[Review] AquaComputer cuplex kryos XT Tested

[Vapor]

First review of the new i7 CPU Block Roundup #3....new testbed, some changes in testing method, etc., etc. More importantly, this is my AquaComputer cuplex kryos XT review



A big thanks to Shoggy for the review sample

Full review is here, but all the results and some impressions are in this thread of course

First up, great performance from this block....beat the EK Supreme HF P1 with CPU temps at typical/medium pumping power (though barely). Restriction isn't bad and not a deal breaker in the slightest. The block is sharp and shiny and solid, can't say I know a better way to describe how it looks and feels; it's really, really nice.

I'm not fond of the mounting system, but it is better than the EK and Heatkiller mounting system. The biggest problem is with assembling the backplate/hexnut configuration. I just don't think assembly should be required for mounting, this is reinforced by the fact there are great mounting kits out that do not require assembly.

The barb spacing is really tight...Bitspower and Koolance 3/8-1/2" compressions are a no-go and even 3/4" OD tubing with clamps is really, really tight.



The base with the thicker o-ring has a noticeable bow:


Anyway, onto performance!











Small win for the kryos XT, but a win is a win

Despite the mounting system, I'm really digging this block for its performance and its looks. The rectangular pin matrix is obviously providing a lot of performance and is really well done. The bent pins should really be of zero concern, there's no measurable performance loss and they're not going to break off

[imersa]

Nice testing there. Really good quality images of the block too!

[Martinm210]

Amazing work as usual..that pumping power vs temps is really ideal. Amazing how close the results are.

Thanks for the detailed work.

[Vapor]

Thanks Martin!

Here's the tables of data from all the tests so far....Supreme HF P1 vs. kryos XT (thick o-ring) are within hundreths of a degree at the four highest pumping power settings

[Alexandr0s]

Amazing job Vapor, love the test . I'd buy it if I didn't like the EK Supreme HF better. I really doubt I'd notice 0.05C difference .

[BlueAqua]

Very nice review Vapor.

[Church]

Hmm. I somehow expected a little bit bigger win hearing from Shoggy about big effort put in developing block and mirriads of variations tested. Current result combo (of both temps+restriction) runs a bit short for 'replacing king' in my eyes. Imho HF still occupies 'sweet spot' of overall performance. I won't count silver version, as it's price simply is in too ridiculously expensive area, but half degree win of this more common variety of block in temps probably can be catched up by full-nickel or full-copper versions of EK's HF, and you'll get as bonus better flow (especially now when one can buy replacement top to fit two 1/2"ID BP CFs aswell). I aplaud efforts of Aquacomputer, but my choice of full nickel HF as top block remains unchanged.
Imho this test shown that microchannels like that of HF's && CPU-360 have more potential to get to 'sweet spot' of temps vs restriction then that of micropins design, which was refined further from Apogee XT into this new Kryo WB, and i'm guessing that next iterations of Koolance's and EK's cpu blocks will be able to reach little bit better results with even more thin and deeper microchannels then current ones in both flow & temps unlike micropin arch that seemingly has hit limits. Afterall, it's easier to make thinner channels then thinner pins material strength & machining difficulty wise ..

[nikhsub1]

Nice review Vapor

[mk-ultra]

but half degree win

i don't know how you guys can declare a winner with such a small gain/lost, it is WELL within the margin of error from the testing

[Alexandr0s]

i don't know how you guys can declare a winner with such a small gain/lost, it is WELL within the margin of error from the testing

Who said half a degree? It's ten times less for the Kryos XT with thicker oring: 0.05 degree. I think the best at this point would be to declare a tie.

[Church]

mk-ultra: that's where the several mounts and average temps come in to eliminate errors.
Alexandros: or loss, as it's way more restrictive

[Vapor]

Who said half a degree? It's ten times less for the Kryos XT with thicker oring: 0.05 degree. I think the best at this point would be to declare a tie.

That would be like declaring a 110-109 basketball game a tie.

