[eddiecatflap]

..seems as if , eventually the water WILL escape , that's what i've read on many forums so far

..if in , say 2-3 years , the pipe cracks etc - how do you stand with apple ?

..could you argue the design is fundamentally wrong ?

[Maflynn]

Originally posted by eddiecatflap:
..seems as if , eventually the water WILL escape , that's what i've read on many forums so far

..if in , say 2-3 years , the pipe cracks etc - how do you stand with apple ?

..could you argue the design is fundamentally wrong ?

Without seeing the full design or a working unit that's been operating for years you (nobody) can argue that the design is fundamentally wrong.

Since water cooled units is not a new phenomenon apple hopefully took into account water (or what is being used) loss in the system.

Don't get me wrong, its a brand new cooling system that's pretty complex, I opted for the 2.0 instead - go with the tried and true.

Mike

[videian28]

this has also got me worried, kinda forces you into applecare...

[memory-minus]

Originally posted by videian28:
this has also got me worried, kinda forces you into applecare...

AppleCare is a must with any Revision A -- I am counting the dual 2.5 as a Rev A because of the complex cooling system. I wouldn't even think about getting a dual 2.5 without it, personally. However I just went with one of the "new' 2.0 systems.

[Luca Rescigno]

I am still pretty surprised that there is a water cooling system in the PowerMac. It just seems like the sort of thing that an amateur throws together - I wouldn't want a totally sealed-off one sitting in my machine, and I don't care who made it. Water cooling is for hardcore overclockers, not over-the-shelf retail systems. Nevermind the noise - I'd rather have a noisy system whose fans are working at max to cool it than a water cooler.

I guess time will tell if it ends up being reliable or not. I hope this doesn't happen, but I am afraid it's possible there might be a large number of failures shortly after people start getting their dual 2.5s.

[Spliffdaddy]

The same spring clamps are used in automotive applications at temperatures and pressures that greatly exceed those of a desktop computer - and they don't typically fail within 5 years...at which point your PowerMac will essentially be worthless, anyway.

The reason for G5 watercooling had little or nothing to do with noise reduction. It had everything to do with cooling the smaller die (.09). Without a lot of surface area for direct contact, the CPU becomes more difficult to cool. Until now, a large surface area from which to transfer heat had masked the inherent inefficiency of metals in that regard. Water is 10X better at transferring heat, when given the same surface area to work with. Simply put, the thermal transfer abilities of even the best metals isn't adequate for the G5's future needs. I think you're seeing an early 'beta' version of what's to come for the Power Mac. They could have made a heatpipe/heatsink solution...one more time. But they knew the effectiveness was nearing its limit.

Don't feel bad. The newest Pentium4 is in the same boat. 100watts of heat has to be moved through a one-square-inch contact area. It ain't easy to do using metal as the heat transfer medium. Watercooling will be the norm starting now. Being an avid overclocker, I have to stay one step ahead of the norm. And I do...

I can tell you that 4GHz ain't happening from the current .09 P4 using anything short of sub-ambient (sub-freezing, actually) watercooling.

[videian28]

I think you're seeing an early 'beta' version of what's to come for the Power Mac.

who wants to be a beta tester of a $3000.00 machine?

[Leonard]

Originally posted by Luca Rescigno:
I am still pretty surprised that there is a water cooling system in the PowerMac. It just seems like the sort of thing that an amateur throws together - I wouldn't want a totally sealed-off one sitting in my machine, and I don't care who made it. Water cooling is for hardcore overclockers, not over-the-shelf retail systems. Nevermind the noise - I'd rather have a noisy system whose fans are working at max to cool it than a water cooler.

I guess time will tell if it ends up being reliable or not. I hope this doesn't happen, but I am afraid it's possible there might be a large number of failures shortly after people start getting their dual 2.5s.

You contradict yourself, first you say a liquid cooling system is something an amateur puts together and then you say it's something hardcore overclockers do (overclockers are not amateurs in my book, they have to be highly technical). Liquid cooling is technology used by very professional computer makers. Even IBM makes computers that are liquid cooled - we had one of their liquid cooled mainframes in our computer facility. Never had a problem with it.

I doubt a liquid cooled computer is the same as a car engine. With a car engine you have other things that are putting pressure on the water hoses - the car moving (jostling, jarring, bumps), the engine violently vibrating, dirt, extreme heat, extreme cold, oil.

The only two possible problems that I see that could happen are:

- water splashes on the motherboard and drips down - a splash shield could be in place to prevent this. And since the motherboard is vertical in the case, water cannot pool on the motherboard.
- water drips down to the bottom of the CPU compartment of the case and then seeps into the power supply compartment - it looks like the bottom of the CPU compartment is sealed. Water may not be able to get to the power supply.

I assume if the water does leak and empty out of the cooling system the fans will accelerate due to the increase in heat and therefore the CPU will not be damaged.

[ink]

Originally posted by Spliffdaddy:
Don't feel bad. The newest Pentium4 is in the same boat. 100watts of heat has to be moved through a one-square-inch contact area. It ain't easy to do using metal as the heat transfer medium. Watercooling will be the norm starting now. Being an avid overclocker, I have to stay one step ahead of the norm. And I do...

So it's not Budweiser running through those pipes, is it?

[gururafiki]

Originally posted by videian28:
who wants to be a beta tester of a $3000.00 machine?

figure of speech

[djohnson]

Originally posted by videian28:
who wants to be a beta tester of a $3000.00 machine?

I would like to be one please

BTW, I highly doubt they have regular H20 in those pipes...

[Leonard]

Originally posted by djohnson:
I would like to be one please

BTW, I highly doubt they have regular H20 in those pipes...

The article here says it's 80% water and 20% other things.

[greenamp]

I think it's just the idea of water being in there that scares most people.

As an avid offroader, I work on my truck quite a bit , and can vouch for the type of clamp being used on that cooling unit. I dread coming accros them, b/c they are very tight and very hard to get off. (You need pliers unless you can bench press 100# with your fingers)
Even without the clamp, the hose is difficult to get off the pipe b/c the pipe is belled at it's end. So that combination on a low pressure system should last forever. I guess time will tell though.

[Leonard]

I just ordered AppleCare with my dual G5 2.5Ghz, I won't be using it for mission critical work and I back up my data regularly (even on my current G4), so I figure I'm covered. I doubt it'll break down, but any computer can have it's problems.

[nJm]

I'm sure it will be fine. Looking at car's cooling systems again, they are pretty rugged. The only times I've known pipes to burst/leak is when they have been neglected with age (think 10 years old). The rubber pipes used in your engine bay have to contend with extreme heat, bumpy conditions and oil. If you have an oil leak and that drips onto a rubber hose then it will start to 'cook' essentially and won't be long before you start getting leaks.

Think about turbo cars - they need the intercooler to keep them at some form of sane temperature. Turbos get hot. After a quick mountain drive in mid summer the turbo in a friend's Volvo XC70 was glowing!