Slashdot Mirror


Commercial Water Cooling, And Quiet

Aleaxander writes: "I followed this link about three weeks ago and may have found something I've been looking for for a long time; a way to make my PC's quieter while still keeping my systems cool. A commercially available case that is water cooled, and which according to the above review seems to perform well on the cooling front, while also running quietly? Will I be able to finally stand having an overclocked PC running in the room while I am watching TV, listening to MP3's or simply waiting out a long download while doing something else? I hope so. Anyone who has bought or tested one of these cases should post their impressions, because the specifications listed on their site probably don't tell the whole story (the chart on cooling in the Comparison section that uses a graph from Tom's Hardware, and tags their Ambient Temperature on the bottom scares me)." And really, there's no reason that water-cooled computers should be any more exotic than radiators in cars (well, except for the energy-chuggin' high power use they would imply).

45 of 128 comments (clear)

  1. They are nice and quiet by Anonymous Coward · · Score: 2

    My company bought two of these cases about three weeks ago. We had bought some AMD systems with fans that were *really* loud. At the time, we couldn't find any fans that would do the same job quietly, so we got these cases.

    They are very quiet, even when the fans kick in to cool the water (which isn't that often). The temperature gauge on the front usually reads about 98 when the computer is idle, and we can get it up to about 101 when we run CPU intensive tasks for a few hours straight.

    We aren't doing any overclocking, nor are we cramming as many hard drives and cards in the box as possible; they're just your basic desktop computer. So, while I can't tell you how good a job they do cooling in extreme situations, it's more than adequate for normal use and it probably makes less noise than the computer you're sitting at right now.

    (YMMV)

    1. Re:They are nice and quiet by baptiste · · Score: 2
      I believe the display on the front is driven off a sensor embedded right into the cooling paddle that sits on the CPU - so it indicates the CPU temperature itself (or pretty close) Better than those blue thermistors.

      --

  2. Re:Quiet cases with water cooling by Paul+Komarek · · Score: 2

    Getting your monitor to power down in X is different than getting it to power down at the text console. To get X to power down the monitor, include the "dpms" argument on the command line to X. For instance, if starting X with "startx" or "xinit", use "startx -- dpms" instead.

    If that doesn't help, well, I'm out of ideas.
    -Paul Komarek

  3. Quiet cases with water cooling by Paul+Komarek · · Score: 3

    I've seen a few posts saying that water cooling doesn't really reduce the noise enough to make it worthwhile, partially because of hard drive noise among other things. One advantage of water cooling over aircooling is that you can use the water to move the heat away from the computer itself (within reason).

    Imagine, then, acoustically insulating your case using foam, cardboard, carpeted wood, or whatever. With an air cooled case, you have just screwed up your air cooling because of too many holes in the acoustic insulation. With water cooling, you only need to allow the water cooling hoses out of the acoustic insulation, and they can go to a big radiator *outside* of the acoustic insulation.

    Of course, you will want a way to access the CD, floppy, and power switch. ;-) I would think that making the front of the acoustic enclosure into a door would work.

    -Paul Komarek

    1. Re:Quiet cases with water cooling by BlueJay465 · · Score: 2

      Wouldn't those noise dampening sheets that you can find for cars (ala Crutchfield) work in this scenario? They would be good since they are only 0.075" thick. And to line the inside of your PC wouldn't cost more than $50US

    2. Re:Quiet cases with water cooling by baptiste · · Score: 2
      I've seen a few posts saying that water cooling doesn't really reduce the noise enough to make it worthwhile, partially because of hard drive noise among other things.

      And there are others saying you can hardly hear the 3 small fans in this unit. Yes, the hard drives make some noise and that a fact, but every spinning device in your PC is adding to the noise level. By reduceing the fan count in your average geek box by at least 3 or 4, it makes a huge difference.

      Most review sites publish stats on HD noise so if you really want a quiet system you can shop for the best value drive with performance AND noise taken into account.

