The future of cooling

nodle

Administrator
Staff member
Member
Joined
Jan 28, 2006
Messages
42,670
Reaction score
1,474
Where do you see the future of cooling going for processors? With all the advancements in processor speeds, that usually means more heat. Air cooling is going to reach a max I believe even with the larger fans and heatpipes. Alot of people have gotten into watercooling, which when done right draws heat away much fast than air cooling. But I believe the future is with submersive cooling. You have probably seen this done with oil bathed computing. It's where you submerse your mainboard into a non-conductive oil and then cool the oil. Since it's non-conductive your electric circuits will not short out, theres for the oil pulls the heat away instantaneous resulting in excellent cooling. Now let's face it this type of cooling is messy and also dangerous if not properly done. But I foresee systems slowly switching over to watching cooling, then in the distant future slowly moving over to submersive cooling. Again this is just my speculation on where cooling technology is going. The other option is they come up with nano cooling which then requires no heatsink at all. If your interested I will post some pictures I have at home of some submersive system that amatuers have done.
 
I have seen this and it is pretty interesting. I just don't know the real need for it though. If water cooling is efficiant enough and in it's early stages of being used then why go inot anything more advanced and complicated?

ANother thing they are trying is making processors that use less energy and disperse less heat. Either way, cooling is very important and the cooler the better. I use water cooling now and love it.

 
I have seen this and it is pretty interesting. I just don't know the real need for it though. If water cooling is efficiant enough and in it's early stages of being used then why go inot anything more advanced and complicated?
Because with the higher the clock speed, the hotter the chip (usually). So there will come a day where even watercooling won't be able to stand the high heat. Again though, if they would find a new way to change the way the chips are made, and get into nano cooling etc. then there would be no need.

 
Here's what Anandtech had to say on the subject which may be pertinent:

In the past 15 years, architectural improvements have made sure that the Pentium 4 issues and retires about 6 times more integer instructions each clock cycle than an Intel 486 could on average. At the same time, the die size would have been 15 times bigger if there were no advancements in silicon process technology, and even those aggressive advancements could not avoid the fact that the Pentium 4 needs almost 20 times as much power.

Clock speed increased from 33 MHz to 3800 MHz, so it is clear that clock speed, not extracting more ILP (Instruction Level Parallelism), has been the main reason why a Pentium 4 performs so much better than an i486.

However, the next generation of CPUs will be based on a completely different philosophy. The Xeon MP Version 2007, alias Whitefield, will have 4 cores, and run at speeds at around 2.6 GHz. At that speed, there are reports that it would consume less than 90 Watt. Intel will use its P-m know-how to keep the power dissipation so low. Each core is not really a P-m, but it is clear that the pipeline will be shorter than the one of Willamette, the first implementation of the Pentium 4s Netburst architecture.

AMDs K9 seems to be a slightly different beast. Andreas Stiller of Ct reported that this Quad core CPU monster would have a TDP of 140 Watt, and run at about 3 GHz.

So, it seems that clock speed will no longer drive performance, but higher IPC and more cores will.

 
Don't you think they will need to increase the core speeds down the road though? I mean you can put 4 cores or even more on a chip, but basically your making your system a multiprocesor system. In time even if you got up to 8 cores, there gonna have to up the speed of the processor. But again with all the changes in technology they may go to something totally different. Again I was just basing off of the technology now.
 
I can forsee many manufactures switching over to water cooling as the cost of such cooling systems drops and the need for more efficient cooling rises. I've also seen pics of a submersive machine and it is an interesting idea. I'm not sure that it will become mainstream, or even close to popular for quite a while.
 
Another question one must ask is processing power and speed the only thing that needs to be analyzed when performance is concerned? What I mean is how about faster hard drive solutions and faster ways for data to be sent and retrieved throughout the whole system. I know they are working on ways to speed up everything but as a whole, processing power is only a part of the performance equation.

How about the software vendors as well? What I mean here is better coding that is more efficient and utilizes the system resources better. That way, systems and programs run better without having to always depend on systems with the biggest specs. I know this is hard to control but maybe should be looked at a bit more.

 
I can forsee many manufactures switching over to water cooling as the cost of such cooling systems drops and the need for more efficient cooling rises. I've also seen pics of a submersive machine and it is an interesting idea. I'm not sure that it will become mainstream' date=' or even close to popular for quite a while.[/quote']That's what I use to think of water cooling back in the day. But more and more companys have started producing kits now that even the average joe can install safely (well almost).
 
