Ouch 103 Watt heat from the Prescott 3.6GHz CPU

[wetcat007]

http://www.xbitlabs.com/news/cpu/display/20030725155324.html 103 watts of heat will be blasted from the prescot cpu, as well as record ammounts of power required to run it. When will CPU's actually go back to producing resonable ammounts of heat??

 

[Corporate Thug]

i dont think prescot is gonna be an OCer...

 

[Tab]

I wonder what Intel's included HSF will look like. It prolly a massive aliuminum block

 

[vaporize]

so thats how intel is going to increase its speed, by pumping more current into the sucker.

 

[mechBgon]

Furthermore, that's the "thermal design power" if I'm reading it correctly. Absolute peak output on the Pentium4's was 4/3 of the TDP, according to Intel's datasheets. Wingz (or was it pm) has remarked that it's not really possible to get a CPU much beyond the TDP in real life, but for the record, 4/3 x 103W = 137W. That's still not the end of the world... the TDP for the Itanium is 130W, last I saw, and those can be crammed into 2U cases.

At any rate, it will be important not to restrict the airflow to/through the case, methinks. There should be a law against those computer hutches that coop the system up in an unventilated space! :disgust:

 

[Acanthus]

Originally posted by: shady06
i dont think prescot is gonna be an OCer...

I do. .09um and a voltage drop. If northwoods do 3.5 why would prescott have trouble going over 3.6?

 

[degeester]

That is the reason the "old" Canterwood and Springdale motherboards likely won't be able to handle the increased loads.

 

[touchmyichi]

Thats crazy...... Well psu technology will certainly improve with all of these new power heavy requirements. In fact I might wait out until this comes out before I upgrade my psu (mine seriously bites right now).

 

[Soulkeeper]

i wonder if the big pc companies like dell or gateway will start outfitting their prescott systems with cooling systems like that seen in the apple G5

looks like we have no choice but to get hotter and hotter with our current silicon technologies if we want to increase speed till something better comes along

 

[mrgoblin]

Originally posted by: Tabb
I wonder what Intel's included HSF will look like. It prolly a massive aliuminum block

Intels heat sink fan wont be a heat sink fan at all, itll be a hose to connect it to your ac. Also it will have a special case where the side ofthe case is actually an industiral strength fan. That thing is gonna be a blowtorch. Looks like intel really screwed up this time. Maybe theyll include watercooling kits for a small fee . I really think amd has em beat this time so it should be interesting to see what intel does. All I know is my amd stock is going up so I say burn baby burn!

 

[THUGSROOK]

looks like ill be skipping another Intel platform.

 

[MercenaryForHire]

YAGDMFPHOT
Yet Another God Damn Mother ******* Prescott Heat Output Thread

- M4H

 

[MrThompson]

Originally posted by: mechBgon
Furthermore, that's the "thermal design power" if I'm reading it correctly. Absolute peak output on the Pentium4's was 4/3 of the TDP, according to Intel's datasheets. Wingz (or was it pm) has remarked that it's not really possible to get a CPU much beyond the TDP in real life, but for the record, 4/3 x 103W = 137W. That's still not the end of the world... the TDP for the Itanium is 130W, last I saw, and those can be crammed into 2U cases.

Amen. I am glad to see others realize this. Have a look at the difference between Thermal Design Power and Maximum Power in Intel's latest data sheet. For example the current 3.06 GHz CPU has a Thermal Design Power of 87 watts and a Maximum Power of 96 watts.

 

[WarCon]

I imagine a great portion of the added heat will be from the additional 512k of cache memory the the Prescott is going to be sporting. With the PNI, new branch predictor, the 1 meg cache, and the strained silicon on a 7-layer low-K metal interconnect process in the .09 micron flavor, I am expecting these processors to perform very well. Just hope the cost comes down on these quickly.

Unless these aren't going to scale well, I see no reason (unless Intel implements something to stop overclocking) that these processors won't overclock. Its the range in the stabilty of the circuit design that helps the P4 overclock, not its thermal properties as much as it seems to be with AMD processors. I.E. When you are cool enough to reach your maximum overclock, only extreme cooling seems to matter to how much farther you can go. My previous 2.4B was stable at 2.8ghz with the stock heatsink and with a powerful water cooler and no more. Only thing that varied was the temperature. Where my old Athlon 1ghz T-bird was only stable at 1.26ghz or so with air cooling, but went all the way to 1.4ghz with this water cooler.

I don't imagine Intel is going through a tape out design phase for each speed bump in this processor series, just to gain the necessary stability to add 200 mhz. (Doesn't Intel plan to hit 5ghz with these processors?)

 

[stevejst]

(Doesn't Intel plan to hit 5ghz with these processors?)

Nope according to roadmaps published so far. This will go over 4.0 GHz when it will be succeeded by Teya which will be the one going for 5.0GHz.

