Intel's boxed coolers...now with 20% more suck factor?

[joefriday]

I've noticed a small change to Intel's half-height heatsinks. Can you spot the difference?

New:

Old:

I'm throwing the question out there to the Forums readers who've bought recent Intel boxed Penryn and Wolfdale CPUs: Anyone get one of these new coolers with their boxed CPU? Has anyone tried it out? The 1st pic is a shot of the cooler that came with an engineering sample e8500. I noticed this same style of cooler in a review of a retail boxed Celeron e1200. Notice how there is no polished bottom, and no chromed spring-like apparatus securing the push pins. The push pins are now part of the fan housing.

 

[dragonsprayer]

ya the copper ones are alot better
but if do not oc these work

 

[badge]

I installed the new version of the 775 HSF on an old Pentium 4 3.0 GHz 915/775. The chip was not particularily hot by nature, but no problem so far. The kid has not brought it back.

 

[dagger]

Lol, who uses those? They're one third the size of stock coolers that comes with old P4. No one in their right mind would slap those on their new expensive cpu. Feel the burn...

 

[badge]

Well, I want to protect my investment with a great performing fan, but the new 45nm processors do run a lot cooler than previous releases. I see where some guys are using the larger heatsink like the 120 and NO fan!

 

[joefriday]

ya the copper ones are alot better
but if do not oc these work

errm...ya...I don't think you spotted the difference...both of these coolers pictured are half-height, both aluminum cored. Of course the copper cored coolers work better than the aluminum versions, but I'm comparing the current, lowest-of-the-lowend intel box cooler to this even cheaper looking unit (i just changed the old cooler pic, to hopefully make it more clear). Just when you thought the box coolers couldn't get any cheaper, Intel found a way to cut another corner.

Lol, who uses those? They're one third the size of stock coolers that comes with old P4. No one in their right mind would slap those on their new expensive cpu. Feel the burn...

Well, Intel puts them in with the e8500 cpu. Apparently they think it can handle the heat at stock speeds.

 

[badge]

Intel puts them in the Q9450!

 

[badge]

You know with the 45nm units if you have superb case air flow along with a close up strong exhaust fan that is near a large heatsink like the TR 120 (or even the smaller version), and you are not overclocking, you can run without a fan. I guess you realize this. But hey yeah, what an El cheapo HSF to provide with such great CPU's.

 

[jaguarskx]

Meh, I always stick an after market HSF on my CPU.

 

[badge]

THe newer, smaller stock units sound like a toy. You know a wooden airplane with a rubber band driven propeller? Like that when they first start up.

 

[dragonsprayer]

the stock intel cooler with a copper core is great for 3.4ghz dual core systems - i used them all the time with out issues
http://s63.photobucket.com/albums/h138/4rothrocks/?action=view&current=DSC_0095.jpg

i agree that a better cooler may extend the life of the cpu 5% or so

 

[Mondoman]

Not sure why you think the new coolers perform worse. As long as there aren't any more problems with good even pressure over the middle for good contact with the heat spreader, I would think that the new coolers are better. Since the idea is to maximize heat transfer from the CPU to the cooler's fins, one less additional plate between the CPU and fins eliminates one interface where heat conduction might not be optimal. My guess is that the chrome-plated metal plate in the old cooler also doesn't conduct heat quite as well as the aluminum of the current one.

 

[pcgamer12]

At stock they perform well. My E4600 came with the 1st cooler, but I was skeptic. It keeps my processor at 23-24C typing this on FireFox (BTW pledge for Firefox 3, details here, they trying to set a Guinness World Record) http://www.spreadfirefox.com/en-US/worldrecord/ And on Prime95 torture test for max heat I maxes out at 45C. But once I overclock it to 3.0GHZ by raising the FSB to 250 from 200, temps skyrocket to 65C+ load. To conclude, these intel stock coolers are good for stock speeds and nothing else. If you OC then you need a 3rd party HSF. I cant wait to get a ACF7P to overclock using auto settings, I dont want to waste time trying to find the lowest stable voltage on my cpu to bring heat down on this crippled HSF.