Nice input Corey. Instead of changing and realing on hardware, make the code perform faster and more economical. It makes sense. I also agree with the storage, they need to start to move towards ram based harddrives or solid state drives, for instant data transfer. Instead of waiting for a hard drive to spin.
 
I can forsee many manufactures switching over to water cooling as the cost of such cooling systems drops and the need for more efficient cooling rises. I've also seen pics of a submersive machine and it is an interesting idea. I'm not sure that it will become mainstream' date=' or even close to popular for quite a while.[/quote']That's what I use to think of water cooling back in the day. But more and more companys have started producing kits now that even the average joe can install safely (well almost).
I haven't looked into available kits lately, but I think right now the cost is still too high for manufacturers to switch. It is a little surprising that some of the high end offerings from each aren't available with at least a water cooling option. Right now though I think the manufacturer's need to offer their PC's at the lowest possible price is driving them to continue with strictly air cooling.
 
Well let's face it air cooling is cheaper, less dangerous etc. I agree about price point. But I know SONY has a water cooling computer out there. It came out a few years back. Also I think there was a version of a MAC that had also a sort of water cooling system in it, if im not mistaken.
 
In response to Corey's message I agree completely. As processor speeds increase there is an inevitable need for the speed of all other hardware to increase as well, since your PC will only really be as fast as the weakest link. Along with increased speeds there will undoubtly be an increase in power consumption and the amount of heat created.

As for software, there isn't one OS out there that is terrific at resource usage, and of course the software available for them isn't great either and varies depending who has written it. Unfortunately the software used by the majority of the populous is the worst at it (windows and software for it). I'm willing to bet that a new Mac will run warmer with XP than OS X. Most linux distros and OS X are more efficient at utilizing system resources in my experience.

 
One more thought on the code aspect of improving the over all heat performance is code bloat. If anyone has ever used Dreamweaver or GoLive to create a web page, take a look at the source and you'll see that there is far more HTML thrown in than is required to get the job done (the reason for this and my annoyance at it is another story all together that I leave alone to keep this concise) At any rate many development tools are guilty of the same thing. I can't say for sure how much extra/irrelevant code is thrown in by using different visual development tools in comparison to hand coding, but improvements to reduce the code bloat and force efficient resource usage will be a key factor in creating more efficient applications.
 
Everything should be written in assembly language. :wink:
 
Great another name changeover, I hated when AMD went to there XP series of chip. When we went to sell them customers were like" So this runs a 1.8ghz because it says xp1800?" Were like no the true megahurtz is actually lower but performs at that speed, because clock for clock AMD out performs Intel. Everyone always misunderstood that. I wish Intel when they have changed thier "P4" processor would have changed the name of it. I'm so sick of people saying to me "But it's a P4 chip it has a sticker on the front of my case" I'm like "They have changed it many times, a P4 is not the same as a new P4." :smt076
 
Everything should be written in assembly language. :wink:
Um how aboutmov A1, 4E

mov A2, 4F

or (in strict machine code... binary)

10100001 01001110

10100010 01001111

Either way it's a resounding NO (well i don't really care, as long as I don't have to do it)

P.S. that wouldn't really put NO out to the screen, but i believe it is legit assembly code syntax (it's been a while) and 4E or 01001110 are "N" and 4F or 01001111 are "O"

 
Assembly doesn't look so palatable. Is it not true that older games used to be written almost entirely in assembly? Steve Gibson (grc.com) used to write his utilities in assembly claiming speed and a small footprint.
 
Assembly doesn't look so palatable. Is it not true that older games used to be written almost entirely in assembly? Steve Gibson (grc.com) used to write his utilities in assembly claiming speed and a small footprint.
I'm sure that writing it in assemply would work toward a small footprint and speed (much like the old Gentoo stage 1 install where you bootstrap the system and compile everything to your computers specifications - power to console login to fully loaded KDE in under 60 seconds)As for how difficult doing it would be I have no clue, all i know is what i wrote before moves bits from one place to another in memory... I only stuck around assembly class a couple weeks before i decided to go to CIS over CS... and now look at me... programming again... woo hoo

 
Back
Top