Enzotech-LUNA CPU waterblock

[ElMoIsEviL]

Test in progress ... for this moment:
- luna has better temp in idle than gtz - 1,2 degress
- other important thing: Swifty is more flowkiller:

RPM of MCP 355:
- GTZ: 4055
- Luna: 4355

I think I just had an orgasm looking at your setup... it's simply beautiful. You know you're a geek when....

[saxile]

bitspower is always a sexy work of art, the p0rn of the computer world.

[ElMoIsEviL]

bitspower is always a sexy work of art, the p0rn of the computer world.

Tell me about it.

If only they made blocks for the Gigabyte EX58-Extreme. I'd be all over it like a fat kid on ice cream.

[ElMoIsEviL]

They do:

http://www.ncixus.com/products/35581...1366/EnzoTech/

Comes with a backplate also. Only issue is, make sure you install your chip first, then the adapter, it blocks the socket lever that holds the chip down, so once installed you can not remove the chip.

You just sold me on that and as for the Bitspower fittings.. I'm going to test some out.

[Malik]

Some info about test method:

- Program for reading temps: RealTemp 2.90 ( maximum temps was marked )
- Idle mode: 2h work with internet, winamp - Fans on 4V
- Stress mode: 10 loop on 3Dmark vantage with stock gpu -Fans on 12V
- Stock CPU mode: QX9650 3Ghz 1.18V
- OC CPU mode: QX9650 4.4Ghz 1.42V

And something about flow in RPM's ( higher result is better ) :

[Harshal]

Images are giving me trouble...
anyone else having same problem??

[ElMoIsEviL]

Images are giving me trouble...
anyone else having same problem??

I was until I visited his gallery now I can see them.

Pretty much the Luna creams the GTZ. I currently am using a GTZ and had ordered a Luna to use it on my Q9550 build.

As you can see above I ordered the Socket 1366 adapter. Luna is going in my main rig. Hellz yeah.

I also just ordered the Enzotech SNBW to replace the stock water block on my EX-58 Extreme. I saw a user on here rig his up with an SNBW and it looked pretty sweet.

Never used an Enzotech product before but it appears they make great stuff.

[Mjhieu]

Wow. another pretty block, it's so cool, but pls make a test!

[Eddie3dfx]

Malik, can you repost those results in the forum under a seperate thread?
I think thats excellent that people may skip over.

[ci2kla]

I'm using this block right now but I don't know if the result is impressed..
For my setup:
Jingway DP-1200 Pump
1/2 hose (changed)
Rad: BIX 360 w/ 3x 2k rpm 69CFM Fan

Q9550 @1.2v

IBT full load @ 50C. Ambient 17.5C

(not really a review but just for reference only)

[WaterFlex]

Wow. What about Fusion v2 vs Luna ?

[Hondacity]

malik

what back plate were you using?
how tight did you mount the waterblocks?
what paste were you using?
what paste application method did you use?
how many mounts did you do on the waterblocks?

my opinion is that the swiftech mounting isn't good, i have to say it, even martin had problems with it, i tested it too and it caused me problems too.

[NaeKuh]

The Luna 3 C better than the GTZ? Questionable that any of the lastest blocks would beat another by 3 C in a straight up comparison.

Maybe in a mixed block loop starved for flow?

andyc

whats my inital guess.

Malik can i see your load temps for each core?

i want to see how even you got it.

[Waterlogged]

I can't put my finger on it but, something just doesn't seem right. I just tested one of my DDC3.2's for flow vs RPM and was completely unable to sway the RPM's (zero, zip, nada) by adding restriction (ball valve on my testing res). How you managed to achieve a 263 RPM difference is....well....

[OdinEidolon]

may i suggest that enzotech ships one of these beauties to Skinnee for some testing that everyone will accept? (not that i'm questioning your results Malik )

[NaeKuh]

I can't put my finger on it but, something just doesn't seem right. I just tested one of my DDC3.2's for flow vs RPM and was completely unable to sway the RPM's (zero, zip, nada) by adding restriction (ball valve on my testing res). How you managed to achieve a 263 RPM difference is....well....

DC motors, as you get more restriction, the motor spins less. (you also get lower heat dump and yada yada)

Reason why its playing so well on his system tho is:

his bitspower freezer.

has to be one the the most restricitive blocks made in history.

[Vapor]

DC motors, as you get more restriction, the motor spins less. (you also get lower heat dump and yada yada)

With modern electronics it's really easy to implement logic that dictates motor speed based on changes in current draw (which occur when restriction changes). This goes all ways--some motors may speed up and some may slow down and some may be controlled to not budge at all.

