SWIFTECH's new waterblock
 

Swiftech released a new WB design...

Take a look...

http://www.swiftnets.com/_borders/Mo...ifications.gif

http://www.swiftnets.com/_borders/MC...1-2_elbows.gif

http://www.swiftnets.com/_borders/Co...stall_asus.gif

More info here:

http://www.swiftnets.com/MCW462-U.ht...0compatibility

Flow gets reduced by a half when you hook them up like the bottom picture did.

All your fittings are belong to us?

ouch half flow would be bad given those fittings are for 1/4" ID tubing.... Can you say single digit gph ;)

I don't understand, if flow is as important as we think it's, these maufacturers (except DD and Fixxit ofcourse) don't seem to care about it... especially the European manufacturers, such as Innovatek... Does it cost extra to them to manufacture the same block with bigger fittings or is it because it's not too much important... ie. Blade Runner is using small diameter hosing in his system...

I'm planning to order 1/2 parts to see the diff. myself... (when i have seme extra cash to blow ofcourse)

http://www.swiftnets.com/_borders/internal_chamber.gif

How is -that- supposed to be a good design?

I don't even understand where the inlet and outlet are. lol

Now that is definately a 1/4" inlet and outlet...

I dont know how it's going to perform(probrably pretty well) but I can definately say that swiftech always seems to have a great fit and finish...

Inlet in the middle.

Since the distance between the inlet and the plate facing the CPU the waterflow might be quite high (same effect as the nanoblock). It will also create a lot of turbulence.

I'm looking forward to see it benchmarked.

What a shame on those fittings. This block appears to be the first to use design that would cause turbulant flow. The fit and finish of this block like all other Swifties is outstanding, and it is pretty cool looking compaired to the older rectangle blocks. I was thinking of drilling the "irregular pattern" into the base of an old MCW462 to see if it would add turbulance. Looks like Swiftech beat me to it. I can't tell from the pics on the page, but I bet one could replace those fittings with 1/2" barbs.

I personally think that Swiftecs design is pretty outdated.. its a new fancy case but the technology is weak. also I see a big problem with alot of air gathering in that block. Since the outlet and the inlet are so much lower than the top of the "tank" the area above those holes will just be air. no coolant will get in that area unless you tip your machine upside down to get the air out.

But then air will gather in there. Air in a WB does not help it at all.


Also I HATE those fittings... such POS's.. no offense LC, I know you love em, but they SUCK if you want to use silicone hose.

Never been convinced by the Swiftech blocks myself, they seem to develop it when problems are found by users, like the corrosion issues with the original, (so now it's all anodised). The open chamber is less than ideal even with the dimples, but they've made it harder to fit normal barbs as well. Worse still are these statements:-

Not only do most users not want these type of fittings, They cannot possibly be safe with vinyl tubing unless the vinyl has an insert, (which would further reduce the ID.)

Vinyl will Compress after a while forming a set shape, and if moved to a different position can leak. I'd also be unhappy if they could be pulled out fairly easily, like vinyl can if I try it in my push in fittings, which indecently look better quality than the Swiftech ones.

Although it looks quite nice from an image point of view, and well made, it's another missed opportunity for Swiftech to come up with something truly revolutionary imo. They seem to be always one step behind the game. I'd be quite happy to eat my words if it blows everything else away http://www.zytec.worldonline.co.uk/p...otions/LOL.gif. The point Joe makes about trapped air is also very valid :)

On the barb or fitting size issue, I still maintain that larger can be better but it wont necessarily be so, once you have enough flow rate more is not required, bit like fitting larger fuel lines from your cars petrol tank to the engine.

http://www.zerofanzone.co.uk/picture...ojja3small.jpg

I'm running this block I bought from NZ recently, (and with a few mods). Its running 1/2" barbs reamed out to 10mm ID. It's performing much better than my OCH Z3 block, a lot of which is down to clamping issues I think. After testing two OCH blocks one with 3/8" barbs one with reamed out 1/2" barbs (10mm ID) the best I could say is it was possibly 1°C better with the reamed out 1/2", but that's with the socket diode rubbish so is debatable.

The picture with the Duallie Asus board truely turns me on, the swivel fittings do have one use I suppose.

Otherwise, looks fairly poor I think

Unfortunately I know for a fact that the fittings cannot be changed out. Not without machining and tapping the holes.

If you check out swiftechs site you'll see the block without the elbows on it. The push-in type fittings can come in a "cartridge" format. This means that you can machine your part to accept just the push-in part of the fitting. In other words the fitting is pressed into the block. There are no threads at all.

I dont know why they did it this way. It would cost about the same to use threaded parts. It does make the block thinner, but then the tubing has to make a bend so your really not gaining anything. Also like Joe said they suck if you dont have the inserts with silicone tubing. I had thought about doing this, then realised what I was thinking was stupid :) Nobody can modify the block now without doing machine work...

If I got two I'd contemplate taking off the fittings, but I still like that duallie pic

I think Swiftech's main target is still air-cooling, so i think they're more concerned about how to deal with Alpha Pal8045 and other new heatsink stuff, than what's really going on in the watercoling arena. Such a shame but I guess that's the reality. But I still don't understand why they don't offer blocks with bigger fittings like 1/2".
There's definetely a big demand for it, so how come they don't do it??? that's a big question mark for me?

In response to this debate..

Preamble: unfortunately I won't be able to respond individually, but here are some general answers to the questions raised in this debate:

Re. 1/2" vs 3/8" controversy: someone said we are not concerned about liquid cooling, reason why we do not offer 1/2" blocks. This couldn't be farther from the truth. The decision to stick to 3/8" OD is a strategic one on my part. I want to promote liquid cooling not only among extreme overclockers, but into the industry at large. I feel that 1/2" tubing is too cumbersome. But being a hardcore overclocker myself, I also know that 1/4" OD is really too restrictive. Therefore, 3/8" represents a good compromise in my opinion.

