Good review but I don't see the point. I saw the point with a 980X stock cooler which is a cooler comparable to other high-end solutions.
Everyone knows that stock coolers from low-end Intels cpu's and AMDs suck big time.

Thanks for the review. I guess I won't be using the AMD stock cooler for my next build.

Thank you. The review reflects my view of Intel & AMD stock cooler: AMD is quieter than Intel, and 65nm CPU's stock cooler is better than 45nm CPU's counterpart in terms of cooling efficiency across the board.

I've been using AMD stock heatsinks for years even though I'm an avid SPCR regular. Speedfan almost makes it a non issue for me.

The point of a review like this is to know what value heatsinks actually make the situation better instead of being about the same or worse. The worst thing in the world of heatsinks is a fried CPU, the next worse thing is buying a heatsink, spending the time and effort to install it, and not getting your moneys worth.

Now the real kicker is most AMD CPUs come with a heatsink that has no heat pipes. Especially true when you buy on the budget end of the price spectrum.

So the AMD stock heatsinks I'm using right now on multiple systems in my house are worse than both of the two in this review. They have 70mm fans, are solid aluminum, no copper base, no heat pipes.

Without a review like this it'd be real easy for someone with a non heatpipe stock AMD heatsink to think it might be worth paying $10 or $15 to get the latest AMD heatpipe version. This review gives enough information to know instead of guess which one is worth spending money on.

Haha, only us Canadians understand bear skin rugs.

BTW, it’s FOUR heat pipes, not two.



//

A correction:
The coolers use four U-shaped heatpipes, not two, as can be easily spotted in the pictures by realising that there are four pipe-ends on each side.

An addition:
The stock cooler that came with my Athlon64 X2 4600+ for socket 939 is just about identical to the older design tested (might be a different fan), and mine is definitely made by Foxconn (as is the stock cooler for my previous Athlon64 3200+). (The name "Foxconn" is printed on the coolers.)

Been there, done that!
Replaced the all aluminium stock cooler of an Athlon XP 1800+ (Palomino) with some Thermalright(?) copper core heatsink and larger fan to reduce the noise. Made no apparent difference in the noise level, although the temps might have improved.

Replacing the Athlon64 X2 cooler with a Scythe Ninja 2 made a huge difference though, far more than I expected given that the case I used at the time had plenty of noise dampening and noise traps.

I think these stock coolers could benefit from a lapping, since the contact surface is very rough, so that would be my suggested micro budget alternative to buying a $<20 third party heatsink.

Cheers
Olle

Why didn't you test the heatsinks with reference fans installed?

SPCR doesn't have reference 70mm fans, last I checked.

Measuring performance in °C Rise/dBA favors larger & heavier heatsinks. But if you evaluate performance by weight & volume, °C Rise/weight*volume or °C Rise/$cost, then the AMD cooler is the record-holder.

Here is a review, the AMD cooler is an AVC Z7U7414001
http://www.frostytech.com/articleview.c ... 962&page=5


And 2 years later, it still ranks near the top of the list of all-time best CPU coolers, about equal in performance with the Thermalright Ultra 120 Extreme and the Tuniq Tower 120.
http://www.frostytech.com/articleview.c ... 244&page=4


If AMD had designed a funnel adapter for a 120mm fan, this would be the best top-down cooler ever in its weight class.

Oh, I meant the way the Intel stock cooler was tested with the 120mm Nexus fan in LGA775 heatsink roundup: http://www.silentpcreview.com/article976-page3.html

It would have been nice to see how the modded AMD heatsinks performed compared to ones with the stock coolers.

Thanks for the review. It was very informative.

I do believe there is a mistake made in the review. I have both of the reviewed coolers.

The fan speed of the two reviewed heatsinks need to be reversed. The Phenom II AM2+ Stock Cooler max rpm is 3300 rpm. I have the same heatsink that came with an Athlon 64 X2 6000+ that I have been using for over a year. The heatsink is made by Foxconn and the fan is either made by Delta, Nidec, AVC, or possibly more.

Just two days ago I bought a Phenom II AM3 955 with the latest C3 stepping, and it has the lastest AVC heatsink. I hammered on it fairly hard with Prime95, and x264 encoding overclocked to 3400, and the fan max rpm was 4000.

Anyway, I just don't see where the reviewer got the numbers that were posted unless, in fact, they were reversed by mistake.

Before I read this review, and during the installation of the AMD 955 BE just a couple of days ago, I compared the two heatsinks. And right off the bat I could tell that the older Foxconn was heavier, and I thought it would do a better job at cooling. I would like to point out to the reviewer that part of the reason that the latest cooler is lighter is due, in fact, to the copper heat plate being thinner. I am sure the point was noticed during the review but was not mentioned.

The Foxconn heatsink also has the fan embedded into the heatsink, and it is enclosed by a shroud. In my opinion, in the way it is designed, it would loose less pressure than the latest AVC heatsink with the larger fan.