But if you are looking for extreme cooling Pentium IV has broken the boundary of 5.0GHz already, using liquid nitrogen.

 

[Wingznut]

First off... Let me go on record saying that I'm not going to confirm if this rumor is true or not, and that I'm not going to speak specifically about future products. But for the sake of this post, let's presume that it is true, and that the TDP is ~100w....

So what? Do you think that if PSC is ~13w more than the current Northwood, that is going to change ANYTHING? We're talking 13w here. How many watts do you suppose your overclocked/overvolted 3.5ghz P4's are dissapating?

And when people talk about overclocking, do you realize that heat dissapation is not the roadblock? It's the gate leakage and elecromigration... And that's only going to become more of an issue (exponentially) with every process shrink.

Are we done overreacting to a rumor now?

 

[Wingznut]

Originally posted by: stevejst

(Doesn't Intel plan to hit 5ghz with these processors?)

Nope according to roadmaps published so far. This will go over 4.0 GHz when it will be succeeded by Teya which will be the one going for 5.0GHz.

Btw, it's "Tejas".

 

[stevejst]

Thermal dissipation is the issue but hey Intel engineers have been proving themselves worthy of challenges. We had gate clocking and thermal throttling so it is to be expected that this processor will eventually have some type of thermal control perhaps altogether a new one. As far as cooling goes it has been bigger and better all the way. Check that heatsink you had on Pentium II or AMD K5. What did you expect then?

 

[Mickey21]

Originally posted by: stevejst

(Doesn't Intel plan to hit 5ghz with these processors?)

Nope according to roadmaps published so far. This will go over 4.0 GHz when it will be succeeded by Teya which will be the one going for 5.0GHz.

But if you are looking for extreme cooling Pentium IV has broken the boundary of 5.0GHz already, using liquid nitrogen.

As Intel will probably wait till the last minute to point out though, that MHz is not everything. AMD already points this out, albeit sometimes less warranted in their "PR" schemes. As well as Apple showing that clock rate does not equal performance. Newer processors though starting at similair clock rates will no doubtely outperform their predecessors. Even though a P4 has reached 5 GHz, it was done very unconventionally and unpractical.

130W - 140W of heat dissipation is not unheard of, nor is it impossible to deal with. I admit it is going to be the highest in personal computing yet, but that's progress for you. It's also the upper range of standard usage. The processor surely will not try to dissipate that much heat at idle.

Besides, looks like my watercooling system has another upgrade still left in it...

 

[stevejst]

130W - 140W of heat dissipation is not unheard of, nor is it impossible to deal with. I admit it is going to be the highest in personal computing yet, but that's progress for you. It's also the upper range of standard usage. The processor surely will not try to dissipate that much heat at idle.

I see more blackouts coming to California.
And the President signing a law that outlaws computer game playing.

 

[jswjimmy]

i heard that it will also have a 1000fsb aswell as a 800

 

[Mickey21]

Originally posted by: jswjimmy
i heard that it will also have a 1000fsb aswell as a 800

People like me can run 1000FSB or more now, that is no big deal, but good nonetheless.

 

[stevejst]

Prescott suppose to have 800 MHz FSB. Teja will have 1066 and even higher later. But you never know.
Prescott brings in new instruction set as well. Depending on what AMD releases Intel will be waiting with their releases. In either case you'll see probably 3.4 GHz 800 FSB Prescott in the late October.

 

[RalfHutter]

Originally posted by: Wingznut

And when people talk about overclocking, do you realize that heat dissapation is not the roadblock? It's the gate leakage and elecromigration... And that's only going to become more of an issue (exponentially) with every process shrink.

There's been two or three brave souls raising their voices about this but they've been drowned out by the "OMG 103W! Chicken-Littles" out there.

Originally posted by: Wingznut
Are we done overreacting to a rumor now?

Unfortunately, probably not. People love to dogpile on this kind of stuff.

 

[THUGSROOK]

Originally posted by: Wingznut
First off... Let me go on record saying that I'm not going to confirm if this rumor is true or not, and that I'm not going to speak specifically about future products. But for the sake of this post, let's presume that it is true, and that the TDP is ~100w....

So what? Do you think that if PSC is ~13w more than the current Northwood, that is going to change ANYTHING? We're talking 13w here. How many watts do you suppose your overclocked/overvolted 3.5ghz P4's are dissapating?

And when people talk about overclocking, do you realize that heat dissapation is not the roadblock? It's the gate leakage and elecromigration... And that's only going to become more of an issue (exponentially) with every process shrink.

Are we done overreacting to a rumor now?

thx for the great posts in this thread

the reason we are concerned?
cause we see whats happening with northwoods @ 3400+mhz ~ heat and power consumption galore.
if its gonna get worse with prescott, we wont be able to OC them (much).

:disgust: