Water cooling over hyped ?

What I was saying is that if anything the room temperature would increase after switching from air to water. Consider the following:
1. Water cooling typically uses more power than air because of the pump, and this means more heat is being generated.
2. Your water cooling setup is able to transfer more heat from your components into the air. Instead of having a bunch of heat trapped in the processor and its heatsink, it gets dissipated into the air.
3. Even without voltage increases, I'm pretty sure that simply increasing the clock speed results in more power draw and thus more heat being generated. Certainly veteraned overclockers will back me up on this.
4. Depending on the setup, your water cooling may have better air flow / ventilation, meaning more room air is making its way through the computer / cooling components and getting warmed up. Instead of a bunch of hot air being concentrated in and around your computer, the heat is being dissipated better throughout the room. This would make it cooler (than before) around the computer and warmer (than before) everywhere else.

CoW]8(0);1031443884 said:
The amount of processor heat dissipated into the air would be same regardless of cooling.
Click to expand...
This is false. See number 2 above.

Dan_D said:
Where did I say anything about power? At the same frequencies, the system runs cooler. I'm able to get a higher overclock out of my CPU, but no voltage increases were made.
Click to expand...
See numbers 1 and 3.
 
cheezer said:
1. Water cooling typically uses more power than air because of the pump, and this means more heat is being generated.
Click to expand...
Probably correct in most every scenario imaginable, but possibly not in absolutely every scenario imaginable.

cheezer said:
2. Your water cooling setup is able to transfer more heat from your components into the air. Instead of having a bunch of heat trapped in the processor and its heatsink, it gets dissipated into the air.
Click to expand...
It's always dissipated eventually, regardless of the cooling system. If heat were to be "trapped" within the processor/IHS, the temperature of the processor would continually rise until it reaches a point of failure.

cheezer said:
3. Even without voltage increases, I'm pretty sure that simply increasing the clock speed results in more power draw and thus more heat being generated.
Click to expand...
This is accurate.

cheezer said:
...This would make it cooler (than before) around the computer and warmer (than before) everywhere else.
Click to expand...
In the short term, possibly, yes. A water cooled system may yield a higher ambient (room) temperature more quickly than would an air cooled system. However, hot air won't stagnate inside a case for any extended period of time (due to obligatory exhaust fans and basic physics). Eventually, it will do precisely what the air exhausted from a radiator does (and we all know what hot air does). Within a certain time period, there would be no significant difference in room temperature between both systems, even when factoring in the heat output of the pump. 15-20 watts is fairly insignificant.

cheezer said:
This is false. See number 2 above.
Click to expand...
No, it's accurate -- to a point. A cooler-operating processor will tend to require less power to function than a hotter-operating processor, and will thusly expend less energy in the form of heat. So, the reality is that a water-cooled processor will generally require less power to operate than a similar air-cooled processor. We're only talking about a few watts or fractions of watts here, of course, which will be totally offset by the power demands of the pump(s), but the pumps have no impact on the amount of heat expended by a processor. His statement is correct.
 
Hardly. I'd like to see what data you have that suggests an 800 watt power supply is the minimum for a water cooled system.
Click to expand...

I am running my system on 500 Watts Ultra X-Finity now, but @ 4.1ghz Q6600 i fell the need of 800 Watts PSU, because my load power consumption is almost 400watts, unless i have to ditch my X1900XT 512mb video card, to keep my 500Watts PSU.
 
phide said:
It's always dissipated eventually, regardless of the cooling system. If heat were to be "trapped" within the processor/IHS, the temperature of the processor would continually rise until it reaches a point of failure.
Click to expand...

Well yes of course, it is dissipated thanks to our good friend physics, but my point was that air cooling constantly leaves a bit more heat in the processor than water cooling does (if done right). It's not much heat at all, but it's still more heat that gets dissipated into the room by water cooling, and all I was trying to show was that there's no way his computer room got colder after switch to water cooling.

phide said:
In the short term, possibly, yes. A water cooled system may yield a higher ambient (room) temperature more quickly than would an air cooled system. However, hot air won't stagnate inside a case for any extended period of time (due to obligatory exhaust fans and basic physics). Eventually, it will do precisely what the air exhausted from a radiator does (and we all know what hot air does). Within a certain time period, there would be no significant difference in room temperature between both systems, even when factoring in the heat output of the pump. 15-20 watts is fairly insignificant.
Click to expand...

Again, my entire argument was to show that the room doesn't get any colder because of a change in cooling.

phide said:
No, it's accurate -- to a point. A cooler-operating processor will tend to require less power to function than a hotter-operating processor, and will thusly expend less energy in the form of heat. So, the reality is that a water-cooled processor will generally require less power to operate than a similar air-cooled processor. We're only talking about a few watts or fractions of watts here, of course, which will be totally offset by the power demands of the pump(s), but the pumps have no impact on the amount of heat expended by a processor. His statement is correct.
Click to expand...

Well ok, not that it matters (as you explained), but that's a good point about processor heat affecting power usage. To be clear about the pump part, I never implied that the pump's power consumption affected the processor's heat output (except very indirectly and insignificantly based on how well it's cooling the processor). It is another source of heat though, which is what I was pointing out.
 
cheezer said:
Well yes of course, it is dissipated thanks to our good friend physics, but my point was that air cooling constantly leaves a bit more heat in the processor than water cooling does (if done right). It's not much heat at all, but it's still more heat that gets dissipated into the room by water cooling, and all I was trying to show was that there's no way his computer room got colder after switch to water cooling.
Click to expand...

As Phide said, if this was the case the CPU temperature would slowly increase until it burned up.

A less efficient cooling system will remove all of the heat that your CPU produces--it will just stabilize at a higher temperature (relative to the air that it's dumping heat into) All passive cooling systems try to make that delta (between room air and CPU temp) as low as possible.

Of course I'm speaking of the case where the CPU has reached a stable temperature. While it's still heating up, you are correct, heat energy is still building up in the CPU. Only once it hits a stable temperature is the heat output obviously equal to the heat input.

As long as we're picking nits, note that fans add heat to the loop as well, unless they're exhaust fans. :) Probably not enough to worry about, though.
 
To me, one thing that biased the Anandtech review is that they only cooled the CPU.

With a lot of the better modern air coolers, there is still a massive heat sink combined with coolant pipes and fans. This takes up a LOT of space in the cabinet, but there is usually a lot of room above the CPU for this kind of stuff. So, they can get a very effective air cooler installed that can compete well with some lesser WC setups.

On the other hand, let's look at GPUs. The amount of space between PCI slots is much more limited. GPU temperatures run WAY hotter than CPUs now. And, if you run dual GPUs (which, I am guessing a lot of folks in this forum do), you are limited by fixed spacing between your high-end GPU cards. As a result, air-cooled GPU options tend toward the width of a single or double PCI slot out of necessity.

As a result, there are two problems with GPU air cooling that aren't addressed in the Anandtech review:

1) GPU air cooling won't compete well with water cooling: Either the air cooler will be much noisier to move the required amount of air, or it won't cool as well for a similar sound level.
2) With an air-cooling solution on dual GPUs, a significant portion of the radiated heat from one the lower GPUs affects the higher GPU (heat rises), since a lot of the heat isn't being pulled away from the lower GPU as effectively in an air-cooled setup. For example, my dual stock air-cooled 8800GTS 340MB (Superclocked) GPUs idled at 59 Deg C (lower card) and 64 Deg C (upper card). My water-cooled setup with the same mobo & GPU cards now yields idle temperatures of 42 Deg C (lower card) and 44 Deg C (upper card).

I haven't had the GPUs stressed out yet (I am waiting for a replacement SLI flex jumper to arrive), but I'm sure they will perform much better than my air-cooled setup where the upper GPU would regularly reach 85 Deg C under load.

Another thing that crosses my mind is that since modern GPUs run much hotter than CPUs, the margin for improvement is much higher. Trying to cool your CPU down to within a degree or two of the case temperature is probably a very difficult task, where cooling a GPU to within 10 Deg C of the case temperature would be an easier task. It seems like this is where water cooling shines above the air-cooled options. I don't have proof of this, though.

