new swiftech block?

[snowwie]

I know billa often lurks around these forums, so I thought I'd ask here

I was just browsing the swiftnets website, looking at latest prices for the MCW5002, and saw the MCW6000

well, anyone have any explanations or opinions?

what I can see from the site description:

-signifacantly better than the 5002, ~2 degree celsius better performance
-all copper, one-piece design, even the barbs are copper
-innard design remeniscient of Hoot's thin-pin design (see overclockers.com)
-center inlet, both barbs are for 3/8" ID hose
-versatile hole mounting (but it still uses socket lugs for AMD)
-40 bucks msrp, apparently available now

well, what do ya think?

[AngryAlpaca]

ALMOST 2C better performance? 0.015C/W better, I see at the bottom, at 10LPM. I'm disappointed. Really disappointed. Twice as much restriction!? I hope this wasn't the big thing that BillA alluded to earlier. There's not even anything noticeably innovative about it! What the hell? :shrug:

[bigben2k]

It's based on the tried and true pin fin block, which is an excellent performer, especially for the price.

Note how the target market here is servers. This product would serve that need above high performance, which really, it already has. A "Cascade" block would be a poor choice here (high risk due to very thin baseplate).

It would be nice if a block-maker could break into the server market: it'd open things up for everybody.

[snowwie]

well, first off, thin-base "almost-direct-die" blocks were always an enthusiast product.

I never liked them solely because I want a broad base plate cooled to my expectations.
The cascades, white waters, and RBXs I always had qualms about blocks of that design.
I still do.
I will buy a maze or mcw5002 or aqua joe thank you very much.

I never expected to swiftech to make a block of such a design either, doesn't seem their cup of tea; if they would make one like that, it would only be to grab a peice of the market where every enthusiast and their mother are now obsessed over an RBX/WW block.

I do see them all over the "turbulence" and "impingment" buzz words though

why can't anyone accept the fact that surface area and water velocity are two extremely important factors in water block design? no, they only accept the simple words that cathar once used for briefly describing the way his blocks work

every time I see those words in the water block design/construction forums I cringe

anyway, I dunno if I'm so keen on it

[AngryAlpaca]

Watercooling is enthusiast! Hell, non-stock aircooling is an enthusiast product! So, you don't like good blocks because they aren't built the way you think good blocks should be? You're ignoring facts? There's this strange thing about water velocity. When blocks use impingement, water velocity is drastically increase, creating more turbulence and reducing the boundary layer, improving cooling. Surface area is placed throughout the WW, RBX, and Cascade. The fins and cups are there for a reason. Utilizing a small amount of surface area is much more effective than just increasing it. Have you learned nothing over the past few years?

[prandtl]

Originally Posted by snowwie

I do see them all over the "turbulence" and "impingment" buzz words though

why can't anyone accept the fact that surface area and water velocity are two extremely important factors in water block design? no, they only accept the simple words that cathar once used for briefly describing the way his blocks work

errr Impingment is used solely to increase water velocity and hence turbulence...
with higher turbulence, read lower thermal resistance, you dont need (or cant have) bigger surface area. You either work with block wich have lots of area and lower velocity or smaller area with high velocity (smaller surface area where water velocity is high).
Why you cant use both you might ask (lots of surface and high velocity)? Because the result will be a hell of a restrictive block.

[BillA]

hey Ben, what does a quick 'eye balling' reveal ?
jk, jk

AA
all things are comparative
look at the wbs available today, their performance, and their price
95% of the wb market just got redefined, big time

and you missed it, but Ben did not

snowwie
the bp is 3/16" thick, not exactly thin by any measure

[snowwie]

well, I didn't really expain myself too well, I know

But an example is this:

some guy posts about making a thin-fin water block

using a minimal amount of copper, he wants to place grooves parallel to the flow of water, refer to attached pic

[BillA]

Originally Posted by snowwie

well, I didn't really expain myself too well, I know

But an example is this:

some guy posts about making a thin-fin water block

using a minimal amount of copper, he wants to place grooves parallel to the flow of water, refer to attached pic

not Swiftech, no comment

[snowwie]

but some guys says:

