TDX or Dtek Whitewater
 

I have a few questions about both of these blocks. Right now I am running a 2500 Barton and i will be upgrading in 2 months, and i want to use this stuff im buying this week hopefully, when i get it. Im pretty sure if i get the TDX i would need to buy some more mounting hardware if i was to the the socket a version now. Im pretty sure (from what it looked like) that the whitewater block if i was to order the version with the a64 top piece on it all ready if i would be able to take it off and the socket holes would be usable so i did not have to buy any extra hardware.

Which block do you guys think is better. So far i have a 2-302 HC and some 1/2 id tubing. I am going to get the Aquaextreme pump and the HDPE reservoir from dtek, cause they seem to have the best price on the pump. That leaves me with a choice of blocks. I can order the Whitewater from dtek as well, or i can order just the TDX from up here in canada with both tops for about the same price after shipping. Which block would you recomend me for my system.

BTW sorry if half of this did not make sense, really tired and wanted to get it posted before i fell asleep :) TIA

This should answer your question:
http://www.procooling.com/html/pro_testing.php

(In short, go for the WhiteWater).

I prefer the TDX over the WW because it only has 2 barbs and doesn't need a Y adapter. Both blocks should perform pretty similarly if TDX is equipped with #4 nozzle. DO get the #4 nozzle insert if you use the TDX. You will need to buy an A64 top for the TDX though which I think costs around $10

and he will have a flow rate of <1.5 gpm (probably ~1.35 or so)
??

I would agree with that flow rate Bill. With a loop containing a BIX, 1/2" tubing, TDX+#4 nozzle, a swissflow flowmeter, and the AquaXtreme pump I measured 1.33. The flowmeter's pretty restrictive so real world should be a little higher than that

hmm, could be a bit higher then (would estimate the flowmeter at ~0.15 at max)
and so with a flow rate in that neighborhood, . . . .
it would seem that some $ could be saved and yield better performance
??
- of course the poster did not ask this question, lol

save money? better performance? tell me how.

with the white water i was thinking that if i got that reservoir it has 2 inlets on it so i wouldn't probably use a Y adapter. and with the white water if i get the one with the a64 adapter, i can take that off for now and it will fit on my socketa mobo right?

poke around the link

It seems the good Mr. Adams is erring on the side of prudence, but I'll make it obvious. The MCW6002 is a better, less expensive block than either the TDX or Dtek WW. If you can be patient and wait until you upgrade to an A64, it works out much better in the long run to just get the A64 version of the MCW6002. You'll be glad you did.

heard this comment from another forums...any comment on it???
--------------------------------------------------------------------------------------------------------
Another thing is the design of the two blocks(MCW6002 & TDX). Sustained overclocks wise, I believe the DD TDX may very well do better than the MCW6002 due to the "impingement" design reflected in the usage of distributed nozzles, along with the distributed cooling effect of the nozzles. This has been shown as per the Cascade to have an effect on cooling & o/cing of cpus. The other way around it is to have a very thick copper base which the MCW6000 series has done away with and which the MCW5000 series actually had. The thick copper base aids in evening out the localised heatspots and heatspikes as experienced by CPU cores and does aide in o/cing too somewhat.....
--------------------------------------------------------------------------------------------------------

I believe Phaestus's testing results in the link above should tell you what you need to know.

it not about temperature only...but also the ability of the waterblock to reduce heatspots as well as heatspikes... not really sure if this is true that TDX handle it better...

The design of the TDX w/ #5 nozzle and the cascade are not as similar as people believe although similar LOOKING. The #5 nozzle results in a better dt/flow curve, but it's a lot more restrictive so that actual performance in a loop is similar to the #4. What I mean is that the #4 and #5 perform about the same because the #4 yields higher flow rates. The #5 also has the problem of easily getting clogged up with debris in the loop. Combine these two facts and the #4 is a better choice.

The RBX and TDX from Dangerden are really tweaked Whitewater designs and do not share much in common with the "jet in a cup" design of the Cascade and the Storm other than "holes at the inlet".

If you look at the MCW6000, they did decrease bp thickness compared to the 5000 series. They also put a central inlet in that block to get some die area impingement too (just as the DD and Little River blocks do).

Here's a kind of interesting post on my testing of all the Danger Den and Swiftech blocks I have:

http://forums.procooling.com/vbb/showthread.php?t=10872

It's kind of a moot point anyway unless you want to buy 2 waterblocks. The MCW6000 that fits the Socket A is not compatible with AMD64. So you're left with WW or TDX and I'd guess the TDX is easier to use but requires a separate top for AMD64 and the WW is more of a hassle to mount but cheaper.

there's more to it than temperature- heat distribution also counts when factoring in stability when overclocking

first, the purported descrsption of the wb's baseplates is just incorrect
see where this is headed ?
who gives a flying f what that author "thinks", stuff it and him for being an ignorant ass
- for reference, which commercial wbs has this author designed ?
got any TEST RESUTLS to substantiate the patter ?
another internet expert

He he, simply gotta love Bill's comments in situations like this. You just brightened my day man.

Overclocking-wise, getting the base-plate as flat as possible, and mounting the block as flat and as firmly as possible against the CPU, tends to have the most effect before the block's internal design comes into it. Addressing the mounting inconsistencies and base-plate flatness should be the first port of call for anyone who wants to assess the overclock potential of a design.

A casual observer attempting to predict visually which design allows for better overclocking is fraught with peril. There are a great deal more aspects to it than simply saying "block X has impingement", or "block Y has a thick base-plate".

thanks bill and cathar for the info :)