      So a system like this - for the price, is a godo deal if you enjoy peace and quiet even in the room your PC lives in.

      Of course if the vendors would ever get power mgmt working in a reliable and std manner - our PCs woudl shut down when we weren't using them without requiring a reboot :) I've used dozens of PCs and it always seems like some suspend nicely, others don't - and RedHat - heck I can't even get it to shut off my monitor (and yes I have power mgmt configured in my kernel!)

      --

  4. PSU Limitations by drix · · Score: 2

    On the Koolance website they say that the water-cooled power supply only works for systems under 1ghz and with non-DDR RAM, even though it's rated at 300W. For people who want to go higher, they reccomend an air-cooled PSU at... 300W. So it's not like the cooler itself is sucking too much power. Can anyone shed some light on this? I posit that no one who cares enough to buy these cases are going to be running under 1Ghz, either before or after overclocking. On one of the distributor's web sites they say only that you shouldn't use DDR with it, meaning I guess that 1Ghz is okay. I really want to buy one of these cases for the noise reduction, and of course the PSU is one of the biggest sources of noise. OTOH I must have the latest and greatest, which means a P4 in the 1.5Ghz range. Anyone aware of what thes technical limitations stemming from a using liquid cooled PSU (even of identical voltage) are, how they relate to RAM and processor speed, and any ways to flout them? Thanks!

    --

    --

    I think there is a world market for maybe five personal web logs.
    1. Re:PSU Limitations by drix · · Score: 2
      --

      I think there is a world market for maybe five personal web logs.
  5. Read the FAQ by Quarters · · Score: 2

    4) Q: I am interested in over-clocking my system. Would a Koolance PC help me with this?

    A: The current PC addresses our intended customers: the home and office user, who admire a quiet system with the added efficiency of liquid cooling. For the high-performance user, we do not feel our present system can accommodate the additional stresses placed on it by overclocking the components. Please check the news page for updates on our performance-oriented PC model coming in August.

  6. Re:First Impression by avdp · · Score: 2

    Looking at the resellers that sell these things... They have additional cooling units ($15-$25 extra for each) for additional CPUs, chipset, video card(s), and hard drive. That pretty much covers anything that might cooling.

  7. Re:Leery by Brento · · Score: 3

    Does anyone know what happened to those cooling systems that used a crystal that would absorb/emit heat as electricity was applied?

    I think you're talking about Peltier coolers. They're definitely gaining market share - they're on the rack at my local CompUSA, even. They go for around $25-$50. They work as advertised, but they use a lot of electricity. They're no quieter than a regular CPU cooler, though, because they still require a fan.

    --
    What's your damage, Heather?
  8. The next post we'll see... by gsfprez · · Score: 3

    ...is that someone in here has found a way to squish their computers into a 8 inch by 8 inch by 8 inch cube!

    Speaking for the Mac users...we all thought that it wasn't a real need to have a computer run silent? No, sorry let me rephrase that..

    We had it jammed down our pieholes that when Apple started making fanless computers, that it "was no big fucking deal, bitches" and why don't you worry about getting 3-fucking button mice instead and you Mac users give a shit about the stupidest things...blah blah blah

    Jessis - get your stories straight... i mean, if the poster wants a silent computer so that he doesnt have to hear the fan blowing all the time.. they call them iMacs and fucking Jobsian G4 Cubes... there's no need to bother with some weird ass hydro-powered case like it was a fucking Russian sub with James Bond at the tiller.

    And yes, it can run linux

    --
    guns kill people like spoons make Rosie O'Donnell fat.
  9. Isn't it the opposite? by GauteL · · Score: 2

    I may be wrong, but doesn't making things smaller actually _decrease_ heat output?
    Every time the transistor size decreases, it has been possible to increase the clock-rate, and decrease the amount of voltage required.
    One of the reasons Athlons will soon be viable as processors for laptops is because of decreased transistor size.
    I would guess that if Moore's Law DOES NOT continue, liquid cooled computers will become necessary.
    Why are processors hotter today than they used to be? Because the extreme competition has made the manufacturers increase clock-speed and add features without always being able to decrease the size of the silicon.
    Could someone confirm this, or perhaps prove me wrong?