If I were to retest each block 5 times (not 5 mounts, but 5 tests of 6 mounts), maybe the results would move a little bit. After that many mounts, however, I'm sure my ability to 'master' an ideal mounting technique for either block would come into play even more.

The results are insanely close (look at the Low through Very High pumping power results--5 hundreths of a degree is the largest separation), but it would be disservice to the data to call it a tie (especially if I were to round down to tenths of a degree....then it would be 75.9C vs. 76C ).

[NKrader]

that would be like declaring a 110-109 basketball game a tie.

cheers!!

[@lex]

I somehow expected a little bit bigger win hearing from Shoggy about big effort put in developing block and mirriads of variations tested.

I'd say these new WBs have reached technological limit, don't you think?.. I would be surprised to see a new WB outperforming Kryos or EK HF by one degree within one year, at least.
Hmm, maybe a fully silver base?..

[Alexandr0s]

I doubt that's the same. Because during basketball, there aren't that many variables that come into play. The fact is, even over 5 mounts, 0.05 degrees difference can be caused by anything, even you exhaling anywhere near it. I know it's 'better' according to your tests (which I hold highly btw ), but IMO 0.05 isn't enough to declare a winner, not with that many variables coming into play. My main reason for this is that CPU's have a little something called rounding errors, they only give full degrees. Mounting it 5 times gives less of an error, but I still consider 0.05 degrees difference a rounding error by the CPU.

[Vapor]

If you can't tell, I have a stubbornness about ties

You can call it a wash, you can call it within margin of error, you can call it imperceptible, just don't call it a tie

I don't think the .05C difference means diddly, but one block did have a better performance for CPU temps at that setting in the data I have

[Alexandr0s]

Fair enough . I guess you got to be able to tell the masses something. No one likes a tie, I'm aware of that. Makes it look like nothing changed. But just as you don't like ties, I don't like rounding errors. Since it's your test, it's up to you how you want to present the data , it just won't change a thing for me. To me it's still a tie .

[randomhro]

I'm with Vapor on this one. The way he takes the averages means the only way for a tie is for them to be equal. If it was just one or two mounts per block it might be different, but scientifically speaking rounding error isn't as prominent when averaged out with the worst mount disregarded. 0.05C might not be much of a difference, but it is a little and that may be all we see for a while from block manufacturers.

Another good review.

[NaeKuh]

Guys the block is excellent... no doubt about it.

The only flaw i had was you cant use jumbo tubing on it.

But shoggy did say they made a new one with wider spacing, so i would honestly wait for that.

Hehe... this is probably one of my favorite shots.

[mk-ultra]

mk-ultra: that's where the several mounts and average temps come in to eliminate errors.

ELLIMINATE ERRORS...that's a pretty bold claim here, how can such a test be 100% errors free, not 99.9%, 100 fricking percent errors free. There's always a margin of error, of a degree at the very least.

there's too many variable to not have a margin of error,(temps readers, ambient temp,water temp, humidity, cpu load applied,(i'm sure i'm forgetting a lot)
what if ambient changed from a tenth of degree it would be a world of difference for those small gaps data.

i'm not looking to dish out those reviews, i love the effort put into them and everything have been made to have the best data possible. i'm happy to see another good block on the market, being superior of a tenth of a degree tho doesn't make a world of difference. You can call the better performing block a winner but with those gaps, i take it with a grain of salt. The datas are more important than the decision at the end.

ok my rant is over, keep up the good work vapor

[Shoggy]

But shoggy did say they made a new one with wider spacing, so i would honestly wait for that.

You don't have to wait for it. It is already available

[longh]

hey Shoggy, i'd like to know if there's a nickel plated version of the Kryos HF in your roadmap.

Great work on that waterblock and nice review vapor

[quattro_]

grate work as usual Vapor, thanks .

[Captain H.N.]

You don't have to wait for it. It is already available

Where outside of Europe is it/will it be available? :/

[masska]

Where outside of Europe is it/will it be available? :/

it will actually be available next week or the week after that.
I'm sure most of the main retailers will eventually stock it, as are we and we ship international. PM me if your interested