Sign in with
Sign up | Sign in
Your question

which would you prefer?why?

Last response: in Overclocking
Share
July 5, 2007 1:51:36 AM

1. enzotech ultra-x
2. thermaltake big typhoon vx
3. Scythe Samurai Master CPU Cooler

:? 8) :? 8) :? 8) :? 8) :? 8)

More about : prefer

July 5, 2007 1:29:54 PM

Which of those I'd prefer, would depend on what my needs/criteria/priorities were for my particular system configuration. But since you didn't list any of that, here's some general notes...

First, depsite several other sites showing conflicting (if not the complete opposite) results, AnandTech is of the stated opinion that "side-blower" heatsinks outperform "top-blower" heatsinks. My best guess is that this has something to do with differences in test setups (the sites showing top-blowers outperforming sde-blowers all seem to have cases with side vents/fans and/or otherwise good airflows... AT has no side vent and turns off exhaust case fans for reviews). So, I don't know what the real answer is, but it's worth mentioning.

Having said that, the 3 you list are all top-blowers. You also appear to have listed them in what I would consider to be performance-order (the Scythe Andy being the lowest performance of that bunch), and possibly pricing order as well.

The best comparison data-point I've seen for those 3 coolers would be this article. At least in that test, the Ultra-X outperforms the other two by a decent margin (5-6C inside a case with a side fan, at reasonable noise fan speeds (~1300rpm), on a E6400 overclocked to 3.5Ghz @ 1.45v). The other two are probably neck-n-neck, given the difference in fan speeds.

The Ultra-X is the heaviest of the 3 (not by much), weighing in @ 835g (vs 822g for the VX and only 685g for the Andy).

The Ultra-X has the best mounting mechanism - the other two (particularly the VX) are getting on the heavy side of things for the Intel push-pin system IMO.

The Ultra-X has the shiniest lapping job and/or build-quality.

The Ultra-X has a PCI-slot fan-speed controller - the VX has a controller, but it's mounted on the heatsink (would have to open the case to get to it). I don't believe the Andy comes with one.

So my personal preference, based on my own criteria and system configuration would be:

1. Ultra-X
2. Andy (lighter weight on push-pins, with virtually same performance as VX)
3. VX
July 5, 2007 3:46:31 PM

I just want to clarify, you noted that Anandtech said side blowing heatsinks are better than top blowing heatsinks. This is not entirely true.

Anandtech said Heatpipe Towers are the best heatsinks, a result of this is the fan is almost always blowing sideways and not down. You'll notice that most if not all down blowing heatsinks have less heatpipes and therefore perform worse.

So just because it is side blowing doesn't mean much, if its a heatpipe tower, then you're almost guaranteed that its going to be, on average, better performing than other down blowing heatsinks.
Related resources
July 5, 2007 6:20:48 PM

I was paraphrasing various (and multiple) quotes, such as...

Quote:
The heatpipe towers with side-facing fans are the top-performing coolers we have tested, in both overclocking ability and cooling efficiency. The best designs with down-facing fans perform in a second tier category just below the top heatpipe towers.


...the reason I shortened it to "side-blowers" is probably because thier charts also have the Zalman 9700 out-performing any top-blower.

I just wanted to point out that thier charts/results look nothing like the ones you find at some other sites, like FrostyTech, X-bit labs or Hexus.

Are those other reviews just getting it wrong/reversed? Is AT just getting it wrong? No, I think the answer has something to do with thier differences in testing methods. And I think AT (as a matter of professionalism, if not pure curiosity) should figure out what that difference is, before making such conclusions.
July 5, 2007 7:58:31 PM

They say the best down blowing coolers are on a second tier to the top heatpipe coolers. That is exactly in line with what Frostytech came up with. I have no clue what other towers Frostytech has tested but they rank the Ultra 120 better than the Enzotech. Now the Ultra 120 E would be even better than that.

Also i would question what how many quality heatsinks Frostytech have tested cause they have that Alpha Novetech PRE9060 for #10 on its top 10 intel heatsinks. It doesn't even have any heatpipes and the fins are made of AL.

I'm not sure about Frostytech's standard way of measuring the heatsinks, but i do know that Anandtech has used the same cpu, mobo, memory and software since day one with the Tuniq Tower 120.

It's been pretty much concluded that the Thermalright Ultra 120 Extreme is the best air cooler available to date. A quick look at Xtremesystems forums reveals that is the air cooler of choice. They are a no bs place where performance only matters if they thought there was something better they would be all over it. I still think there is a lot of questions to be answered from hexus, frosty , x-bit labs, and anandtech as to how the performance results differ so much.