    1. Re:Isn't it the opposite? by lucius · · Score: 2

      I also may be wrong, but I have the feeling that the relavant quantities here are also not the density on the chip and some function of the switching speed and the switching current. So while the transistor area may decrease by, say, a factor of 2, the switching current by a factor of sqrt(2) and the speed increases by a factor of sqrt(2), say. Then the heat given off would be proportional to the heat per transistor times the density, which will be the same heat per transistor as the old chip times double the density: so we get double the heat output.

      This is all on the absolutely shady side of things of course, I'm assuming that the current is proportional to the linear dimensions of the transistor, so it goes as the square root of area, which is the relevant quantity when calculating density.

      Also, I had to assume that heat is proportional to the switching current times the speed, and that the speed varies inversely with the current.

      But the point is, it is conceivable that heat output would increase with decreasing size.

  10. Re:Leery by Kartoffel · · Score: 2
    (And yes, I do realize that the primary CPU in a dual system generally runs under a greater load & therfore would require greater degree of cooling,)

    Not necessarily. In fact, unless you're running a poor SMP implementation, you shouldn't be loading one CPU heavier than the other.

    As long as you have at least 2 active threads (and decent SMP), there is no reason for CPU 0 to be loaded any heavier than CPU 1. Note that Linux 2.2 and FreeBSD 4.x don't really do SMP all that well. If you're talking about those OSes, then yes, you'll see CPU get loaded unevenly.

    IMHO, as long as you can switch context and toggle semaphores quickly, you're better off with zillions of threads. Sling those threads around all over the place and don't bother with funky worker pools and coarse 10 millisecond time quanta. Alas, most operating systems don't see it that that way :P

    FLAMER DISCLAMER: Linux 2.4 and FreeBSD 5.x both address SMP shortcomings and do things much more efficiently.

  11. Re:Inevitable by Polo · · Score: 2

    Whoops, you're forgetting something...

    This system replaces several fans: power supply, case fan, video fan, (hard disk fan), cpu fan, etc...

    But... Smaller fans have much more noise and move much less air.

    Actually, really large fans are quieter and can move prodigious quantities of air.

    So the best system would have 1 large fan.

  12. Re:Inevitable by Polo · · Score: 2

    Many hard disks have a "quiet mode".

    Here's some good info

    You give up some performance, but sometimes that would be worthwhile.

    Also, here's a bunch of reliability vs. heat info from pc power and cooling

  13. Re:So much for efficiency by Polo · · Score: 2

    Yeah, but then they'd need more machines.

  14. Re:But will it.. by -Surak- · · Score: 2

    (Yes, I know you were kidding, but it's a point that should be made early in this discussion)

    No, it won't cool your beer. This is NOT a phase change system like some of the systems posted here before. It just uses the (water-based) liquid for heat transfer (exactly like your car radiator), so it can't get the temperature below ambient room temp.

    A phase change "heat pump" system (like an air conditioner or refrigerator) needs a compressor, coils, and special fluid. It would hard to make quiet.

  15. Re:well now... by s390 · · Score: 2

    I once _had_ an (OK, used) RX7 Turbo that unfortunately developed a sticky fuel injector. I didn't get it fixed in time, so the car burned up one day.

    However, water-cooling systems are _much_ simpler. I wouldn't hesitate to water-cool a hot Athlon chip. Swiftech (www.swiftnets.com) markets wicked air-cooling heatsink/fan assemblies and water-cooling (block-only, pieces, kit, or bare-bones system) - with or without Peltier assist with water-cooling - that have gotten good reviews. I'm running an old 500 mhz Athlon at 700 mhz right now as I post this, with a Swiftech heatsink/fan and Peltier.

    Bear in mind, though - a Peltier won't help much at high thermal loads such as recent Athlons throw off. At most, you might see 5% chip efficiency improvement from adding a 155-watt TEC, and you'll have to dissipate over 180 watts from the system!

