water cooling worth it?

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
Seems that we have another expandable AIO now: the Raijintek Triton. Too bad I can't find any reviews atm, but the specs look alright (4W pump, as opposed to the 1-2W used in most CLCs).
 

Cerulean

[H]F Junkie
Joined
Jul 27, 2006
Messages
9,269
Water cooling only worth it if it's integrated with your swimming pool / house water pipes. :D
 

Tsumi

[H]F Junkie
Joined
Mar 18, 2010
Messages
13,378
4 W is still a far cry from the 18 W MCP 355 and 24 W MCP 655.
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
I know, I'm just saying that for the Triton's $105 price tag and expandability along with a 4W pump, it makes all the CLCs on the market right now much less attractive, and is a lower cost alternative to the H220X and H240X, although with a less powerful pump.

In fact, this is what I'd probably use right now had I not already setup my custom loop.
 
Last edited:

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
H140-X, H220-X & H240-X use an 8w pump.

4w pump will have a hard time pushing though 2 blocks .. and single radiator is not big enough to cool both CPU and GPU.

Triton warranty is only 2 years and is at Raijintek's discretion if seal is broken to add even their supplied dye. It's almost as if they dye is included to entice the buyer to break the seal so they have an easy way out of warranty.

I would suspect 2 years is about all it's good for .. same as other CLCs.
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
The Swiftechs use a 6W pump, not 8W, and the Apogee XL is also a high restriction block. The Swiftechs are single radiator as well.
 

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
My bad, yes they are 6w pumps.
I believe the pump is still able to move more volume even with the higher resistance of the Apogee block .. which is also a much better cooling block than CLCs are using.

Edit:
What do you mean by the all use single radiators? Not all radiators are created equal. :D
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
The Apogee XL is one of the worst blocks you can buy, its performance near or at the bottom across 3 different CPUs, in addition to being high restriction. How it would compare to the Triton's block I don't know.

Both the H220X and H240X use a slim radiator, and in addition one fan space is blocked off due to the pump/res so you can't do true push-pull.
 

Tsumi

[H]F Junkie
Joined
Mar 18, 2010
Messages
13,378
Holy crap, I didn't think Swiftech would be making a major step backwards in performance with the XL. The chart does make me content to stay with my Heatkiller block though, lol.
 

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
The Apogee XL is one of the worst blocks you can buy, its performance near or at the bottom across 3 different CPUs, in addition to being high restriction. How it would compare to the Triton's block I don't know.

Both the H220X and H240X use a slim radiator, and in addition one fan space is blocked off due to the pump/res so you can't do true push-pull.
Please research what you post a little better.

The Apogee XL is not one of the worst.

Slim radiators are not all the same. Core count, core size, fin count, etc all contribute to how well they transfer heat.

As for single push versus push pull, a slim radiator has no need of push/pull when appropriate fans are used.
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
Well I have another review that puts the Apogee XL at the bottom again and disagrees with your HiTechLegion review. :D

As for the single radiator, well you brought up the point that it's not big enough to cool both the CPU and GPU. Well the Swiftechs also use a single radiator. Yes FPI and all that, but point being you can't claim anything either way because there are hardly any reviews out on the Triton, much less a direct comparison between the Triton and the Swiftechs. In any case, a single Alphacool UT60 280mm rad was found to be enough to keep a 3930K @ 4GHz and THREE 780s under control. Granted it's a full custom loop with a proper pump and a 60mm rad, but it can be done. You're not going to be pushing 1 CPU block + 3 GPU blocks with a dinky 4W or 6W pump in any case so that point is moot.

Also even at 2200 rpm the UT60 only has a 14% lead over the comparable ST30 slim radiator in terms of thermal dissipation, so thickness doesn't really help too much, at least if you only do push.

The point about the push/pull is that with the Swiftechs you lose 1 fan mount permanently. All else being equal, it will have slightly less cooling capacity. You don't "need" a lot of things, doesn't mean it isn't nice to have the potential to do such things.
 
Last edited:

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
I don't know how good or bad it is. I do know a single push or pull fans of 70cfm with 6.0mm H2O static pressure will move more air than 2x push / pull fans of 70cfm with 2.0mm H2O.

If radiator was a 3" thick radiator than push / pull might be needed to overcome the resistance. but a 1" thick radiator doesn't create that kind of resistance.

Have you read any of Martin's testing and tutorials on Martin'sLiquidLab? very imformative. ;)
 

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
Because "all else is not equal." Nothing is equal because all things are different;, different coolers, different radiators, different fans, etc.
 

Nobu

Supreme [H]ardness
Joined
Jun 7, 2007
Messages
4,600
A slim radiator with more fpi (~=surface area) will generally perform similar to a thick radiator with fewer fpi, however the thin radiator may need a better static pressure fan because of how close the fins are together. A thick radiator will generally perform better than a thin radiator with similar fpi, assuming both radiators have an appropriate fan configuration.

fwiw, my H80i (in push-config) keeps my A10-6800k at around 20C below critical (when stress-testing) with vcore at 1.43. That's with cheapo thermal paste from RadiosHack. Definitely not going to push it any further without upgrading. :eek:
 

Tych-0

Gawd
Joined
Jul 7, 2010
Messages
544
Didn't have the time to read the entire thread, but I figured I'd drop my 2c in on custom WC setups.

