Printable View
You could cool the die directly by using another fluid than water. Gun oil is a great non conductive coolant, but pricy. In Denmark it's around 800 USD for ½ liter, around 800 USD for 1/9 of a gallon.. Tom's hardware made and oil cooled system by lowering the entire system into a bath of frying oil..
Viscosity and conductivity are not at issue for direct die (actually viscosity is, but thats a more general problem not limited to direct die).
What is then the issue? I thought the main problem was conductivity, viscosity and how the fluid used infectets the die, does it corrode the die etc..
It's very difficult to seal the area around a direct-die setup, and then you have to deal with different cpu's being different sizes, etc. Which is why there hasn't been any comercial water blocks like this.
Seal isnt actually that much of an issue, just different die sizes and your injector style.
Well in case anyone wants to play with diamond powder/diamond slurry as a thermal compound or try molding some here is a source
http://www.smg-diamond.com/index.html
and a link to masterbond for the thermal epoxy
http://www.masterbond.com/tds/ep34an.html
Interesting. He said some things that made me doubt him. Then he said some things that make me hold my opinion until we see results.
All in all very interesting.
When does he plan to have it finished?
Am I the only one that LOL'ed at this picture from Xoxide?
http://lib.store.yahoo.net/lib/xoxid...ater-all-2.jpg
And let's not forget those horrible looking BGA ramsinks that this guy makes too... I find it hard to believe that he's actually created a block that performs up to his big mouth. We'll see.
whoa looks different :DQuote:
Originally Posted by FragTek
....crist :rolleyes:
that water cooling loop makes my loop cry. it cries for the pain its fellow brother must endure, all that un-needed restriction. we must liberate that poor pump from slavery!
i think WhiteBoy88 misunderstood my point in this postQuote:
Originally Posted by AzraelDarkangel
http://www.xoxideforums.com/916793-post134.html
i know this guy's NOT a 15yr old boy....LOL
the second point is more important................................why would you use wording like that on a forum................we (XS people i guess) don't go around shouting from rooftops we are the best or that some tweak i've been testing will take Fugger down......it's just pointless yacking if there is nothing to show..................credibility is questionable due to his possible motives or what he can gain from drumming up hype about his product. We all welcome better ways to cool CPUs but that's it it has to be proven to be worthy to make such extraordinary claims.
We are far from myopic here. It is just hard to listen to the bloat without any proof. Please provide proof that you can keep a LOADED CPU at ambient temps. It's very important to have skeptics around otherwise where are the checks and balances for bull:banana::banana::banana::banana: artists..............NW may not be one but he has to have proof to stay credible. I will endorse and accept his block if it really is as it's said. If it isn't he will get a massive serving. That's the cost of lying to people.
Let us see some proof................DON'T show us pics but results for now................preliminary at least.....when you patent the invention i hope you'll be as open as Stu. He is enormously respected because he doesn't talk :banana::banana::banana::banana:. The guy knows his stuff and can back up everything he's confronted with. Maybe MW should take a leaf from his book and act appropriately so that I am not forced to make such unfortunate conclusions about him.
i've said too much here
read this line again.......i mean wtf is that all about LOOOOL.....how can i keep a straight face while reading this garbage. What is the average age on Xoxide forum...
i think he's practiced this in front of the mirror ;)Quote:
it's not as far away as you think... the Storm is over... watch and see the future be unveiled!
http://img97.imageshack.us/img97/595...driver3lc6.jpg
Yours is just pegged? Mine detonated and left a 100yard blast crater of BS.Quote:
Originally Posted by ghent915
Its called heat flux and it sucks, heat source will ALL WAYS be hotter then the cooling source due to heat stratas forming (for thoughs who know just replace strata with flux, I figured strata would be better understood by most)
you have no idea what you are talking about mate, I had to fight with a science teacher of all people, took him 5 years to admit he was wrong but here we go again: Metal feels colder due to it absorbing heat faster then skin or air, thus as it picks up heat it rejects it at the other end thus, placing a thermomiter on metal or your finger; The metal will feel colder to the tuch then it is in reality, what you are feeling is the metal taking in the heat and moving it through out its mass and rejecting it to the air, if the air where 100% static and no convection, the metal would saturate and messure room temp perfectly. Allas it will be as hot as the coolest air touching it is.Quote:
Originally Posted by mnewxcv
I didnt read much about the topic, but I wouldnt say its impossible to build a better block than the storm.
In fact there are already a few blocks out that perform better than the storm. I have seen several reviews in german hardware magazines. Its been a while,so I dont know exact numbers, but the storm was outperformed by some German coolers, by a few degrees. Test were made on a dual core AMD and a IHSless AMD. I will try to dig up the test and post the numbers.