    If you want to really cool a system well - and quietly too - look at evaporative water cooling (think of nuclear plant cooling towers) or vapor phase-change cooling, perhaps with a liquid (50/50 to 80/20 water/glycol) buffer stage: cpusite.exmedia.nl/sections/steve/its-cooling-time .html . . . There _are_ ways to get your chip die down to -40 C.

  16. Re:well now... by s390 · · Score: 2

    Um, the actual reference is: cpusite.examedia.nl/sections/steve/its-cooling-tim e .html

  17. Re:Inevitable by overshoot · · Score: 3

    The basic rule is W=CF(V**2) -- the capacitance of the switching nodes times the switching frequency times the square of the supply voltage. Capacitance has been going up as oxide thicknesses have come down, along with more layer of metal. Frequencies have also gone up; you might have noticed. Voltages have come down, but not nearly as fast as the other two have gone up.

    My 500 MHz K6-2 ran on 2.2 volts; my 1200 MHz Thunderbird runs on 1.8 -- even if the capacitance had stayed constant (and it hasn't, by a long shot) the power would have gone up 60%.

    --
    Lacking <sarcasm> tags, /. substitutes moderation as "Troll."
  18. I'm about to order one of these. by NetJunkie · · Score: 2

    My case is too loud. I'm an overclocker and I've got about 8 fans in my case right now. It's just too loud as it is. I didn't really want to build a water cooling rig myself, but now that these guys have one integrated into the case I'm going to do it.

    Anyone know what the lead time to get one is right now? I'm sure they are busy with all the press they've been getting lately.

  19. cooler methods (IMHO) by joq · · Score: 2


    I posted this before somewhere but it applies more here... One way to make a nice little set up for your machines is to create your own crate like addition which could follow the typical dimensions of a desk. Now it would have to be enclosed at least the section where your pc is going to be placed which makes for cool security too if you want to add a lock to it. No one could remotely root it via booting in single modes...

    Anyways what you'd do after making the enclosure is insulate is with fiberglass insulation sold at a typical hardware store (dirt cheap), drill about an 4 inch slot into the side somewhere or the back if you have enough space or the side and attach a floor standing Penguino air conditioner along with a hose used for clothes dryers. The insulation keeps both the cold air in and noise out. Not only that but when your not using your PC's you could remove the hose and cool your house.

    New Blackbox Themes

  20. Re:Leery by ct · · Score: 2
    It almost sounds like your 'crystal' based is in fact a Thermoelectric (TEC or Peltier) cooler.

    Without getting into details here, you can read an intro about them here:

    http://thetechzone.com/articles/peltier_intro/inde x.html

    and how to make one here:

    http://www.thetechzone.com/articles/how_to/peltier _cooler/

    Once thing that I'd be cautious is running their dual CPU liquid cooler. According to the picture on the right of this page :

    http://www.koolance.com/Clickable%20Case/largeview /CPUjacket.html

    the heated "exhaust" coolant from the first CPU is directly looped into the intake of the second CPU, rather than having separate coolant loops for each. I'd be interested in knowing exactly what the difference in CPU temp was between the first and second CPU's when running full tilt.

    (And yes, I do realize that the primary CPU in a dual system generally runs under a greater load & therfore would require greater degree of cooling,)

    -ct

  21. Koolance Cases: Early adopter's experiances by rrhal · · Score: 3

    If you're interested in these cases check out this thread at ars-technica. They seem to be everything you could want in a water cooling system. Allthough the Liquid Nitrogen debate still runs hot.

    --
    All generalizations are false, including this one. Mark Twain
  22. Not now by Animats · · Score: 2
    Not after IBM finally got rid of all the water-cooled mainframes in their product line.

    As a practical matter, this cooling system doesn't offer cooling for the graphics board. Recent 3D boards consume more power than the CPU. Heatsinks on the RAM and fans on the graphics chips are common.