 

[pcgamer12]

oh, i use AS5 if that matters.

 

[joefriday]

Not sure why you think the new coolers perform worse. As long as there aren't any more problems with good even pressure over the middle for good contact with the heat spreader, I would think that the new coolers are better.

The reason I believe it to be less effective is the rough mating surface this new style presents to the CPU.

one less additional plate between the CPU and fins eliminates one interface where heat conduction might not be optimal. My guess is that the chrome-plated metal plate in the old cooler also doesn't conduct heat quite as well as the aluminum of the current one.

I think I need to clarify this. That chrome bracket is not a thermal barrier, at least not in the sense that you imply. It has a punched out center, and is interference-fitted around that highly polished aluminum slug which lies under the thermal paste.

That cylindrical aluminum slug is driven all the way through the heatsink as you can see in the following pic displaying the top of one of these coolers with the fan removed:

In the new version, not only is the bracket gone, that slug of aluminum is also gone. The cooler is now a single chunk of aluminum, allowing Intel (or whoever makes these heatsinks) to forgo the time consuming and more expensive process of the performing the interference fit. I'm not necessarily saddened by any of this, as like you said, a single chunk of aluminum in theory should be a better thermal conductor than two pieces of aluminum stuck together, even if the two pieces are sandwiched together as tightly as this slug to its radial fin housing. What does erk me though is that the bottom of the heatsink is no longer a nice, smooth, polished surface. It simply looks crude. Some home-brewed heatsink lapping of these new versions might prove to be highly beneficial.

One positive aspect I see out of this new version? Without the steel plate in the way, the fan should be able to push air more effectively through the heatsink, as the passing air no longer gets coldcocked by that steel plate blocking the underside exit of approximately a quarter of the heatsink's finned area, theoretically allowing more airflow, less turbulence and quieter operation. This may make the new version more efficient than the old, despite the lack of a smooth mating surface. although, without testing the two , we won't really know.

There is also another possible negative of this new style cooler. Those housings encasing the push pins in the plastic fan look like a potential point of failure, especially during installation, where one has to typically push the poo out of those pins to get them to snap down securely. I can imagine some ham-fisted knuckle dragger breaking them clean off the fan because he pushed on 'em too hard.

One thing is for sure: this new version is a much simpler, cheaper to make design, and almost assuredly a bit lighter, saving Intel even more money on freight costs.

In case you were wondering, the title of the thread is more for a bit of sensationalism than about how I actually feel about the redesign. I find that most of the kiddies in the forum are much more likely to read and respond to a thread when the title is intriguing, rather than when the titles are the dull, matter-of-fact headings I normally make for my threads, and I really wanted to get some interaction and discussion of ideas about this topic. It's also a bit of a pun. If the removal of that steel plate actually does improve airflow, then perhaps Intel's boxed coolers now really do have 20% more "suck factor" . I thank you all for your thoughts and comments (well, except Dragonsprayer, who was kinda off in his own conversation). It was nice to have a thread for once that didn't revolve around "OMG my CPU is on fire!!!11!!", "Check my built", "CPU running slow" or the ever-popular "e8400 vs Q6600". Thanks guys.

Joe

 

[Reynod]

The old P4 coolers for the last of the Netburst line were very efficient coolers ... much better than these newer el cheapo crap.

Why?

Because the older coolers had to dissipate more heat ... consistant heat.

The core2 line idle much lower, work harder for less power.

If you can get an old 3.2Ghz or higher P4 cooler and it fits your less likely to be cooking that overclocked beast.

Can someone tell me which of the older coolers fit the new mobos?

It's of interest I'd say.

 

[mi1ez]

By your logic, my Pentium D cooler must be a beast!

 

[Scotteq]

I guess it's hard to make the leap in logic that since the new CPUs run cooler then the HSF can be downsized. That the stock heat sink remains adequate for stock usage, and therefore you still need an aftermarket one for overclocking.