Anyway, I am disappointed in the fact that AMD had to go the route of making the heatsink lighter. 80 grams of copper and aluminum per unit does add up fast.

Actually there is no spring, its a cam mechanism that provides tension on the retention cross clip.

And I see no point in testing heatsinks rated for 125W+ chips with a 95W TDP X4 620. The 95W chips ship with the aluminum heatsinks, and the 125W come with the heat pipe heatsinks, and should remain so if you want to test it under "stock" conditions. That changes the results significantly.

Considering how these coolers did with the 95W TDP part, why would you care how they do with the 125W part?

How do they determine that max RPM?
The data says 3800 for the Am3 - and just yesterday while I was playing mass effect, I took off my headset and wondered if my girlfriend had left her laptop running since there was a high fan noise - then I saw that the RPM measurements from PC Probe II was showing 4800 RPM for the CPU fan :-/

and yes, I have afterwards installed a Nexus intake fan and I'm looking for a new CPU cooler

Thanks, this justifies what I've been doing with AMD and CoolerMaster blow-down coolers for years- removing the ridiculous 50, 60 and 70mm 4-5,000 RPM fans and replacing them with 80mm case fans.

It may not have been high tech or near-silent, but the bang-for-buck decrease in noise was phenomenal.

-HF

I've witnessed no less than 20 models of AMD stock coolers and 50 models of Intel coolers. They just didn't care the detail of coolers if cooler's dimension was within spec. To complicate the matters, they often switched to different fan/cooler manufactures in different batches, making a fair comparison is not feasible. For example, I've two 4850e which bundle 2 slightly different coolers. To make matters worse, they did shrink cooler's size when migrating from 65nm to 45nm which is ridiculous to me.

And yes, I've witnessed 6000RPM LGA775 cooler and 5600RPM AM2 cooler, so your 4800RPM isn't that strange to me.

This must be cost cutting in action. Considering how little $ they get for their top CPUs, this isn't surprising I guess.

I've found the stock coolers from both AMD and Intel to get the job done fairly quietly thanks to today's mobos all having automatic fan control and the low idle power of modern processors. It's not like the old Athlon/Duron days when OEM coolers were frequently running a 60mm at a constant ~4500 RPM (eek).

Of course, I'm more picky when it's my home system sitting in a quiet room.

In my case, the "newer" fan could spin up to 5,000-6,000 RPMs easily under load during the hotter summer times. Anything above 1,400 RPMs was fairly audible... to say the least. Every time my processor was hard at work... I wanted to kill myself, literally. Unbearable pain, it was.

It is now cooling a 45W processor as we speak.. and doing so quietly. No idea, how AMD could possibly come up with such a dreadful design for their high-wattage flagship models.

There must be a way to buy a high-end processor in retail, minus the mediocre cooling system. If AMD want to be true to its "economical strategy", then that would make the first logical step. People end up buying aftermarket heatsinks, spending more $$$.

There are 10 editions of Windows... but I can't buy a processor with MY choice of cooler. To be this long in business and still have not realised that people may want this option... is beyond me.
I rest my case.

I wanted to buy either one of these on eBay for my HTPC, running an Athlon 64 X2 5600+ with it's all-aluminum stock cooler that was actually nice and quiet (it has the curved blades like the "old school" heatpipe type cooler in the review), but had idle temps (due to the cramped case) in the 48-50 C and max temps in the 65C range. I was hoping to maintain the sound level but drop temps a few degrees since these only go for $20-$30 on eBay.

Based on this article I bought the one on the right, the "newer" model with the more square blades since it appeared to cool slightly better with lower sound levels.

What I got looks *exactly* as the picture in this review, however, performance wise it's *nothing* like it!

The max speed on the one I got is over 6000 rpm, and it sounds like a bloody harrier jump jet taking off under load, and like a normal jet spinning up under idle. The best I could do was to get it to idle at 52-54 degrees with about 3000 RPM, and 68 degrees at max load and 6000+ RPM. And honestly, you can hear this fan screaming through the whole house, even on the second floor when the machine is in the basement!

I was about to put my old cooler back on, but I wondered if the fan was to blame - I took the fan off the old stock aluminum block and compared it to the one from the newer heatpipe model, and surprisingly, other than the fan blade design, they are virtually identical (thickness and other dimensions, even the flange where the tiny plastic clips hold it on).

So before pulling the new cooler off, I simply swapped out the fans, and it made a HUGE difference.

The new heatpipe cooler with the older curved blade fan from the original aluminum cooler is as silent as it was before, but my idle temps are now in the 38-40 C range and the max temps are in the 58C range. I'm very happy with my Frankensteined version of this cooler now, but when I first installed it out of the box, boy was I disappointed!

Which makes me wonder why AMD would use such a crappy fan??? With the right fan, which they obviously had access to before (and both are 4-wire PWM fans), this is a great cooler.

The article should be updated to say that this particular cooler can be a crap shoot due to the variability in fans used, and that you might end up with something that sounds like a jet taking off at idle.

Cheers
The REAL Joe