[zubro]

see that guys!

http://www.theregister.co.uk/content/3/33252.html

I want one! I want one! ;o)

[DekuDekuplex]

Here's the relevant quote from zubro's posting on The Register, a British source:

"The Apple connection is interesting since Cooligy's could play a major role in allowing the company to ship G5-based PowerBooks. The 64-bit CPU requires a major computer-controlled cooling system in its desktop incarnation, rendering it effectively useless for mobile applications."

--DekuDekuplex

Originally posted by zubro:
see that guys!

http://www.theregister.co.uk/content/3/33252.html

I want one! I want one! ;o)

[parsec_kadets]

What do you mean finally? The G5 hasn't even been out that long. Heck, people didn't even really start getting the 2GHz ones until just a couple weeks ago. Also, there's no chance in hell that I would buy a liquid cooled laptop. There's just way too many things that could go wrong with that, not to mention that it would probably cost around $4,000. Apple knows this too. One of the news reports out there mentioned that Cooligy was working on 1U rack servers and work stations. To me that spells XServe, not PowerBook. Also, don't forget the reports that IBM has made G5s that draw dramatically less power than the current chips. Those are the CPUs that will be used in the PowerBook. Apple would have to be nuts to use the current ones in a PowerBook.

[Super Glitcher]

Originally posted by parsec_kadets:
Also, there's no chance in hell that I would buy a liquid cooled laptop. There's just way too many things that could go wrong with that, not to mention that it would probably cost around $4,000.

I doubt anything would get released that was problematic. This kind of technology has to get extensively tested before it ever even gets to a prototype model let alone in your hands, so I wouldn't be worried about it. Whatever tech and engineering tricks they use to get it into a powerbook will probably be rock solid. Check out the size of the heat sync in your G5, I'm pretty sure they will use a tamer 'mobile' version of the G5 when it eventually gets there and it will likely be expensive (that didn't stop me from buying the 17" G4 though, after ram upgrades it was $3650 - $4000 for a tuned G5 I wouldn't blink twice)..

Let alone the brag factor.. My laptop's so fast it's liquid cooled... sounds cool to me

[michael_on_mac]

Sounds silly to me.

Look at P4-Laptops - no watercooling whatsoever. From my point of view, the main problem is about power supply. There are surely ways to dissipate >30 W in a notebook design, but few ways to provide >30 W*5h (to challange Pentium-M (Centrino) machines).

BTW, I feel like Apple did a big mistake by reducing the battery capacity with the new Al-Books, just when people start seeing the efficiency of the Pentium-M design. Casts shadows on the old Apple-PB nimbus of real portability...

Just my 2ct,
Michael.

[zubro]

max respect to all of you!
You sound like you know quite a lot more than I do regarding this stuff... was just a suggestion!... (no please stop bashing me! Pleeeeeeeese!)

But anyway, can t we agree that the G4s will be replaced "soon" in the PBs???

[Fellow2000]

But anyway, can t we agree that the G4s will be replaced "soon" in the PBs??? [/B]

Not bashing or trolling you at all, but I highly doubt that you will see something else in a powerbook other than a G4 anytime 'soon.'

I know that there is supposed to be an update to the ibook soon, if it happens and they put g4's in them, then maybe ya, in a year or 2 we might see g5 powerbooks, but if they just update them with a faster g3, i would see the powerbook staying g4 for at least 1 and a half to 2 years.

[xylon]

The actual design of that Cooligy plans to use is incredible. Solid-state electro-kinetic pump? I'm having trouble wrapping my noodle around that one, but it makes it sound like it'd pose less of a problem in a PB than a pump with moving parts.

I definitely think we'll be seeing this in the Xserve first and in the PB last, if we even see it at all. Just so many more problems that must be accounted for in a PB, but it would be so awesome to have a water cooled PB.

[parsec_kadets]

Originally posted by Super Glitcher:
I doubt anything would get released that was problematic...

Are you kidding? How many complaints have we heard from PowerBook owners over the years? Off the top of my head I can think of the CD eject problem, excessive heat, peeling paint, dead pixels, and hinge problems. All of these problems happened to a very small number of people when compared to the number of PowerBooks sold. And yet everyone who has been around since the original introduction of the Ti remembers what a big fiasco each one was. Now, if Apple will release a computer where the CD drive was installed BACKWARDS, that tells you that something could go wrong with water cooling, no matter how much Apple tests it.

Originally posted by zubro:
But anyway, can t we agree that the G4s will be replaced "soon" in the PBs???

