water cooling

[HardWarrior]

Going though the added hassle of firing two blocks into buckets, against 0 head, won't represent anywhere near a real test, GM. You know that. Coolant WILL seek the path of least resistance and favor it to the detriment of the path with most resistance. So, a "small amount of pressure in both lines" is enough? Again, all contemporay GPU blocks have some sort of impingment structure. This means that a small amount just won't cut it, if you want any sort of performance, that is.

Let's move this out of the realm of theory, shall we? How many of you people actually USE Y connectors on SLI'd cards? Is so, what sort of temps are you getting? Please list your room ambient as well.

 

[HardWarrior]

Originally posted by: aigomorla

Originally posted by: Greenman
A Y splitter shouldn?t make any real world difference in flow rates. If the blocks are matched, the flow should be within a few percent of identical, as long as your pump is strong enough to keep a small amount of pressure in both lines. It?s easy to test, just hook it all up outside the case, turn on the pump and let each gpu block dump into an empty can. Check the water level in each can after 2 minutes; they will be close enough as not to matter.

like i said, a Y splitter shouldnt be used unless u have a fairly decient pump. And if ur going to loop more then just CPU, your going to need a strong pump. Personally all the rigg's ive seen with SLI used a Y splitter. My Rig however doesnt include the vid cards in the loop. I didnt want to drop in a big and bulky 2 fan radiator.

🙂 You don't use Y's, but it "seems" like it'll work just fine, with a powerful pump? Great...

 

[HardWarrior]

Just what I thought. Some of you guys are MORE than willing to fluff yourselves at someone else's expense.

 

[Topplayer]

go with teh zalmen fanless..the tubes are much smaller and they make no noise because of the NO FANS rofl.....