    Koolance also doesn't give info about allowable inlet air temperature, so you don't know what this is really good for. Inlet air temp specs today are lousy; most Compaq stuff has a 95F inlet air temp limit, which will be exceeded in non-airconditioned environments in most parts of the US. You really want a limit around 115-120F.

    1. Re:Not now by BlueJay465 · · Score: 2

      Yes they do offer a module for the graphics board. Look at their website again.

  23. WHO NEEDS THIS? by aiken_d · · Score: 5

    I RUN A DUAL PROCESSOR DESKTOP WITH 3 HARD DRIVES AND I DON'T SEE WHAT ALL THE FUSS IS ABOUT . MY SYSTEM IS VERY VERY QUIET. THE WEIRD THING IS THAT EVERYONE I KNOW SEEMS TO TALK REALLY QUIETLY, TOO.

    -b

    --
    If I wanted a sig I would have filled in that stupid box.
  24. Re:Water in a power supply.... by sik+puppy · · Score: 2

    If a tv transmitter can have a water/antifreeze based cooling system moving around in a 30kV environment, whats the worry with 120/240?

    --
    The first thing we do, let's kill all the lawyers. Shakespeare, Henry VI, Part 2, Act 4, Scene 2
  25. Re:Instead of Exotic, be practical by alleria · · Score: 2

    The Dynamat is a good idea, and so is the hooded vent at the back, but they hardly make a PC completely silent, which Home Theatre PC users and college audiophiles demand (I never turn my comp off, and most other ppl don't either.)

    If you want some more good hints and tips, head over to AVS Forum and the Home Theatre Computer Section, and do a search on the word "quiet." Lotta people talking about all sorts of various solutions.

  26. Re:Inevitable, not :) by -Harlequin- · · Score: 2

    The fact of the matter is that the fans have been moved one or two steps aways from their original locations. At some point the exchanged heat will need to be dissipated - and as of yet, I haven't seen a solution that doesn't produce noise. You're assuming that because there are still fans, the noise is pretty much the same. There are several reasons why this is not the case. A good PSU can be so silent you can barely hear it with your head next to it, yet it has a fan. How can this be? Fan noise is the product of many things, speed, and air turbulence over edges being the worst offenders. A larger fan, spinning slower, will pump more air, yet be much quieter than a small fast fan. This is why CPU fans are so noisy, while good PSU fans are silent. The design of the cpu heatsink obviously also affects the noise. In other words, fans don't preclude a silent computer, excellent engineering can overcome the problem. The case is question seems to be well engineered, and people say it is silent because when they power it up, they notice that they don't hear it. For the record, a cooling solution that doesn't produce noise is convection cooling. And this is possible to do with modern computers via a water cooling and a Big Fucking heatsink, or simply a 10 litre water reseviour :)

  27. I've seen these bad boys in action (Koolance) by tshak · · Score: 3

    I live 5 minutes away from the first vendor for these units (Infotech) so I've been researching the Koolance cases for quite some time.

    They are quieter then a normal power supply, but they still make some noise. They seem to be very well manufactured, just by eye-balling the unit. Definatly not a "hack job". I personally haven't bought one yet (my PII350 is cool enough thank you), but it's on my list if I ever go for those cool 1GHz Athlons that can be safely OC'd to 1.4GHz+. $180+ depending on configuration - but that's the price you pay for being cool (pun intended)!

    --

    There is no longer anything that can be done with computers that is nontrivial and clearly legal. -- Paul Phillips
  28. Re:Inevitable by boaworm · · Score: 2
    Im not really sure liquid cooling will be that widely adapted. Sure it reduces the temperature better then air-cooling, but how do you cool the water.. fans anyhow. There are of course alternatives to this, like huge metal wings in the air and stuff. So, it gives better (and more quiet) CPU cooling, but not more quiet

    My experience with watercooling is though, that even if its more effective, it doesnt reduce the "total" sound of the box. The power supply and the disks makes quite a lot of noice as well... There's no real need to cool CPU in a quiet way, if disks still makes noice.