No. I'm pretty confident that the earliest we could see a G5 PowerBook would be August 2004, and even that's pushing it. Personally, I'll be elated if we see one before May 2005. I won't start to be disappointed until September 2005. Call me a pessimist if you like, but I feel expecting a G5 PowerBook "soon" goes beyond unreasonable and into unbridled silliness.

[slider]

I would say predicting a G5 in a PowerBook in a year sounds about right. I would be shocked to see watercooling, but Apple has often done that to me.

[Super Glitcher]

I'm not saying there won't be a few localized issues- there are ALWAYS bugs and messups. I'm saying if they take the water cooling route, it will be tested and it will work, not that every single unit will function perfectly- that's retarded.

Joz implied in a recent interview that the avenue for production was clear, but there was some work to be done.. so I'd guess mid next year.. no sooner!

[Cincojoe]

With all the problems thay are having wih the new AL books, I can only imagine the issues with a water cooled G5...lol I can see the boards now. Instead of people complaining about how hot their laptops are on their laps, they will be complaining about the wet spots their G5 left... or threads like "How many drips per minute are you getting?".....lol

[SplijinX]

If you read up on the company's info it sounds like they've been working closely with the manufacturers and are likely to be in a prototyping stages. According to their press release: "The company will begin supplying qualification systems to computer systems developers and manufacturers later this year."

So Apple at the earliest will get something close to a finish product by early next year and work on refining it and adapting it to suit their needs.

Check out there press release here:
http://www.cooligy.com/images/Coolig...ot%20Spots.PDF

[cdhostage]

I wonder how this electro-kinetic pump works.

[stevesnj]

If IBM can't get the G5 to work without water cooling than it's time to get Motorola to do it. Water cooling is just asking for trouble.

[zubro]

Well, no to go in details but I think that Apple surprised ALL of us about a thousand times now, in hardware and software! ;o)

So I will offer myself this dream of a water cooled G5 PB! :o)

...

[michael_on_mac]

Originally posted by Fellow2000:
Not bashing or trolling you at all, but I highly doubt that you will see something else in a powerbook other than a G4 anytime 'soon.'

Sorry, no offense meant from my side. I get sometimes a little bit harsh...

But anyway: The point with G5s in PBs is, in my opinion, not about cooling down that "insane" power (marketing !). Keep in mind that if you only reduce the clock speed the power requirement for the G5 gets equally smaller. G5s in PBs is not prohibitive by design! Also think about IBM moving the G5s to a 90nm process or beyond, which would also reduce the power requirements. And if they increased the cache further (as we have seen with the recent PB-G4s and Pentium-M) and had a real DDR-interface to RAM, the major bottleneck would be removed, giving more power at lower clock speed.

So, from my point of view, there is no reason why not to use the G5-design in Powerbooks. Not the chips that are in production right now, of course.

Best,
Michael.

[nobitacu]

Well... ok, watercooling in the Powerbook G5? So... does that mean we have to fill up the water from time to time in this thing? How would this work? Weird...

I have a watercooled PC system at home in which I don't even use anymore, my parents are using it now. But I can't figure out how this would work by putting that into the CPU.

Ming

[SplijinX]

I'm guessing that you won't need to refill the machine with water, it will most likely be recirculated so users don't spill it all over the place and damage internal components.

There's some "heat collector" built onto the processor which water probably passes over to cool the processor and then the water is sent to a radiator to get cooled and starts all over. Here's an image on Cooligy's site: http://www.cooligy.com/images/Cooligy_loop_merged.jpg

[Simon]

Originally posted by nobitacu:
Well... ok, watercooling in the Powerbook G5? So... does that mean we have to fill up the water from time to time in this thing? How would this work? Weird...

Hardly.

The water-cooling isn't about creating steam that can be vented and thus cooling the CPU. Who the hell thinks water should get lost? It's probably just an efficient way to carry energy away from the CPU core and inject it into something with a lot more surface. In the end there will always be a passive cooling.

[nobitacu]

Originally posted by SplijinX:
I'm guessing that you won't need to refill the machine with water, it will most likely be recirculated so users don't spill it all over the place and damage internal components.

There's some "heat collector" built onto the processor which water probably passes over to cool the processor and then the water is sent to a radiator to get cooled and starts all over. Here's an image on Cooligy's site: http://www.cooligy.com/images/Cooligy_loop_merged.jpg

Yea... but I still don't see how this would work that we won't need to refill the water. Even if you recirculated, sooner or later, the water will evaporate itself... I don't see how this thing can work out in a Powerbook...

