• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

WhiteWater on P4 with removed spreader?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
you'll probably see little/no difference in max overclock by removing the heatspreader, and with the heatspreader still on you can really go to town with the mounting pressure you can use on the block.
 
I would think that with or without the heat spreader you will get very good temps. I think Intel does a very good job with the installation of their heatspreaders so that they do not cause a serious thermal impedance.
 
Yeah, if you check out the latest snafu at [H]ardOCP (the article where they popped a heatspreader off a 3.06 PIV), you can see that Intel did one heck of a lapping job on the core to mate perfectly with the IHS. I agree with the other posters here, you'll probably see very little, if any difference at all.
 
I'm hearing a lot of "probablies" here.

I bring this up since WhiteWater mentions that one of the strengths of its block is the thin copper on the bottom. It seems to me that keeping the heatspreader on is just adding an additional layer of thermal paste and about 1.5mm of copper between the chip and the waterblock. These don't seem like terribly advantageous items.

I like that the heatspreader (which AMD will begin to employ soon) adds to the chip's durability and helps with all aluminum heatsinks, but it also strikes me as additional distance for the heat to travel.
 
PatrickBateman said:
I'm hearing a lot of "probablies" here.
Click to expand...

I'd say the reason you are hearing "probablies" is because nobody has done a before/after comparison. If you have the block and chip, maybe you should do both and post results. Your line of reasoning sounds good to me, but I've learned that what seems logical isn't neccesarily true in these areas. Thats why I'd be curious to see the results. Although Volenti might not have tried exactly what you're asking about, I know he has taken the heatspreaders off and cooled the die directly with water. With that in mind, I wouldn't take lightly what he has to say about it.

peace.
unloaded
 
I, for one, am extremely interested in hearing before/after results. I don't want to encourage you to damage your chip though. It is true that the extra heatspreader-thickness of the P4 does limit the full potential of what the WW has to offer but this is true of every waterblock on a P4.

Really, the experiment you're contemplating is one of "Will removing the heatspeader from the P4 help?". What cooling device is used is largely not relevant. The answer is "Yes". Volenti is the person with the most direct experience/answer to this.
 
Cathar said:
I, What cooling device is used is largely not relevant.
Click to expand...

Possibly is very relevant.
An edited repost of http://forums.procooling.com/vbb/showthread.php?s=&threadid=5416
Although the example is for a 10x10mm mm die with a 40x40x1mm IHS I suggest similar sums will apply to a P4 die and IHS
"Perhaps the considering as a "Finite Compound Spreading Resistance"* problem and comparing to the "Finite Isotropic" case** may give some insight.Maybe the "isoflux" is questionable but ....
An example of the "Compound" case for a 40x40mm heatsink with a 1mm Cu heatspreader
hsa.jpg

hsFilm.jpg

Which suggests the detrimental effect of the heatspreader increases the better the cooling"

* http://www.mhtl.uwaterloo.ca/old/on...rce/strip1.html
** [url}http://www.mhtl.uwaterloo.ca/old/on...rce/strip2.html[/url]
 
Last edited:
Ok, my personal experience with removing the heatspreader on a p4 resulted in an, at best, 3 degree difference, which gained me nothing overclocking wise (with waterblocks) when I went to direct die I gained some better stability but still no further mhz overclock.

That chip has since ceased to function, presumably from NSDS (running 1.8-1.85v for 9 months +)

In my experience the p4's don't respond to improved waterblock performance and more V-core in the way that athlons appear to, my current [email protected], 1.625v won't go a single FSB higher even with .2 of a volt more.
 
Volenti said:
Ok, my personal experience with removing the heatspreader on a p4 resulted in an, at best, 3 degree difference, which gained me nothing overclocking wise (with waterblocks) when I went to direct die I gained some better stability but still no further mhz overclock.

That chip has since ceased to function, presumably from NSDS (running 1.8-1.85v for 9 months +)

In my experience the p4's don't respond to improved waterblock performance and more V-core in the way that athlons appear to, my current [email protected], 1.625v won't go a single FSB higher even with .2 of a volt more.
Click to expand...

Hmmm, I'm wondering if it wasn't something besides the cpu that wasn't allowing for more of an o/c? I guess once you hit the wall with FSB you'd need an unlocked engineering sample to really tell if there's improvements from lower temps on the cpu. I've never owned an Intel chip, so I'm prolly wrong basing my ideas on AMD experiences.

peace.
unloaded
 
Thanks Les. I guess that explains why the heat-spreader is probably quite an okay thing for Intel to do with air-cooling heatsinks, and perhaps there is a definite advantage to ripping it off for water-cooling.

Unloaded/Volenti, P4's are strange beasts for overclocking. It seems that with AMD CPU's, the better the cooling, the higher the overclock in a fairly consistant manner, whereas for P4's, some react that way, while other seem to do it in steps, with no benefit seen if the cooling drops a few degrees, but all of a sudden another leap in overclock is gained, and then they stop again. Very strange. It seems to be a real mixed bag.
 
Cathar said:

Unloaded/Volenti, P4's are strange beasts for overclocking. It seems that with AMD CPU's, the better the cooling, the higher the overclock in a fairly consistant manner, whereas for P4's, some react that way, while other seem to do it in steps, with no benefit seen if the cooling drops a few degrees, but all of a sudden another leap in overclock is gained, and then they stop again. Very strange. It seems to be a real mixed bag.
Click to expand...

This is because of the smaller micron process their using compared to the current AMD cpu. I'v read up on this but I lack memory to fully explain it. Unfortunately I think we'll see the same with AMD with their next chip. I wish they would stop now with the small process their using and find other ways of speeding it up. Can only go so small before it's not of any benefit.
 
SysCrusher said:


This is because of the smaller micron process their using compared to the current AMD cpu. I'v read up on this but I lack memory to fully explain it.
Click to expand...

If I send you a stick of DDR would you be able to fully explain it?

:D
 
Since87 said:


If I send you a stick of DDR would you be able to fully explain it?

:D
Click to expand...

hmmm You might have a point. Today's DDR technically can't utilize the full bandwidth of the intel?
 
You must log in or register to reply here.

Similar threads

Niku-Sama
old school, K6-2+ cache unlock?
Replies
12
Views
783
Niku-Sama
Niku-Sama
W
FEATURED One more bad experience with EKWB products
Replies
9
Views
696
Woomack
W
Nebulous
Games drive & PrimoCache
Replies
12
Views
667
Mpegger
Mpegger
don256us
Two teams in our sights. Two!
Replies
8
Views
282
Norcalsteve
Norcalsteve
Railgun
TRX50 blocks
Replies
7
Views
375
Railgun
Railgun
Back