    --
    Probable impossibilities are to be preferred to improbable possibilities.
    Aristotele
  29. Really quiet by Alien54 · · Score: 2
    Probably the best way to get things really quiet is to treat the noise problem as a separate isse from the cooling issue.

    Noise as an issues is like best handled by isolating the machine partially in a cool cabinet with sound baffles. There are folks who go overboard into the noise reduction. some recommendation I have seen include reducing the speed of your machine so that the computer can run without CPU fans. I have also seen power supplies that do not have and do not need fans because the design.

    Needless to say, you need to have a handle on your priorities.

    Check out the Vinny the Vampire comic strip

    --
    "It is a greater offense to steal men's labor, than their clothes"
  30. Instead of Exotic, be practical by krystal_blade · · Score: 4
    So many people today simply don't cross over into other areas when they are looking for something.

    The best item I've found (I'm not an overclocker, mind you) to quiet hard drives, and CD roms are the following items...

    1. Teflon tape. Wrap it around the head of mounting screws. It will act as a "threadlocker" for high RPM CD-Roms (which can loosen the screws through vibration.

    2. Thin rubber mat. See above.

    3. DynaMat, or another acoustic absorbing material.

    In fact, the ultimate way to quiet a PC is to make a "snorkle, over the top side of the PC. Take some sheet tin, available at any hardware store, use a razor blade, and a pair of tin snips, or heavy duty scissors. Leave small fins for mounting it to the back of your PC, line it with DynaMat, and you've just knocked out about 60 to 70 percent of your PC's noise, the cooling fans. Make sure you use a rubber, or teflon washer when mounting it.

    Line the inside of the PC with DynaMat, (leave room for vents, of course) and the only sound you'll hear from your PC is when you're swearing at it.

    krystal_blade

    --
    It will be easy to motivate our fellow man; there is hardly anything people treasure more than not being annihilated.
  31. First Impression by Papa+Legba · · Score: 4

    After reading the site and poking at the specs I have some misgivings, some of the specs seemed cooked in their favor to me, for one I noticed that the SCSI hard drive in dual fan case is listed as running at 138 degrees Farenheit, making their HD cooling solution beat it by 34 degrees. The fact is that ALL of my SCSI enclosures run at 125 to 130 degrees. They start to alarm at 130 and warn me of failure. The average is around 127 degrees across about 30 drives. This means that they are beating the HD fan solution by 13 degrees. This also makes me wonder if that is the case since they pushed the higher numeber up did they also push the lower number down, is the total savings closer to 5 or 6 degrees? With this noted, that they are playing with the HD cooling numbers, how many of the other numbers they list are being played with?
    With the fact that their solution can only handle 3 devices total this does not make HD cooling overly practical.

    I also noticed that their case did not contain any case fans for other circulation, with their solution only providing cooling for three devices (let's pick CPU, Power Supply and one HD) That means you have to air cool the rest. Their case does not seem to have good ariflow, and if I am having to add case fans then I am loosing the quieting fucntionalty that is desired.
    this looks like a good idea , and from reading the FAQ they have put some thought into this (especially in the area of condensation) but until they can handle more device off of the pump in the case I think I would personally give this a miss.

    --
    Papa Legba come and open the gate
  32. Re:Leery by aussersterne · · Score: 3

    They're called peltier devices, and they're still around. They're not popular because they require water cooling to work well with the current batch of CPUs.

    See, an AMD Thunderbird will dissipate as much heat as a sizable incandescent light bulb and do it in a very, very small space. Peltier devices aren't magic -- they still have to obey the laws of physics. Yes, one side gets cool, but the heat has to go somewhere. It goes to the other side.

    The catch is that the peltier device draws a lot of current, so it has to dissipate a lot of heat of its own. So, collect all of the heat coming off your CPU, plus all of the added power dissipation (a huge amount) generated by the peltier device. All of this heat will turn up on the "warm side" of the peltier, which really ought to be called the "hot as freaking lava" side.