"no, you want turbulence, because it's better and I don't really know what turbulence is, so I'm going to talk out of my ass"

and suggests this, because the he thinks the water will slosh around more this way, which is the secret to water cooling

[snowwie]

Originally Posted by AngryAlpaca

Watercooling is enthusiast! Hell, non-stock aircooling is an enthusiast product! So, you don't like good blocks because they aren't built the way you think good blocks should be? You're ignoring facts? There's this strange thing about water velocity. When blocks use impingement, water velocity is drastically increase, creating more turbulence and reducing the boundary layer, improving cooling. Surface area is placed throughout the WW, RBX, and Cascade. The fins and cups are there for a reason. Utilizing a small amount of surface area is much more effective than just increasing it. Have you learned nothing over the past few years?

I agree, rbx/ww are high performance and very cool designs

but about watercooling being enthusiast, it's slowly changing

with 75 watt video cards and prescotts, etc., you know that water cooling is in the back of even DELL's mind
edit/
about ignoring facts, it is in my personal oipinion that surface area and water velocity epitomise water block performance, NOT turbulance. the words turbulence and impingement just seem too vague too me

[AngryAlpaca]

I hadn't noticed that this price is far below that of the MCW 5002. Well, I wouldn't have thought that the server market is a big one, for watercooling and such. I would've expected the ultra low cost end to be the market to get. Enthusiasts aren't your main customers, are we? I see that this price is far below the 5002, and the performance is better, so I suppose that would pick up a lot of sales, I guess. 95% of the market got redefined? Or the market just expanded? Hmm... Come to think of price/performance this is a good move, but I still expected better than shrinking it, and changing from the diamond matrix to pins. Hmm...

Edit: (You should try it) What are you trying to say? None of those have to do with impingement, and, incidentally, increased water velocity is caused by turbulence. Also, Dell isn't going to bend over backwards to accomodate Intel. Intel will bend over backwards to accomodate them. High wattage processors are unacceptable.

[#Rotor]

what good will a lot of surface area do you, if you have a nice and thick boundary lair coving it .

Surface area should be the result of a design that takes water velocity (read potential turbulence) and turns it into turbulence.... the water's velocity is useless, if you don't do something with it.... and without a turbulence inducing design. you might as well have a hydrofoil for a block.... minimum drag, maximum flow...
The myth is this instance, as it was for so long, is that a block will perform well if it can let a lot of water through it fast....then explain the strangeness of the fact that a maze1 has nothing on one of Cathar's beauties, i'm pretty sure the maze1 kicks it's ass on the flow-rate scores... not so...

the reason is simple... it's not how much you flow, it's what you do with that flow, that will make your block work well.

[AngryAlpaca]

Goddamnit now that I've thought about it (I'm still disappointed) this block is probably revolutionary, in the marketing at least. I hope that you'll get a sample to JoeC and pHaestus pretty quick.

[snowwie]

Originally Posted by AngryAlpaca

incidentally, increased water velocity is caused by turbulence. Also, Dell isn't going to bend over backwards to accomodate Intel. Intel will bend over backwards to accomodate them. High wattage processors are unacceptable.

I'm not so sure that both of those quoted statements are true

increased water velocity is caused by a decreased cross-sectional area for flow
what is your defonition of turbulence?

and I think that Dell just may bend over backwards for Intel. Intel's name takes a percentage of the reason for why people buy their machines

It's not that Intel isn't trying to improve their processors, but it is a fact.

more performance calls for more energy which calls for more heat management

things will only get hotter in the future

[#Rotor]

I would not say the server market is new to this... it's where this started long ago... even I have had a U1 solution out there for a couple of years now... heck I'm running one of them right now as a replacement for the block that went to PH.