Re. tube inserts: Tube inserts for use with vinyl tubing are provided with the block. The wall of the insert is extremely thin : 0.015".

All this being said, flow is really not as important as some would like to say; the design of the block is much more critical. I guess review sites will report their findings soon enough !

Air gathering in the block (in response to Joe): lengthy research, and testing is behind this block with respect to bleeding. See detailed bleeding instructions page 2 here: http://www.swiftnets.com/installation guide AMD.pdf

In standard single CPU config, bleeding is absolutely flawless. It get a little tricky with dualies, but really not a problem if you follow the instructions...

Finally, in reference to quick-connect fittings: as a user and overclocker, I simply love them for their convenience. No tools, no clamps, absolutely safe (when properly used, like anything else). They are used for most professional installations where there is a need for frequent assembly/disassembly. I feel that convenience is key to promote the liquid cooling solution in the public at large.

While liquid-cooling is becoming a more widespread solution, my priorities are unchanged: performance and quality first, with economics taking the back-seat. Yet, the reality is that economics cannot be totally ignored, and my design reflects what I felt was necessary to produce the best product overall.

Best to all of you!


Gabriel Rouchon
President, CEO

Swiftech Inc.,

Small point on tubing sizes, when we say 1/2" tubing it generally means 1/2" ID, similar 3/8" is ID. The swiftech by these measures takes 1/4" tubing which is VERY restrictive as most people run 3/8 or 1/2" (though I have heard of 1" setups!)

Gabriel,

I think Swiftech is heading in the right direction to help bring watercooling to the main stream. As someone who started manufacturing and selling waterblocks in 1998 (coolchip.com) I can say its a tough market to keep on top of. Unfortunaltly I did not have the financial resources to keep up with the R&D needed to stay competitive in the market.

As the speed of CPU's increase and the price drops, I don't think overclocking is as big as it once was. But with the amount of heat generated by the newer faster CPU's, heat is as big a concern as ever. I am more interested these days in reliable QUIET performance of my systems. This is why I water cool, for the silence....

These days as I convert all my systems to watercooling, I want it to be a clean and simple install. I want the components to fit right the first time. I also like a setup that requires minimal if no case modification.

Just my 2 cents.

Scott

"I feel that 1/2" tubing is too cumbersome. "
When I said 1/2", I meant fittings to use with 1/2"ID silicone tubing. I don't think 1/2" silicone hosing is too cumbersome, but it's just my opinion ofcourse... and BTW I don't like vinyl tubing and I never use it... even with your MCW462 WB that I use in my system...
but thanks for the input anyways..

Al Lallo

consumers love it when the manufacturers speak out! ;)

also guys, U have to relize that just about any watercooling setup is more effiecant then a HSF setup. So, while, like he said some of us are into preformance watercooling, some people like the idea of a system being quiet.

Swiftech has a good solid watercooling setup for someone in this market. And they are "PRETTY" too GRIN

I understand that everything is done with a budget, but if the block had threads then all the performance users in this forum would be happy and instead of complaining about the tiny tubing we could just put some more substantial fittings on it and carry on. That is what I did with an Innovatek block that I got and I don't hold it against the block that it came with 8mm fittings at all.

As far as the utility of larger tubing goes, I think most of us realize that when a block and the radiator and the pump are all tuned for optimum performance that the actual flow rate is not that important. But the Hayden is fairly restrictive of flow and the 1/4" ID tubing is going to hold things up a lot. It would be interesting to see the results with your waterblock and: a larger pump, 1/2" ID tubing, and a way to adjust the flow rate. I would be interested to see where the stock setup (0.2-0.5 GPM I am guessing?) falls in the grand scheme of things.

I think it makes sense to use a high flow block that will be close to optimum performance in the 0.5 GPM range because an Eheim 1048 is much more compact and 3/8" ID tubing is certainly easier to fit into a midtower. As far as 1/4" ID tubing goes I would rather use a smaller pump and 3/8 or 1/2 tubing than a larger pump and 1/4 to get the same flow rate personally.

they need to rethink their design, an open block like that isn't that good. plus the fittings are terrible, you have to use polyethylene hard tubing, which is not flexible at all. With a little redesigning they could have a great water cooling system.

I'd like to see the benchmarks for the block before I conclude that it is no good. It looks like it might not work too well, but the benchmarks should tell the tale.

Has anyone spotted benchmarks for this puppy?

-Ice-

High guys I'm new to this forum but I have to tell you that I got a Barbones from Gabe on Friday. Let me tell you I love that puppy, I'll be writing a review/article on it for Icontic.com. For those who don't know me I currently have the highest published Mhz for an AMD CPU at the Orb 3dMark2001. I hang with the Best of them with the Rig in my sig. Anyway back to the MCW-U. The barebones system is so quiet and so basic you would think it could barley pull off an overclock. I tested that bad boy on Saturday and much to my surprise I got a stable overcklock (Prime Proofed) at 1912 using 2.12v. with a XP2000 chip. Get this, the temps measured by MBM max and were stable at full load for 2hours at 46c room temp at 21c. Pretty dam good for a closed loop using 1/4ID, the Hayden didn't seem to have any problem at all with it :)
I personally use 3/8 I/D on my big rig, 3 loops takes way to much space already never mind having larger tubes running everywhere. So in defense of this product I can tell you that Swiftech blocks work just fine:D
Now to let you guys in on a little inside scope. Gabe has finally decided to design a Block with 1/2" I/d barbs. It's threads like this that has convinced him. Not that it will work any better but that’s just what a lot of clockers want. So there’s the scoop. Check out Icrontic.com for that review in about a week.

OPP