    If you hook a 100 watt peltier up to your Athlon (which you'll need because an athlon may be dissipating 60-80 watts) and you don't water cool your peltier, you'll quite literally melt it to pieces and it will stop working within a matter of minutes or even seconds. Of course, once the peltier has more or less dissolved into little chunks, there will be no "cool side" to keep your CPU cool and your CPU will fry as well.

    Even if you do get a peltier+water rig working, it's never quite safe because the cool side gets so much cooler than room temperature that condensation starts to form all over the place and you end up shorting your CPU out anyway.

    In short, it's safer to just cool things to room temperature using a water rig without the peltier element, which is just too risky and will still need water to cool it with any recent CPU.

    --
    STOP . AMERICA . NOW
  33. Inevitable by Mnemia · · Score: 2

    If Moore's Law is to continue, liquid cooled computers are probably an inevitability. After all, in order to stuff more transistors into the chips you have to make the wires smaller which increases resistance and thus heat output. Sooner or later, our air cooling systems just won't cut it anymore...

  34. No overclocking by sonofepson · · Score: 2
    Seems like this is not yet the answer overclockers are looking for. From their FAQ
    I am interested in over clocking my system. Would a Koolance PC help me with this.

    The current PC addresses our intended customers, the home and office user who admire a quiet system with the added efficiency of liquid cooling. For the high-performance user, we do not feel our present system can accommodate the additional stresses placed on it by overclocking the components. Please check the news page for updates on our performance-oriented PC model coming in August.

    --
    If Godzilla did not exist, man would have had to create him.
  35. It's been around in Taiwan by jsse · · Score: 2

    Taiwanians, being famous for their overclocking fenzy, made water-cooler for CPU years ago:

    Slot 1 version Socket 7 version The Water Pump
  36. Re:But... by baptiste · · Score: 2
    Well, it depends - my Athlon 700 has three fans, one fo which is in the power supply, two case fans (one in one out) and the heatsink fan - It isn't bothersome most of the time, but if you are watching a quiet movie by the fire with an SO - (my desk is in teh den) it is noticable during quiet periods in the movie.

    I definitely see one of these babies in my future!

    Of course my Athlon is nothing compared to my server room - 5 boxes, plus a firewall was a VERY loud fan, and a 4 disk RAID array tower. PLENTY of noise there! - They used to be in my office - too noisy - now they'll be in a room above my garage - Hopefully I won't be able to hear them in the house LOL :)

    --

  37. Re:And.. by baptiste · · Score: 2
    Which is cool - but what they REALLY need to do is provide a fan header that pulses at 4K a minute or somethign to fake the mobo into thinking its a fan - if the pump dies, the signal stops and the mobo will take action (most new mobos will shutdown if the fna dies or at least will raise an alarm) Plus most mobos have temp thresholds to shut down if they are exceeded.

    --

  38. Water in a power supply.... by baptiste · · Score: 2
    I really want one of these babies. But seeing water tubes going into the power supply with 120 or 240VAC dancing around just makes me squimish. Even though I'm sure its not a big deal - the design looked very nice where the tubes stopped at the cases edge and inside was just a metal cooling assembly. I'm sure I'll quickly get over it and get one anyway :)

    Of course - I can't imagine UL signing off on these Power Supplies, but maybe they already have if they are robust enough. Are they UL listed? If not I can imagine this will be an issue of concern in the corporate world.

    --

  39. Leery by number+one+duck · · Score: 3

    I'm leery of plugging anything even remotely close to leaking into a computer. (How often does my car leak, you ask?)

    Does anyone know what happened to those cooling systems that used a crystal that would absorb/emit heat as electricity was applied? I remember seeing ads for them as much as 8 years ago, then they suddenly disappeared...
    They just strike me as the best of both worlds.. an actual cooling affect instead of a fan, without the risk of a liquid based system. (These crystals *were* mounted on a fan, they were energized on the one side and released on the outside so as to yank the heat out of whatever they were mounted on)