[joemac]

Nice but why only use 3/8 barbs :shrug:

[snowwie]

oh, and I never meant to imply that I thought swiftech is going the RBX/WW route

I just saw the word "turbulence" in their description and lost my cool

[prandtl]

Originally Posted by #Rotor

then explain the strangeness of the fact that a maze1 has nothing on one of Cathar's beauties, i'm pretty sure the maze1 kicks it's ass on the flow-rate scores... not so...

strangeness? I'm sure, althought flow is higher in a maze than in a cascade (with same pump), water velocity, on the other hand, is higher in the cascade's cup than in any point of the maze.
That said, I agree that the geometry of a block can improve turbulence, with everything else constant, but keep in mind that turbulence do not cause higher velocity, but the it's the other way around.

increased water velocity is caused by turbulence

that's completly wrong.

[snowwie]

Originally Posted by joemac

Nice but why only use 3/8 barbs :shrug:

I don't know, not so bad, because the cross-sectional area for flow is reduced anyway to increase the water velocity, I'm sure they tested it with 1/2" holes and found lesser performance because of lesser water velocity turbulence

so 3/8" barbs gives more turbulence, duh, and more turbulence is better! yay!

I'm learning already

you will have to buy an adaptor though to use your 1/2 tubing, pity, 1/2 barbs would have been easier
but as this block is possibly catering more to the mainstream and even OEMs, as others have pointed out, 3/8 barbs are indeed easier I suppose

[snowwie]

Originally Posted by prandtl

that's completly wrong.

thank you

i'm not crazed

or am I?

[AngryAlpaca]

My mistake. I was in a rush to post. Increased water velocity has the same effect of turbulence. That is, reducing the boundary layer. It seems I took too long posting as it is. My maki is now cooked.

[BillA]

Originally Posted by AngryAlpaca

Goddamnit now that I've thought about it (I'm still disappointed) this block is probably revolutionary, in the marketing at least. I hope that you'll get a sample to JoeC and pHaestus pretty quick.

"revolutionary" is a bit extreme, such progress is evolutionary in nature
but note that this is a product that can only exist by virtue of high volumes

while we sell thousands of wbs, the design impetus for the MCW6000 was the 'opportunity' provided by the MCP600 pump to utilize an increased head loss in the wb - making finally a more balanced kit design (the goal)

there is a 1/2" barb version which will be announced in several weeks
(quite similar performance, but better for 1/2" ID tubing systems - will have a 1/2" barb rad also, waiting on the res)

note that the MCW5002-T will continue in production as it is a better solution for TEC heads

snowwie, yes

[snowwie]

Originally Posted by #Rotor

what good will a lot of surface area do you, if you have a nice and thick boundary lair coving it .

Surface area should be the result of a design that takes water velocity (read potential turbulence) and turns it into turbulence.... the water's velocity is useless, if you don't do something with it.... and without a turbulence inducing design. you might as well have a hydrofoil for a block.... minimum drag, maximum flow...
The myth is this instance, as it was for so long, is that a block will perform well if it can let a lot of water through it fast....then explain the strangeness of the fact that a maze1 has nothing on one of Cathar's beauties, i'm pretty sure the maze1 kicks it's ass on the flow-rate scores... not so...

the reason is simple... it's not how much you flow, it's what you do with that flow, that will make your block work well.

heh, I missed your post

I liked reading that

my recollection of the word turbulence is an airplane going through some rough weather.

and I was more commenting on the healthy combination of water velocity and surface area. Much of it also has to do with, I suppose you can call it sloshing, where a larger percentage of the water molecules are actually exposed to the copper surface, and I guess that is where the constant reference to turbulence comes in, but now I entering the realm where I have no idea of what I'm talking about. Oh, wait. I entered that when I started this thread. No, when I joined pro-cooling. Now I remember why there was such a long period of time during which I refrained from posting.

lol

[joemac]

Originally Posted by snowwie

I don't know, not so bad, because the cross-sectional area for flow is reduced anyway to increase the water velocity, I'm sure they tested it with 1/2" holes and found lesser performance because of lesser water velocity turbulence

so 3/8" barbs gives more turbulence, duh, and more turbulence is better! yay!

I'm learning already

you will have to buy an adaptor though to use your 1/2 tubing, pity, 1/2 barbs would have been easier
but as this block is possibly catering more to the mainstream and even OEMs, as others have pointed out, 3/8 barbs are indeed easier I suppose

How can that be? Using smaller barbs slows down the flow. How can that be a good thing?