So... What changed, exactly???

 

[dagger]

By your logic, my Pentium D cooler must be a beast!

Wow, that's one useless comment. But you know the sad thing about it is? You're actually right. It actually perform better than the new q9450 cooler. Look it up.

As for newer cpus running cooler, well, that's not true. It actually runs a lot hotter than old p4 for someone who actually use it and not let it idle like most people. Idle heat output is lower, because it's more efficient, but if you keep all 4 cores at 100% load with the stock cooler, as in some encoding tasks, even on stock clock, the chip will overheat on you, guaranteed.

 

[amdfangirl]

Now, before I go off-topic I've noticed that the stock AMD coolers have been getting cheaper and cheaper. Now they don't have screws to hold the fan so its hard to replace them...

Back on topic: :/ anyone got any comparisons... efficiency can be improved alot so... (Besides P4s weren't that hot, right?)

 

[clay12340]

I just got a Q9450 last week(Not sure when it was manufactured) and I'm almost certain that it had a copper slug in the sink. I'm wanting to say it had the metal bracket holding the push-pins as well. I'll have to look at it tonight when I get home from work, I might be wrong on the bracket part.

It was pretty small, but it seemed like a reasonable retail cooler. I ran it for a couple of days making sure everything ran stable at stock speeds and I never saw any high temps. The chip stays very cool according to the bios and asus temp probe utilities.

 

[speedbird]

The cooler that came with my Q6600 was barely able to keep my CPU cool when overclocked. These Stock coolers are really only enough for CPU's that run at the default stock speeds. With my new aftermarket cooler installed my Temps have dropped dramatically when running at overclock.

In general The Intel cooler Push Pin design is probably the worst idea ever, it's fiddly and can be frustrating to install correctly. AMD have a far easier and quicker Heatsink design, just line up the Heatsink holes on the metal plate then simply push the levers down to secure. Intel should abolish this ridiculous push pin rubbish and follow AMD's method.

 

[joefriday]

The old P4 coolers for the last of the Netburst line were very efficient coolers ... much better than these newer el cheapo crap.

Why?

Because the older coolers had to dissipate more heat ... consistant heat.

The core2 line idle much lower, work harder for less power.

If you can get an old 3.2Ghz or higher P4 cooler and it fits your less likely to be cooking that overclocked beast.

Can someone tell me which of the older coolers fit the new mobos?

It's of interest I'd say.

Hey Reynod, the best factory cooler for LGA 775 is the new monster that comes with the QX9650.
http://blog.mpecsinc.ca/2008/02/intel-qx9650-quad-core-extreme-has-new.html
Barring that, according to THG's testing, the best cooler for LGA 775 that us mere mortals can attain would be those used on the P4 EE 3.73GHz and the Pentium 840/955/965 EE. They typicaly came with a high speed blue fan, had a much larger diameter copper slug than regular coolers, and had a dense collection non-bifurcating fins that curved counterclockwise, going against the fan's clockwise rotation. After that I think it's your typical Pentium D/Conroe/Pentium 4 coolers, followed by the Celeron D full height aluminum cored units, then I'm guessing probably the half-height coppers, and finally this sad bunch of coolers I've talkied about in this thread.
Here's a shot of QX9650 next to the old extreme edition cooler for Pentium D:

Source: overclockers.com

 

[clay12340]

I got home and looked at the cooler from my new Q9450 and it definitely has a copper slug, but I was wrong about the metal push pin bracket. The push pins connected to the fan bracket which seems pretty cheap. So I guess the coolers can vary even within the same model which seems kind of odd.

That QX9650 cooler is pretty impressive for a retail cooler. It looks like pretty similar to a lot of the orb coolers etc.

 

[dagger]

That's hardly a monster. It's still slightly smaller than stock cooler that comes with old P4. It still has the circular base, reducing contact area. *not impressed*

Who'd use that on a $1k QX cpu? Those things run a lot hotter than typical stock quads and are meant to oc.