THOR's HAMMER or PROLIMATECH MEGALEHAM

[CrazigXn]

These rae both about $60, but in some reviews the Megaleham wins by 3 degrees in a 3.6 i7 overclock, but Thor's Hammer looks more Enthusiast and uses HDT versus the Heat Plate.

Thoughts, opinions?

 

[Rubycon]

What are you cooling? HDT is a gimmick IMO.

 

[CrazigXn]

cooling an i7 920 that I plan to OC to 3.6 - 3.8

 

[Rubycon]

Definitely go with the Promlimatech. Performance > appearance.

 

[CrazigXn]

true enough... thanks.

 

[Special K]

You might also consider the Scythe Mugen 2. This review shows it beating the Megahalem:

link

 

[JEDIYoda]

Originally posted by: Rubycon
Definitely go with the Promlimatech. Performance > appearance.

who says?
performance is relative....a degree or two varies from computer to computer.
I have read review after review.....its a virtual toss up!

 

[Shmee]

what about a TRUE? mine seems to have very good temps at 3.6. about 43c load, usually less than 30 idle.

 

[Rubycon]

Originally posted by: JEDIYoda

who says?
performance is relative....a degree or two varies from computer to computer.
I have read review after review.....its a virtual toss up!

Sheer capacity. An overclocked i7 puts more thermal stress/demands on air cooling than any other previous release. A TRUE will also give excellent performance. Buy whatever is cheaper if cost is of concern.

 

[JEDIYoda]

Originally posted by: Rubycon

Originally posted by: JEDIYoda

who says?
performance is relative....a degree or two varies from computer to computer.
I have read review after review.....its a virtual toss up!

Sheer capacity. An overclocked i7 puts more thermal stress/demands on air cooling than any other previous release. A TRUE will also give excellent performance. Buy whatever is cheaper if cost is of concern.

cost has never been a concern, I have over 100 different heatsinks and counting.
Sheer capacity actually means little to nothing depending on how a person sets up the airflow in their PC....

 

[TC91]

hmmm so is there a consensus top cpu air cooler out there anymore?

 

[Rubycon]

Originally posted by: JEDIYoda

cost has never been a concern, I have over 100 different heatsinks and counting.
Sheer capacity actually means little to nothing depending on how a person sets up the airflow in their PC....

Sorry but this is incorrect. Once heatpipe capacity is reached increasing airflow will do little to bring temperatures down. Well let's just say (like most people) you're applying ambient air and a reasonable cooling - Delta 74dBA monsters don't count!

Capacity is EVERYTHING on top i7 class cooling otherwise it's water/refrigeration, etc.

Top air cooler currently is the TRUE. The Prolimatech is also a very good competitor with a proven mount and the results show. You cannot go wrong with either one and they both have very similar (if not identical) capacities.

 

[Special K]

Originally posted by: TC91
hmmm so is there a consensus top cpu air cooler out there anymore?

Based on the reviews I have read, these seem to be the top coolers (in no particular order):

Scythe Mugen 2
Prolimatech Megahalem
TRUE 120
Noctua NH-U12P
Thermalright IFX-14

The exact ranking of these depends on the particular review. I'm guessing you probably couldn't go wrong with any of them. Search for reviews on them to get a better idea for the specifics of each cooler.

 

[CrazigXn]

Only concerns with the TRUE is that in a few tests it did not perform well because the base of the TRUS is smaller then the i7 chip heat spreader... The folks above who mention using the TRUE, is this ture? (no pun intended.)

 

[Special K]

Originally posted by: CrazigXn
Only concerns with the TRUE is that in a few tests it did not perform well because the base of the TRUS is smaller then the i7 chip heat spreader... The folks above who mention using the TRUE, is this ture? (no pun intended.)

I have also read that:

1. The TRUE doesn't use a very good mounting system, and that the heatsink isn't really secure on top of the CPU

2. The base of the TRUE isn't very even, and to get the best performance from a TRUE, you really need to lap it

Perhaps someone who owns the TRUE can comment on these?

 

[Rubycon]

Originally posted by: Special K

Originally posted by: CrazigXn
Only concerns with the TRUE is that in a few tests it did not perform well because the base of the TRUS is smaller then the i7 chip heat spreader... The folks above who mention using the TRUE, is this ture? (no pun intended.)

I have also read that:

1. The TRUE doesn't use a very good mounting system, and that the heatsink isn't really secure on top of the CPU

2. The base of the TRUE isn't very even, and to get the best performance from a TRUE, you really need to lap it

Perhaps someone who owns the TRUE can comment on these?

The mounting mechanism is most definitely secure. It will not fall off, for example unlike pushpins that are not positively secure. Some may notice the heatsink may move if twisted but the pressure should be within specifications. Adding a washer between the bracket and the block and cutting down the bracket bosses so the screws bottom out at the mobo pcb is optional and a lot of users may do this to increase mounting pressure.

The base on some were a little more convex and that may have been intentional. I'd leave that up to TR's engineers.

 

[faxon]

the base on the TRUE is horrible, as is seen here http://benchmarkreviews.com/im...treme_Base_Closeup.jpg

it may be designed to put more pressure on the cpu in the center to make the solder under the IHS spread better and provide better contact since the IHS is usually dipped in in the center a bit, but the base is so horribly ground that you can audibly hear your fingernail grinding when you run it across the cooler base. there's maybe 15% actual metal on metal contact the way they finish the base of the TRUE. given, i may be a bit bias, since i kind of lapped one once and the results were nothing short of stunning, but it's something to consider

http://forums.anandtech.com/me...=2297013&enterthread=y

ed: for an update on that, my buddy currently has his 920 at 1.4v and a nice 3.8GHz OC, running like 55C load across the cores on avg with push/pull high CFM fans. in comparison, my S-1283 averages 55c at 1.36v on my e5200 at 3.2GHz, and the e5200 is only a dual core. gives you an idea of the performance delta between the 2 coolers

 

[zagood]

Originally posted by: faxon
ed: for an update on that, my buddy currently has his 920 at 1.4v and a nice 3.8GHz OC, running like 55C load across the cores on avg with push/pull high CFM fans. in comparison, my S-1283 averages 55c at 1.36v on my e5200 at 3.2GHz, and the e5200 is only a dual core. gives you an idea of the performance delta between the 2 coolers

To give you a different perspective, my e5200 @ 3.5ghz with a non-lapped TRUE idles at 37C/32C and maxes at 44C/44C. The base isn't that bad, but lapping can definitely help.

-z