• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Here are the Core measurement between Palomino and Thoroughbred

N

NicColt

Diamond Member
Link to website

I don't know what language this is in but you can clearly see the measurements. The important thing is that the height of the core remains at 0.8mm for the t-bred

sorry if this was already posted.
 
no, hte important thing is we still don't know hte Vcore of the tbred. this site is claiming 1.60V.

either way, you're still looking at a massive contact surface loss, with potential for very high heat output.



MIke
 


<< no, hte important thing is we still don't know hte Vcore of the tbred. this site is claiming 1.60V.

either way, you're still looking at a massive contact surface loss, with potential for very high heat output.



MIke >>



You got that right, Mike! Pulling the heat off a mere 80mm2 core would not be fun. Why can't AMD put a darn heat spreader??
 
Maybe someone can explain how a heat spreader increases the cooling surface. I have taken the heat spreader off of a Celeron 1.0a and it is nothing but an copper plate attached flat on the core itself. The contact area is exactly the same as if a heatsink was sitting on it. How can this help, it just seems like another layer of material (it had some sort of thermal grease) will lower the cooling efficiency when you then slap on some more thermal grease and then a heatsink.
 
the IHS' primary role is for CPU core protection, secondary role is to increase contact surface, allowing for easier cooling for the heatsink.

I've never really pushed amd for IHS implementation on t-birds/xps for one reason: They probably run too hot for an IHS to be effective. It probably provides a barrier preventing effective cooling. Wiht p4s, this isn't really the case because the core is significantly larger to start. With .13mu Celerons and P3s, the heat output is low enough, that despite a smaller-contact surface, and IHS works fine.

But when you're talking about 60-70W on an 80sq mm surface, an IHS may not be a good idea. This does push the heatsink-requirement up though, as you need a high quality heatsink base (preferably made out of copper).



Mike
 


<< I've never really pushed amd for IHS implementation on t-birds/xps for one reason: They probably run too hot for an IHS to be effective. It probably provides a barrier preventing effective cooling. Wiht p4s, this isn't really the case because the core is significantly larger to start.
>>



I thought the Northwood core is almost the same size as the Athlon XP. You may be right about the spreader being inefficient for cooling, though. I am a bit confused. With faster and hotter P4s coming, do you think Intel will have to discard the IHS again? I imagine that a 3 Ghz P4 will put out a lot of heat.
 
Well, I don't see Intel dropping the IHS anytime soon. I don't see a problem with p4 chips and the IHS even at 3+ghz.

There may be some hindrance to cooling, but it doesn't stop people from getting incredible overclocks.

As far as a tbred goes, it may or may not be too high of a hindrance. its really hard to tell, because you're talking a large amount of heat on a miniscule contact surface, so an IHS may or may not help.


Mike
 
like has been said....it wont help...its just a piece of copper on the surface......i would get better results from a swiftech 462....just gotta be careful of the core...like has been said.....lol
 
Thanks. I didn't notice the bobo in the link.

I have not put very many links in posts on the anandtech forums.
 
From Text

1900+ 1600MHz 1,65 V 61.8 W 90°C
2000+ 1667MHz 1,65 V 63.8 W 95°C
2100+ 1733MHz 1,65 V 65.9 W 95°C
2200+ 1800MHz 1,65 V 67.9 W 95°C
2300+ 1867MHz 1,65 V 70.0 W 100°C
2400+ 1933MHz 1,65 V 72.0 W 100°C
2500+ 2000MHz 1,65 V 74.1 W 100°C

My guess is AMD will start using heatsinks with a copper base.
 
thanks. page 23 but 35 of the pdf d'oh...

and ouch... I didn't think the t-breads would be that much higher. but ouch... overclocking yes but serious cooling indeed.
 
True 'bout the page numbers, sorry/d'oh! lol...

Anyway - yes, the Tbreds look like they may be a bit of a biatch to cool. The power outputs look reasonable for the clock speed, but with the die shrink the surface area is much lower. So, power density is up => more demanding to cool. Just got an Alpha 8045 for my XP - hoping that this will do the job for Tbreds OK. Better still - hoping that Barton doesn't fall off the roadmap. Apparently the SOI should reduce inefficiencies and lower power output again.

Waiting on those reviews, guys...! NDA? It's only a bit of paper... 😉 (guessing the sites won't get their chips for a month or two... whaddya reckon?)
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top