Also i just noticed this but the Hexus.net writer/editor guys must be smoking crack because they rate the Thermaltake Big Typ VX number one on their charts as the best cooler regardless of money.

Here

Then they say that for the £20-£35 Segment the TT Big Typ VX is the best cooler,

but for the £35+ Segment they say the Zalman 9700 & Titan Amanda are the best coolers?! What on earth were they thinking? Cooler's that are more expensive and perform worse are rated over the better alternative?

I mean do they think that someone is just looking to spend a dollar amount in a specific range regardless of performance in other price ranges! Talk about not thinking!
July 6, 2007 8:35:10 PM

I don't really mean to (or want to) nit-pick the specifics or belabor the point to much, because I'm mostly talking about a matter of generalities and assumptions/statements being made, based on what I consider incomplete dataset (and I think we both agree on that, to some extent) But to help illustrate my point...

Quote:
They say the best down blowing coolers are on a second tier to the top heatpipe coolers. That is exactly in line with what Frostytech came up with. I have no clue what other towers Frostytech has tested but they rank the Ultra 120 better than the Enzotech. Now the Ultra 120 E would be even better than that.


Yes, Frosty has the Ultra 120 cooling better (at the 150W level, at least - the Enzo is the best performance they've measured to date at the 85W level), but it's a very small margin between the Ultra 120, Zalman 9700 and Enzo. Now look at the margin in the AT charts - it's huge (or at least not proportional, as you might expect).

As for 'top' heatpipe coolers, in addition to the previous X-bit Labs link, copy this link and paste it into google's translation page, and set the language from Russian to English... note the results in that test (once again in a case with side vents/fans).

The first thing you might notice is that they were using a slow (reasonable noise level) fan on the eXtreme, but ignore that result for the moment and look at the results of the Enzo on low speed, vs the eXtreme with one or two fans... I'd call that a 'top' heatpipe cooler in my book :) .

Quote:
Also i would question what how many quality heatsinks Frostytech have tested cause they have that Alpha Novetech PRE9060 for #10 on its top 10 intel heatsinks. It doesn't even have any heatpipes and the fins are made of AL.

I'm not sure about Frostytech's standard way of measuring the heatsinks, but i do know that Anandtech has used the same cpu, mobo, memory and software since day one with the Tuniq Tower 120.


Yeah, but they haven't updated thier 'top 10' page in a while... and have recently reviewed a plethora of new heatsinks, so ignore the top 10 page for now. As I've mentioned before, I think FrostyTech is an interesting data-point (just as I regard every other heatsink review/site)... thier synthetic heat-source testing methodology has both good and bad points...

Good:
- it's completely repeatable over long periods of time
- it completely removes the case, air-flow, cpu and even "app to generate load" from the equation.

Bad:
- it completely removes the case, air-flow, cpu and even "app to generate load" from the equation. :) 

...I'm also unclear on and would like to hear more about how they manage the various mounting mechanisms (so that we get a true representation of mounting pressure).


Quote:
It's been pretty much concluded that the Thermalright Ultra 120 Extreme is the best air cooler available to date. A quick look at Xtremesystems forums reveals that is the air cooler of choice. They are a no bs place where performance only matters if they thought there was something better they would be all over it.


...and if you go look around there, I think you might be surprised how much they are "all over it". It's highly regarded by many there (Edit: the Enzo, that is. And again, I'm not claiming - and don't think - that the Enzo can cool better than the eXtreme, but we're also talking about ~$80 worth of cooler+fan that has a shitty mounting system and needs to be lapped vs a $60 plug-n-play solution that even comes with a fan controller and AS5... this makes it "worth looking into", depending on your requirements and configuration).


Quote:
I still think there is a lot of questions to be answered from hexus, frosty , x-bit labs, and anandtech as to how the performance results differ so much.


That's my only real point. In many/most instances, a lot of the differences can be accounted for, if you look closely enough at the tests being run, fans in use and other variables, but this top-blower vs side-blower issue seems to show very contrasting results. Since AT is the only one drawing conclusions and stating opinions about the subject, it seems to me that they'd want to figure out 'why' thier setup is so unkind to top-blowers, relative to other sites before making such claims.

Should I just assume that everyone else is just getting bad results and AnandTech is the only reliable source of information on the subject? Should I regard AT's charts as the gospel truth on heatsink performance? Personally, I'm not comfortable answering yes to either of those questions. Again, I'm not trying to knock AT and I do believe they are seeing the results they are reporting, my only concern is with them drawing conclusions (and due to thier status/popularity, influencing readers towards those same conclusions), when there's conflicting results that need to be accounted for.
!