What blocks should I buy to test? Need 5

[extremecorvette]

Looking to do some Waterblock testing here at Iowa State University. I need to find out what you guys think are the top 5 water blocks out there. I've personal had three water blocks made and want to compare them to the best out there. I've had a test die made also and have full use of the labs to do my testing. The best part is that Iowa State is help paying for all of this and giving me credit for this research.

[Voodoo Rufus]

1. Cascade, if you can find one.
2. WW
3. DD RBX4
4. Swiftech MCW5002
5. Dtek TC4


Look here too:
http://www.overclockers.com/articles373/wbsum.asp

[zip22]

id say the list above but instead of the tc4, the asetek antarctica

[Korndog]

how about Silverprop or polarflo instead of tc-4?
tc4 is like 25 bucks now, not fair to compare it to ~50 blocks and imo, cascade shouldn't be included because its not mass produced. whats the point of testing something thats so hard to get? to top it off its total out of the price range from the other blocks. We all already know its the best block..

[LtBlue14]

slit edge?

[Arkangyl]

Cascade
Whitewater
RBX
Antarctica
PolarFLO

with the alternates being;

MCW-5002
DD Maze 4
TC-4
Astek normal

put them in in that order, prolly going to be using the 5002 instead of the cAscade

[UberBlue]

Leave the Antarctica out of it! It's probably an exact copy of the White Water anyways. Why test two of the same block?

I like Voodoo's list.

1. Cascade, if you can find one. (I've got a spare you could borrow as long as I get it back)
2. WW
3. DD RBX4
4. Swiftech MCW5002
5. Dtek TC4 (or Maze 4)

It's well rounded with representitives of the major design families.

[zip22]

the antartica is not exactly the same block, the antartica machining/finish seems a lot nicer internally than dteks and the design is undoubtedly different with different base and fin thicknesses etc. i would not brush it off as an exact copy. its a new block and should be tested instead of the older tc4 and maze designs which have already been proven to be inferior


edit: a SPARE cascade :drool:

[UberBlue]

Where are the dimensions for the fins/channels/baseplate published at?

It really boils down to what extremecorvette wants. He said the top five. That automaticaly excludes the Antarctic as it hasn't been tested to place it in the top five. Yes I know it's bound to be a good performer, but IMO the difference isn't great enough between the WW and Ant. to justify bumping a different block out of the lineup.

The Antarctic is bound to tested by a trusted source soon because of the poo storm its stired up. No need to lump it in with this test.

[zip22]

if he is looking for the top blocks then i don't think those older models should be included. the results for the white water and the antartica will most likely be similar, but i think it would be interesting to see how they differ.

the dimensions are not published anywhere, but i highly doubt they will be identical to the whitewater.

although it will most likely be tested numerous times, so have the other blocks mentioned.

if he is looking to study the top products, i think the newest main stream commercially available options would be the most interesting to look at (i think the cascade might make an interesting addition though). i don't think blocks that are being sold off are valuable additions

[UberBlue]

I could be persuded into thinking the TC-4 could go the way of the Dodo, but the MCW5002 is still a top 5 performer.

If asetek is claiming the best performance in the world, I'm betting it's an exact WW clone where it counts. They didn't bother to look harder to find out the WW isn't the top performer in the world anymore. When was the last time you saw the Cascade mentioned on any front page? It's a forum entity, fairly unobvious to non-forum goers. Sure the boutiqe manufacturers like Swiftek and DD know what the Cascade is, but asetek?

[Voodoo Rufus]

It'll be a forum entity until someone figures a way to mass produce a similar version. I wonder if Dtek will make them someday.....

I also wonder why we haven't seen the TC4 ever tested on the main page. Probably the last best maze style made.

[johan851]

I like Voodoo's list as well. It's nice to have the TC-4 in there as comparison to the other blocks, kind of a reference point. It'll give some indication as to whether or not the other blocks are really worth that extra $25.

[Voodoo Rufus]

I won't speculate as to its performance (as testing makes that kind of irrelevant these days), but I'd bet it's on par with the non-impingement blocks.

Is it worth it? To me, it depends on how big your wallet is. :)

[pauldenton]

Originally posted by UberBlue
Where are the dimensions for the fins/channels/baseplate published at?

It really boils down to what extremecorvette wants. He said the top five. That automaticaly excludes the Antarctic as it hasn't been tested to place it in the top five. Yes I know it's bound to be a good performer, but IMO the difference isn't great enough between the WW and Ant. to justify bumping a different block out of the lineup.

The Antarctic is bound to tested by a trusted source soon because of the poo storm its stired up. No need to lump it in with this test.

the other complication with the antarctic is that it comes with their 3/8" push fittings... so you either change them for 1/2" ID ones (in which case the block isn't stock...) or you're testing it in a different setup to the other blocks (step down fittings, or whatever)

[Cathar]

Originally posted by UberBlue
Sure the boutiqe manufacturers like Swiftek and DD know what the Cascade is, but asetek?

Oh, Asetek are aware of the Cascade. They were made aware when I ran that overclocking challenge between a Vapo PE and the Cascade SS, and the Cascade SS came out better for Barton cored CPU's, and was about even for the T'Bred B.

The guy with the Vapo PE had been posting at the Asetek forums to discuss the results with one of the Asetek afficiando's (an actual Asetek employee) asking about just how such a thing was possible.

All of the major manufacturers world-wide are aware of the Cascade. They do follow the forums. The thing that would probably bother them about most it is that it takes about 60 minutes of CNC machine time, ~15 minutes of manual non-CNC machine time, and about 30 minutes of hand-work to make each one, and it's absolute hell on tools with it blunting tools every 10 blocks or so.

The White Water derivatives can be machined fairly quickly with arbored slitter saws, or in the case of the Slit Edge or RBX, CNC machined fairly quickly with larger mill bits. Either way, it's about 15-20 minutes of total machine time to make a White Water derivative.

[vonkaar]

So... in a sense, the Cascade could never be mass-produced unless it had a retail price of at least $200? No wonder nobody is really trying to go anywhere with it...

Of course, as soon as some sweatshop operator in Mexico figures this out, he'll discover some way to operate a CNC at 12 cents an hour and we'll have jet-impingement blocks all over! Go go capitalism!!

[extremecorvette]

I would like to stay under $100 for each block.

[vudoodoodoo]

Originally posted by extremecorvette
The best part is that Iowa State is help paying for all of this and giving me credit for this research.

:cool:

[deathBOB]

I would like to see you include one of the silverprop blocks, simply for personal curiosity...

[extremecorvette]

I think I'm going to go with

1. Danger Den RBX

2. Danger Den Maze4

3. Little River Cascade

4. Swiftech MCW5002

5. Cyclone Evolutions SE

[extremecorvette]

I have my CPU, NB, and just getting my 9800pro waterblocks done. The water block for the 9800pro water cools the memory also. I hope to get some pictures up in a month.

[Cyrix_2k]

Boot the DD Maze4 outa there and go with a D-tek WW.

[extremecorvette]

Done

[JFettig]

yes definately, dont test the maze4, the WW would be a great block to test against those.


Jon