Ming

[Simon]

Originally posted by nobitacu:
Even if you recirculated, sooner or later, the water will evaporate itself...

Why should it? It's in a pipe.

[nobitacu]

With the heat when going through the CPU, unless they can somehow make a way to conceal the pipe and the cpu really really well that the vapor can't escape from it.

With the watercooling system I have, even though everything is all sealed up and in a pipe as well, we will still have to refill the water once every half an year or so... it takes time for the water to vaporize, but... it does...

Ming

[Super Glitcher]

If you read the article it says "water, or any other fluid for that matter" - I'm sure there are plenty of water alternatives that would not evaporate. BTW don't you think they've already thought of this? As if they would release a pb where you had to service the cooling system on a regular basis....

[michael_on_mac]

If you want to see some liquid cooling, AFAIK (only read about it) some of the latest MDD Powermacs should feature two heat pipes through the copper stacked above the cpu(s)...

<theory>
Filled with some fridge gas that evaporates (=consumes large amount of energy) near the CPU-die and liquefies (while dissipating the heat) near the outer parts of the copper cooler, they efficiently transport heat. The liquid then flows back to the die by capillary force...
</theory>

What about that in reality?

Best, Michael.

[Karim]

This will work like a good old fashioned heat-pump (IE an air-conditioner). A fluid will pass througha pipe and go through the micro-channels in the cpu where it will heat up and possibly vaporize and will then continue to pass through a pipe that leads to a radiator where it will cool down and return to a fluid and then go back towards the CPU. It is a closed-loop system so there will be no fluid loss. (just like in an airconditioner)... Yes, the system may develop a leak and need to be recharged at some point.

[SplijinX]

Here's another short article on all this by Wired: http://www.wired.com/news/technology...,60733,00.html

[Eug]

This is an advance, and it's not the same thing as what we have now, but I will point out that some of our laptops already have heatpipes.

However, the key will not be power dissipation by the heatsink. The key will be reduced power production by the G5 PPC970 CPU. That will come with the tweaked design and die shrink in 2004.

Not much point in cooling a 40 Watt CPU only to have your battery last only 30 minutes.

[xylon]

Originally posted by Eug:
However, the key will not be power dissipation by the heatsink. The key will be reduced power production by the G5 PPC970 CPU. That will come with the tweaked design and die shrink in 2004.

Mmmm, 90nm, what a beautiful number...

But lets say that IBM (and all chip makers in general) are able to achieve 90nm technology, can't they just overclock the chip (and use something like the Cooligy system to chill it) and achieve even faster speeds at roughly the same power consumption we're seeing now?

It seems like the Cooligy system occupies a nice little niche in either 90nm or conventional chip design, or am I missing something?

[Eug]

Originally posted by xylon:
Mmmm, 90nm, what a beautiful number...

But lets say that IBM (and all chip makers in general) are able to achieve 90nm technology, can't they just overclock the chip (and use something like the Cooligy system to chill it) and achieve even faster speeds at roughly the same power consumption we're seeing now?

It seems like the Cooligy system occupies a nice little niche in either 90nm or conventional chip design, or am I missing something?

The problem is that an "overclocked" low power chip becomes a high power chip.

Regardless of how well the cooling system works, the problem is heat generation. ie. The heat is being produced because more power is being utilized, at the expense of battery life.

Cooligy would be a bonus, but it doesn't address the primary, more important issue.

Not that I have any insight into this, but my guess based on superficial information is that we can have a 1.5-1.6 GHz G5 at 90 nm by 2004H2. However, I'm guessing it would use more power than the 1.7 GHz Centrino.

[michael_on_mac]

Originally posted by Eug:
The problem is that an "overclocked" low power chip becomes a high power chip.

Regardless of how well the cooling system works, the problem is heat generation. ie. The heat is being produced because more power is being utilized, at the expense of battery life.

Absolutely. After all, it's only about very basic physics...

Now, there are mainly two ways to increase life on battery: (a) by decreasing the power consumption, (b) by increasing the battery capacity.

More than once, we have seen nice innovations with batteries (NiMH, LiIon). Still I guess progress is easier to achieve fast results along path (a).
-- at least as long Apple does not decrease the capacity, as done recently !

We have seen Pentium-M, so I expect efforts at IBM to increase to get something that can stand against it!

Best, Michael.

[zubro]

I am a fool so I will continue!
fuel cells battery in a laptop?...