SPCR

Thermalright XP120 is the first 120mm fan cpu cooler and it outperforms every other HS at just about any noise level, including the all-important super quiet level.

Nice review Ralf Hütter! Are you going to test the XP-90 soon too?

Edit: (I saw Mike C in the news announcement and didn't think any further...)
I think the XP-90 comes very close in performance, has anyone seen a side by side comparsion somewhere?

kudos should go to Ralf Hutter. XP-90: When we get a sample.

Excellent review as always, good work on this one Ralf. It's seems we have a new top performer and silent champ. Great news as I'm about to buy one to cool my upcoming Athlon 64.

Booyaka!

Chalk one up for Globe

My XP-90 sample is already on its way to me from Jab-Tech. I'll try to have the review available within 11 days, if ambient temps are reasonable.

-Ed

Nice work, Ralf. Now for some controversy. From a post at Hardware Central:
_________________________________
Quote:
"And while I'm no expert on heatpipes, my understanding was that they lead the heat _upwards_. They work sort of like a lava lamp. And if you look at any professionally made silent PC, such as Hush or Shuttle's cases or whatnot, their heatpipes always go upwards.

Now look at ThermalRight's stupidity in that pic: when the mobo is vertical, out of the 5 heatpipes on that heatsink, only 2 lead the heat out of the CPU. One doesn't do anything because it's horizontal, and two could at most lead the heat back to the CPU, if it was cooler than the heatsink. I.e., 3 out of 5 do nothing more than restrict airflow.

(Which still isn't the worst design. Others use only horizontal ones, or worse yet heatpipes bent all around, where basically at most they could lead the heat from the heatsink to the heatsink.)

Basically heatpipes have become a buzzword, like XML is to programming. Everyone has to have them, even in designs where they can't work.

Or where they work only as long as the motherboard is horizontal, which is how some of the [words deleted] reviewers measure. Whee! It cools so well with the mobo on the desk! Too bad it won't act the same in a tower case, eh?"
__________________________________

If you still have it mounted, would it be possible to put the test rig on its side and run one more test?

Good article. I am looking at an upgrade but I noticed that both the ASUS AMD and Intel boards I have my eye on are on the list of "yea it will work if you bend stuff a little" list. If that isn't enough to make me nervous the weight of the thing certainly is a factor for those of us who go to lan parties and bump our cases during transit.

It will be interesting to see if someone with a 3.2 Intel cpu does a test with a 5v fan. I'd like to see the temps in that situation.

I've thought of that too DanceMan, I'm really curious what the orientation of the heatsink does to performance. I guess the test setup Ralf Hütter is using is better than having the mobo vertical like in tower cases.

PretzelB wrote:I am looking at an upgrade but I noticed that both the ASUS AMD and Intel boards I have my eye on are on the list of "yea it will work if you bend stuff a little" list.

So, you're an ASUS fan? Maybe the XP-90 is something for you?

DanceMan --

It's a good point. I think RH will be interested enough to try a quick test run (just one fan) with the motherboard held vertical as in a tower case. Could still be done out of case, RH -- just screw the board down on the test rig and set the rig on its side.

DanceMan wrote:Nice work, Ralf. Now for some controversy. From a post at Hardware Central:
_________________________________
Quote:
"And while I'm no expert on heatpipes, my understanding was that they lead the heat _upwards_. They work sort of like a lava lamp. And if you look at any professionally made silent PC, such as Hush or Shuttle's cases or whatnot, their heatpipes always go upwards. [...]

I don't know which Hush case that guy at Hardware Central has opened, but the heatpipe in the Hush Mini-ITX on my desk goes horizontal. In any heatpipe design I have seen at least part of the heatpipe goes horizontal.

There are several places in these forums where heatpipes are discussed. But a quick summary:

1. The heatpipe works best when transferring heat upward.
2. The heatpipe works perfectly fine transferring heat horizontal.
3. The heatpipe can transfer heat downward. In the worst case it performs as a copper conductor.

Great review!

Ralf,

Is your globe fan thermally controlled? i have 2 globes fans and i find them very loud. Do by anychance have a nexus fan?

great review as always.

~RaNDoM

k I'm doing this as I'm reading -

There's the catch: This mounting system will not work with fans that have closed corner flanges like the Nexus or Globe fans, both top contenders for the coveted "Quietest 120mm fan" title. More about this later.

... GYAH!!!! NYUUU!! *cries in pain* Oh well. If I somehow decide to go back to air cooling with the XP120 i'll just whip out my Nexus fan and my dremel

I'd challenge any shipping company to damage this little puppy!

You WOULD or you WILL? Because I'm sure USPS could damage it, or UPS.

Note that the Globe fan cooled 4°C better blowing up than blowing down.

??? Where is this? The table above this quote on page 4 shows only 1 C difference.

Oooo my Mushkin RAM sticks are in one of those pics! I feel so lovingly popular.... even though my face/name isn't on there, ROFL...