I'm just finishing up my first WC build, and I have to say that I have got a lot of enjoyment from it. It's been like big kid lego for me. Performance wise I can't speak much for, but I know with the crazy amount of rad space I have it will be much quieter than I'm used to.

That said it's been just insane dollar wise, this has run me far beyond what I expected. I went all out on mine though and really spared no expense. Case Labs Mercury S8, RVE, 5930k, 3x 360 + 1x 240 rads, EK blocks. I'm sure it could be done much cheaper, but I can't imagine it is ever worth it price/performance - at least until you re-use parts for the next build, which I'm sure I will.
 

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
Fins per inch per core is not just 2 dimensional, but 3 dimensional. A radiator with cores 4, 3, or 2 core thickness increases airflow resistance with each additional layer of cores.. More layers, more resistance, means more airflow pressure / static pressure is needed to overcome the resistance to have similar airflow. Stacking fans in push/pull increases the pressure. And a single fan with the same static pressure and airflow of 2 fans in push/pull move an equal amount of air .. and cools the same.
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
Because "all else is not equal." Nothing is equal because all things are different;, different coolers, different radiators, different fans, etc.
Let me backtrack. You're saying it doesn't matter if the Swiftechs can't do push/pull because you can just throw on fans with better static pressure, and in push only they'd still do better than with weaker static pressure fans in push/pull. So my question is what's preventing you from using the same high static pressure fans on the Triton where you can do true push/pull config?

Additionally, you may also find this post from Martin informative.
 
Last edited:

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
Let me backtrack. You're saying it doesn't matter if the Swiftechs can't do push/pull because you can just throw on fans with better static pressure, and in push only they'd still do better than with weaker static pressure fans in push/pull. So my question is what's preventing you from using the same high static pressure fans on the Triton where you can do true push/pull config?

Additionally, you may also find this post from Martin informative.
You keep twisting the facts trying to gain an advantage.

Posting a link to Martin's answer
Slim + Push/Pull is pretty much a sure win over thick and just push or pull.
The question he was answering is
So I was wondering what was better, a 60mm radiator with push, or a less than 55mm radiator with push pull? My 420 rad doesn't fit in the top of my case with push pull frown.gif

Also the 5mm slimmer rad has a fun density of 20 while the other one a density of 10
What Martin is saying is that using a good thin radiator with push/pull will give more improvement than going to a thicker radiator with only push. I suspect he is assuming both will be using the same or similar fans.

What he is saying is not the same as what I was trying to explain and discuss with you. I say "was" because I will not continue as you seem to care more about winning than the truth.
 

n=1

2[H]4U
Joined
Sep 2, 2014
Messages
2,388
Uhhh all I said was that his post was informative lol.

But hang on let's recap. I made a mention of the Raijintek Triton, another expandable AIO. You then pointed out that compared to the Swiftech AIOs it's still crap because the 4W pump will have trouble pushing through CPU and GPU blocks, that a single radiator won't be able to cool both the CPU and GPU, and the 2 year warranty is because 2 years is all it's good for.

I pointed out the Swiftechs aren't perfect, have their own flaws, and your criticisms of the Triton are unwarranted because they apply to the Swiftechs as well. And my only criticism of the Swiftech was it not being able to do true push/pull due to loss of a fan mount. You said it didn't matter because you could use a stronger fan to compensate. I simply pointed out there's nothing stopping you from using the same stronger fan on the Triton to do true push/pull.

And actually I linked Martin's post for this part:
Martin said:
Don't get too wrapped up in fin density. Generally the more dense will do better with higher speed fans, but that isn't always true either since there are different fin types. I've seen the opposite in too many examples now to really depend on FPI to be much of a guide. The slimmer the rad the less restriction through the fins as well, so it's a number of variables.
So from my perspective, I simply made a mention of the Triton, then you made some unwarranted criticisms of the Triton and tried to discredit me in the process. So yeah of course I'm gonna be defensive lol.
 

doyll

[H]ard|Gawd
Joined
Feb 4, 2012
Messages
1,229
I think we both misunderstood the other. :D
I'm not saying the Triton is crap, just that is little better than other CLCs. I say CLC because breaking the seal voids the warranty.

I assumed the first part of Martin's answer was what you were referring to because your post was about push/pull and static pressure .. and you started the sentence with "additionally."

More often than not push/pull is over-rated. People will spend more for 4 fans to get 4mm H2O of pressure that 2 fans with 5mm H2O pressure cost. Push/pull sometimes cools better at low speed, but not by much .. and as speed increases this advantage is lost.

The Triton also puts me off because Raijintek is a new company with a line of air coolers. As the quality of their coolers is just mediocre, I have a hard time believing their CLC will be any better.

Even Fractal Design's new Kelvin coolers have problems. And Fractal Design has a history of great cases.
 
Top