Intel "stock" Watercooling Solution by Sanyo Denki @ Toms
 

Interesting article on a fully sealed watercooling system by Sanyo Denki at Tom's
http://www.tomshardware.com/cpu/20041015/index.html

A possible response to Apple's moves towards water cooling?

no, when the market is percieved as large enough these products (will) appear
but a C/W of 0.37 indicates some ways to go
our H2O-120 kit has a C/W of 0.23 with 3dBA less noise - but it is not a single module, and much more than $99

I would hold judgemnet on this device untill someone else tests it.

Tom's methodologies always seem (IMHO) to be lacking.

Something's gotta give though. Web reviewers are reporting that the Prescott 3.6s run so hot that they constantly throttle down to performance ~ 2.6GHz northwoods in CPU-taxing apps (scientific stuff for example). Intel oems are going to have to find something cooler than the retail hs but not any louder if they are going to deliver the performance consumers want and meet OSHA/EU workplace noise level limits. Water starts to make sense at this point doesnt it?

Dunno. What does it cost intel per sink for their current stock ones? A few dollars? It must be cheaper to go to a design like the Thermalright XP120 and a 120mm fan than to do a water cooling system of any sort.

A C/W of 0.37 is much better then 0.51 even if it isn't optimal it is an improvement from stock HS...
At least it isn't worse...
What is the C/W on that heatsink is? I can't find any comment about that on thermalright's website

Overclockers.com rates the 120 at .16 with a high speed fan and .28 with a much lower speed fan...

http://www.overclockers.com/articles1043/


Cool unit, but does this mean intel has given up on making a cooler chip???

a CW of 0.16 is great but your fan will scream at 59 dBA :evilaugh:

I think the thought of mind with this product is good enough to work and not the best.

I wouldnt take that C/W as gospel, really. After looking at it, it seems pretty dubious. As their articles say 'Not comparable with others'.

(P4 diode? come on....)

From thread

"How do you compare this concept with the recent posts on low-end/low-noise/integration?

Many times in this forum the advantages/disadvantages of aluminum vs. copper have been discussed, but is this more frequent use of aluminum (Apple G5 also) in such setups the beginning of a trend?

Will it make a dent on copper WC systems manufacturers, or it will be like a "stock" cooler vs. a "heavy-duty" cooler relationship?"

Stock vs heavy duty
Aluminium is lighter, cheaper and easier/faster to machine than copper, and will always be the first choice in 'cheap' cooling.

59DBa doesn't mean the same thing as it usually does when it's measuraed at 8", Madhacker. The 0.16C/W is merely the results on a P4. It's the CPU case temperature, not the die temperature. I think it's accurate (for what it's measuring) although meaningless. I don't think that WC has a place in mainstream desktop computers, as something like an SLK 800 is good enough and cheap, and will be good enough for some time to come.

Also, "machine time" is not all that relevent, when what they do is cut up plates and solder them to the base.

I would not compare JoeC's C/Ws to anyone elses for several reasons

I suspect the Sanyo #s could be fairly accurate, there are industry (Intel) procedures - and for a few even hardware

their advantage is the ability to intregrate, and the capital to mass produce - a very fancy fan/bracket there

joemac, for sure - they have a specific price/performance point
and probably an OEM case its going into
I wonder how it 'got out' ?

Probably the same way as Apple's G5.

which means there is another OEM WCing box to hit the street
?
whose ?

Wild guess: Sony or Dell

hmmm
if I had such a product Dell would be a good target

On the other hand Apple approach towards WC seems (to me) to be a more consistent one (long-term?- if these words still have meaning in IT)) compared to the Sanyo Denki, which rely solely on Intel plans (not known for their accuracy).

So the question is: what kind of "move" is this?

what makes you say "on Intel plans"
I think the article used a Prescott as an example, and speculated about Intel
did Intel provide that module ?

maybe toshiba or fujitsu? They've been dabbling with wcing for a while. The japanese PC market switched to sff long before us users; maybe water becomes required sooner in the lil cases..

Isn't the emerging BTX form factor a move (by Intel) towards better cooling while keeping the noise down?

can't picture this in a small case, its considerably larger than 120mm - thats the entire backside of one of those cases

Got this link from another forum:http://sanyodb.colle.co.jp/pdf/sanacemc_li_e.pdf

I'm not saying its their design, but I think Intel had a part in creating this thing.

Based on the short tube lengths, (assuming thats a final product) I would think this is designed to be front mounted in a BTX system.

My guess is also Dell.

nice link Jag
BTX has 3 cases, this could only fit in the largest
I'd bet (again) on Dell, have to be very substantial volume to warrant that investment
HP is possible too

The new DELL systems based on 775 have a heatpipe based tower similar to the Thermaltake102 with a plastic shroud pulling air through the fins and out the rear of the case via a rear case fan.

http://thermaltake.com/images/cooler...ower102/03.jpg

Thats just ones that are currently available, right? (after reread: I will accept a stupid answer to this, since it is obviously a stupid question)

My belief in the BTX connection is based (poorly, I might add) on the short tube lengths, and my penchant for intake through the rad.

Considering the premise that Intel is planning on moving to BTX, and from a pure cost perspective, shorter tube lengths are preferable. Though the overall cost of this VS some new-fangled heatsink might negate that savings.

I was under the impression Dell had a special deal with Intel to ensure they wouldn't put AMD in their boxes. Add to that, heat issues with the newer CPUs leading to:
A. shorter product life-cycles.
B. heavy, loud air cooling.
C. relatively inexpensive, relatively maintenence free water cooling. (relative to other water cooling options)

Dell (or any PC retailer for that matter) wouldn't publicly admit to wanting shorter life spans ( though you and I know if they could get away with making people buy 3 new pcs a year they would).

I'm pretty sure they don't want to have to try and sell louder pcs to anyone, and BTX+conventionaly aircooling probably isn't going to cut it.

So to keep Dell from even thinking about using AMD, Intel might have needed to apease Dell with some (moderately) cheap, effective cooling.

So my thought is: A propriatary BTX case from Dell with the width to fit this thing.

Our company has a tier1 purchasing agreement with Dell. I'm an AMD kinda guy (espcially since last Oct and the whole 64bit thing. Good times.) So I've been talking to our senior account rep and asking about getting dual opteron CAD workstations. He won't even say the word AMD. Its something he just won't discuss, not even as a future plan. So, I'd be willing to bet that Intel has a nice deal with them.

I'd also lean towards Dell being the OEM. The requirement for faster business PCs coupled with the OSHA low noise requirements will kill them otherwise. Maybe the heatpipe idea will work... maybe. I just have a bad mental picture of a dust covered heatpipe and a roasted cpu.

Intel has defined BTX in 3 form factors; micro, workstation (flat desktop w/92mm fan), and a biggish tower
there is a paper on it somewhere (sorry, no link)

I don't think Dell wants loud computers as much as they want cheap computers
as long as heat pipes will work they will use them in preference to water
and the only loads that 'might' need water are the EEs or such, I doubt that is a big market for anyone really