DanceMan,

Mats,

MikeC,

From this this page on Thermalright's website:

Thermalright wrote: Our Heatpipe construction was also redesigned to eliminate any gravity effect and to unleash the capability of the XP-120's performance to operate at 100% in any direction you mount it.

Thermalright wrote:Our Heatpipe construction was also redesigned to eliminate any gravity effect and to unleash the capability of the XP-120's performance to operate at 100% in any direction you mount it.

Would it not be worthwhile to confirm?

acaurora wrote:

Note that the Globe fan cooled 4°C better blowing up than blowing down.

??? Where is this? The table above this quote on page 4 shows only 1 C difference.

Well, I learned math via the "New Math" back in the '60's so maybe things have changed since then, but 45 minus 41 equals 4 to me.

Ralf Hutter wrote:

acaurora wrote:

Note that the Globe fan cooled 4°C better blowing up than blowing down.

??? Where is this? The table above this quote on page 4 shows only 1 C difference.

Well, I learned math via the "New Math" back in the '60's so maybe things have changed since then, but 45 minus 41 equals 4 to me.

RaNDoMMAI wrote:Ralf,

Is your globe fan thermally controlled?

Yes. I've used them with and w/o the thermal control in the circuit though, and it makes no difference, assuming the rpms are set at the same speed.

RaNDoMMAI wrote:i have 2 globes fans and i find them very loud.

I have 9 of these Globes and they're very quiet. A few have a very, very slight "scraping" noise that I suspect is coming from the bearing spindle.

RaNDoMMAI wrote:Do by anychance have a nexus fan?

Nope, if I did I would have used it too.

Oh. Didn't see that. See i thought it was like:

[Temp chart]

[Analysis of Temp Chart]

instead of the other way around. -.-;;

*Slaps Ed for wasting SPCR resources for a little (

Not merely a , but a full-blown !

Edward Ng wrote:Not merely a , but a full-blown !

Note that mine was a , yours was a . We're both wasting resources now. WHEEEEEE~~~

Too bad you have no Papst 120mm fan to compare with the others, would be nice for all us European users.
Could you maybe note at what rpm the fans were spinning at the different voltages?

Found a minor typo during my reading on page3:
After that I plugged to fan into the power cable

to should be the

Great review and an exciting product.

Are there any plans for a Socket A version? I hope a Socket A/754/939/940 version isn't too much to hope for.

Tom Brown wrote:Are there any plans for a Socket A version? I hope a Socket A/754/939/940 version isn't too much to hope for.

No socket A but the review mentions more than once about compatibility with all A64s.

MikeC wrote:

Thermalright wrote:Our Heatpipe construction was also redesigned to eliminate any gravity effect and to unleash the capability of the XP-120's performance to operate at 100% in any direction you mount it.

Would it not be worthwhile to confirm?

/Ralf salutes smartly and heads across the hall to Hutter Labs. <indistinct sounds of banging, grinding and the occasional muttered curse>...

OK, I'm back with all the hot poop on horizontal vs. vertical XP-120 temps. First I quickly retested my current configuration of the 5V Globe, sucking mode, vertical and got 28°C idle, 42°C 2xCPUburn load. Then I ripped my test rig apart, jury-rigged a vertical setup and retested the temps. Results are: idle 27-28°C, 2xCPUBurn load 41-42°C. Looks like Thermalright was telling the truth about that one, eh?

Here's a pic of my vertical config. Note the orientation of the heatpipes. That's the only way the XP-120 will fit on this particular board:

/Ralf salutes smartly and heads across the hall to Hutter Labs. <indistinct sounds of banging, grinding and the occasional muttered curse>... Looks like Thermalright was telling the truth about that one, eh?

Thanks for that Ralf.

Now I'll have to go back into the article and add this info <Opens up Dreamweaver, FTP client, etc... <indistinct sounds of keyboard banging, rapid mouse clicking and the occasional muttered curse>

sorenbro wrote: Could you maybe note at what rpm the fans were spinning at the different voltages?

Only rpm data I have is from the Globe fan. The other fans used in the testing have no tach output and were run strictly to a fixed voltage.

120mm Globe info:

5V - 680rpm

7V - 1040rpm

12V - 1335rpm

-All of these rpm readings were taken on a thermally-controlled fan with the thermal control in the circuit, running at it's slowest (coolest) speed.

Now that's service! Thanks for being so responsive, so quickly, Ralf.

MikeC wrote:

/Ralf salutes smartly and heads across the hall to Hutter Labs. <indistinct sounds of banging, grinding and the occasional muttered curse>... Looks like Thermalright was telling the truth about that one, eh?

Thanks for that Ralf.

Now I'll have to go back into the article and add this info <Opens up Dreamweaver, FTP client, etc... <indistinct sounds of keyboard banging, rapid mouse clicking and the occasional muttered curse>

My normal computer workday would be more like... <indistinct sounds of keys banging, continuous mouse clicking/moving, muttered curses, and the occassional burp. *BURP* >