The waterblock that had the best performance was the Alphacool Nexxxos XP, which I own now.
Futhermore using liquid metal instead of AS5 resulted in 6C better load and 10C better idle temps on my naked S754 Sempron 2800+@2800Mhz 1.55V.
So I think there are still some improvemts that can be done over current designs. Maybe that guy had a brilliant idea. Good concepts dont have to be complicated.
I dont believe everything he says. A CPU cooled by water having ambiant temp is not possible at all in my eyes, but his block might be able to perform slightly better than a storm.
I dont have high expections, but its a bit too early to make fun of that guy. We can do so after he has shown the results ;)
:rolleyes: When you are at the top of the mountain, like the Storm G5, there is always going to be someone trying to compare themselves to you to try to take you down. It's amazing how many makers of water blocks in the past have made claims that their blocks outperformed the Storm only for us to pick them apart and test them to find out that their claims were just a bunch of hype in an attempt to sell their product. I for one don't believe anything anyone says till I have seen it tested by multiple sources & the results substantiated in reliable forums. ;)
Quote:
Originally Posted by OCme
things is, most blocks perform the SAME as the storm.
people look at flow vs c/w charts, and see the storm has a better c/w at the same flowpoint as other blocks.
but they are too ignorant to realize that doesnt represent real loop conditions. as loop flow is directly related to block restriction.
and that doesnt even factor how gpu block perfomance goes down due to restrictive blocks.
the liquid metal is a thing that alot of people dont use cause it can be hell to remove.Quote:
Originally Posted by Fr3ak
and about the Nexxoss XP beating the Storm that mostly depends on what pump was used.
Nexxoss XP and Storm are both good blocks but their cooling can be influenced by the pump used.
It is very possible to make a block that cools better then a Storm but the claim of cooling 10 degrees better is what people dont trust.
That is why there is something known as C/W vs. Hydraulic Pumping Power. See the graph here: http://systemcooling.com/swiftech_apogee-10.htmlQuote:
Originally Posted by WeStSiDePLaYa
Quote:
Originally Posted by nikhsub1
that does not represent actual loop conditions. and doesnt even come close to representing multiple block loops.
so why you posted that, who knows.
do YOU even understand what those numbers mean?
take the flow vs c/w graph.
at 1gpm, the storms c/w is .05c/w. at the same 1gpm, the apogee's is .059c/w. that translates to a difference of .9c!
HOWEVER, that is at the SAME flowrate. in a real world loop, the apogee loop would have much greater flow due to the lower restriction. thus further closing the gap, if not completely eliminating it. not to mention with the storm, all other blocks in the loops suffer from the low flow rate.
EDIT:
look at this graph, it clearly shows the delta. at no point does the storm beat the apogee by more than 1c. not even a SINGLE DEGREE. but then when considering how the apogee will have much greater flow rate, that difference is nothing, literally.
and then you factor in gpu blocks in singles or sli. they add restriction, lowering flow, and the already low flowrate because of the storm restriction causes the gpu blocks to perform worse. where as for apogees to equal the pressure drop of a single storm, you would need THREE APOGEES.
so is it worth double the money of the apogee to have the same cpu block perfomance, with lower gpu block perfomance?
http://www.systemcooling.com/images/...image28big.gif
I posted it because it removes the c/w vs. flowrate. This graph shows which how a block performs with a given pumping power. So now you are saying water blocks should be tested in multi block loops?Quote:
Originally Posted by WeStSiDePLaYa
I don't think you know how to interpret the graph i posted. The graph is based on pumping power ONLY. So, forget about the c/w vs. flowrate for a moment. On the small 14mm sq die, at 3w of hydraulic power the storm would be 5c better than the apogee at with the same 3w. Granted this is a very small die, on the 32mm sq die the numbers would be much closer, perhaps 1-2c. The idea of this chart (again) is to take the c/w vs. flowrate out of the equation - this is sort of the ultimate graph to look at in terms of pure WB performance.Quote:
Originally Posted by WeStSiDePLaYa
Quote:
Originally Posted by nikhsub1
read my edit above. and then see how your own link proves more of what i am saying.
EDIT: do the math and you will find the hydralic power chart shows a 2.8c difference at 3w with 80w heat dump. where you came up with 5c i have no idea.
Right. But, It would also take 3.5GPM with the apogee to equal the storm at 1.5GPM... I'll be you anything it is far easier to achieve 1.5GPM in a loop with 3 blocks than it is to achieve 3.5GPM.Quote:
Originally Posted by WeStSiDePLaYa
Sorry I am assuming 100W. The more watts, the bigger the difference.Quote:
Originally Posted by WeStSiDePLaYa