed 1.55v on the gigabyte also the gigabyte refused to boot over 1.6v

In the fwiw department I have not submitted it other than to validate it for proof, really have no interest in air records to be truthfull as they are not records, just showing you guys what you can expect to see in the near future.....

Cheers

Its looks like the Asus board have much greater pwm than the Dfi and Gigabyte board, the dfi board also have some trouble over 1.55v-> maybe its the cpu but i think its more the board after your info.

[Blazer_]

Nice scores.

I will tomorror 4.4GHz, i have now "max" ~4.3GHz,
http://www.hwbot.org/compare.do?resultId=816903

And Wprime 4GHz
http://www.hwbot.org/compare.do?resultId=817230

All Air, Wprime go more..but that was "race" stat..
Tomorror if it goes ~4.2Ghz ?

Ps. Can i change my nick here ?

[Solid Snake]

Awesome Man

Gigabyte rulzzz

nice score man on air

[chew*]

Validations are great, they are even better with video proof....

http://www.youtube.com/watch?v=1G2wr...e=channel_page

Managed to get a little more out of it and a legit validation
http://valid.canardpc.com:80/show_oc.php?id=519461