[hellcamino]

Yep,

And that's why I never consider a CPU block performance test valid unless it's a CPU only loop. With everything Malik has in that loop, the fluid must of been crawling at way less than 1 GPM with the GTZ. I would imagine he saw his GPU temps drop a tad also with the Luna CPU block.

andyc

Being as most people do run multiple blocks in their loops I would call this type of test great for real world results.

[Waterlogged]

DC motors, as you get more restriction, the motor spins less. (you also get lower heat dump and yada yada)

Reason why its playing so well on his system tho is:

his bitspower freezer.

has to be one the the most restricitive blocks made in history.

Sorry NaeKuh, this is wrong. I just confirmed it with one of my DDC-2's. As the restriction rises, so does RPM, so the pump Malik is using is a DDC-2 and he's confirmed my suspicions about the Enzo block being very restrictive (probably on par with the EK). How he got the better temps with it though is still a mystery.

It would seem the DDC 3.x have the logic circuitry that Vapor mentioned so they will always run the same RPM.

[hellcamino]

Depending on impeller design I would think the motor would slow with restriction unless it would just cavitate beyond a certain amount of restriction. Where and how in your loop were you creating the restriction WL?

[SNiiPE_DoGG]

Sorry NaeKuh, this is wrong. I just confirmed it with one of my DDC-2's. As the restriction rises, so does RPM, so the pump Malik is using is a DDC-2 and he's confirmed my suspicions about the Enzo block being very restrictive (probably on par with the EK). How he got the better temps with it though is still a mystery.

It would seem the DDC 3.x have the logic circuitry that Vapor mentioned so they will always run the same RPM.

my ddc 3.2's spin at RPM's 300 apart, one ~4400 one ~4100

[ElMoIsEviL]

Yep,

And that's why I never consider a CPU block performance test valid unless it's a CPU only loop. With everything Malik has in that loop, the fluid must of been crawling at way less than 1 GPM with the GTZ. I would imagine he saw his GPU temps drop a tad also with the Luna CPU block.

andyc

So a full loop means lower results for the GTZ because it's more restrictive. This means that a CPU block test only is useless for those of us with several blocks in our loops.

I have two rads, res, an NB, GPU and CPU.

Luna looks to be the better block for me since it's less restrictive.

[Vapor]

It's pretty easy to test CPU blocks in a low-flow setup, even if it's the only thing in the loop

[ElMoIsEviL]

My point is, maybe so and maybe not. Unless you know and use the exact same blocks and setup, the information and results can be very misleading.

Without the same exact rad, fans, tubing, pump and/or top, blocks and everything else equal, another setup could render totally different results.

Most people would just look at the numbers and automatically think the Luna outperforms the GTZ on a purely thermal basis, which isn't the case. On a ultra low flow system which I see as the test bed here, a low restriction block such as the Luna will outperform a impingement/accelerated type restrictive block like the GTZ. But with flow over 1 GPM upwards to 1.5 GPM, which quite a few multi block loops can run, the Luna wouldn't outperform a block like the GTZ.

So to say the Luna outperforms the GTZ in a multi block loop as real world results isn't very accurate IMO, unless your real world is like the one tested, which is rarely the case.



Sure,

With a flow meter like the King, you can dial it down to anyflow you like. But that won't account for the performance delta between one block dumping heat in the loop and multiple blocks dumping heat. You might see a difference towards better performance on the less restrictive block, but no where near as much I would imagine. At least not 3-4 C.

andyc

My point wasn't that everyone setups is the same but that less restriction can make up for a 3-4C difference. There is no reason why it could not. It all depends on the variables. Either way I have the GTZ and have the Luna coming soon.

As for your claim that with 1 to 1.5GPM the Luna wouldn't outperform the GTZ. Well how do you know this? Because the GTZ is more restrictive? That makes no sense. The design plays a VERY big role. More restrictive doesn't mean higher performing and vice versa.

The Luna uses quite an interesting design that redirects the water stream over a larger area while not limiting flow as much as the GTZ. Interesting design.

[hellcamino]

I understand what you are saying Mcoffey but the average person watercooling is running a single loop with most cooling cpu and nb and second largest number adding a gpu on the end. Most are likely running between 1-1.5gpm making this type of test very useful information, personally I would like to see the tests performed as a cpu only loop and a multi block loop just for the average and high performance comparison of end user type results.