I got into WC for one reason only: Cooling my dual GPU setup. Water cooling my CPU is icing on the cake, but I haven't seen as significant an improvement over the stock AMD air cooler (which came with heat pipes). My CPU is now idling at 31 deg C, where it used to idle at around 36 Deg C with the stock cooler. I think WC is still the best (and quietest) option for cooling GPUs, regardless of the cooling benefits for the CPU.

IH8Spam
 
I'm actually surprised to see people still talking about this article. It was pretty clearly biased and just flat out wrong in many respects. My current machine has a CPU only loop that used to be cooled by an Ultra-120 Extreme. Its a great heatsink, no doubt, but it couldn't beat my water setup.

However, in my case, air cooling was quieter (hopefully this will be remedied soon). I think that its just not responsible to throw out blanket statements regarding cooling methods that have so many options to them, and that applies to the "water is quieter" argument. This isn't always true. In many cases it can be, but in others it isn't. This also applies to the "water is better" argument. It isn't always better, as I would expect my U-120E to perform just as well, if not better than the H2O Compact. Value is another argument used against water cooling. Water cooling has never been about value. But this is somewhat subjective. I would expect that most water coolers, including myself, water cool simply because we like water cooling. Its another project on top of the simple system build. Its an obsession by itself.

My biggest problem with the article was saying that high-end air is better than any water setup based on tests with a low-end entry level water kit. This is irresponsible reviewing. He made conclusions on equipment he did not test. After making a mistake like that, I've been checking the author of every Anand article I open, and skip it if its written by Fink. Sorry if this seems like I'm rambling, cuz I am. :) Hellgate DL just finished, time to go.
 
See my sig. 4 hard drives, a 7900 GT, CD Rom, 2gb ram, as well as my WC setup including 3x120mm fans and 1x80mm fan.... all on a basic Antec 550W.

Even moreso, I am running a basic 1x120mm RAD and get great results... cooling power far superior to air even with both the CPU and GPU on my loop.

WC makes it a bigger pain to work on, but it was worth it to me from a cooling performance perspective. Air might be a touch quieter when at load compared to my setup, but when I turn down the fans on the WC setup using my fan speed controller, it is far quieter than my systems on air cooling. I only run full fans when gaming.
 
I have to say that personally speaking, this guide is something I have been looking for, for a long time. Mostly my posts recently have been involved asking what case is better for water cooling, which would give me ample room and would look nice when it's all setup. I already have a water cooling system, however I have been seeing a lot of debate based on the quality of the water cooling system.

How do I know if my water cooling setup is better than those that were reviewed in the article? I previously won a kit from a raffle which is by danger den. Here are the components (some listed are upgrades of existing pieces I own... particularly the block and radiator)

Pump: Some Hydor 12v Pump
CPU Block: Maze5CPU w/ Brass Core, optimized for Quad core processor (I will have Q6600) (http://www.dangerden.com/store/product.php?productid=280&cat=61&page=1)
Radiator: Black Ice Pro II Radiator (http://www.dangerden.com/store/product.php?productid=162&cat=84&page=1)
Reservoir: 5 1/4" Bay Acrylic Reservoir
 
The thermal transfer capabilities of water is much higher than that of a normal copper heatsink but most air cooling methods nowadays use heatpipes which are many thousands of times more efficient in heat transfer than solid metal.

With a good heatpipe designed heatsink and a large enough fan setup, it's possible to get a quiet but powerful system comparable to watercooling at a fraction of the cost.

Water will always be more powerful, but the return on your investment may not be what you expect.
 
Spare-Flair said:
The thermal transfer capabilities of water is much higher than that of a normal copper heatsink but most air cooling methods nowadays use heatpipes which are many thousands of times more efficient in heat transfer than solid metal.

With a good heatpipe designed heatsink and a large enough fan setup, it's possible to get a quiet but powerful system comparable to watercooling at a fraction of the cost.

Water will always be more powerful, but the return on your investment may not be what you expect.
Click to expand...

It is Hassle Free also, no worry about cleaning the loop or what if the pump die and you don;t know. You can burn your CPU and all other things in the loop, e.g, Video Card.

It happened to me, but luckily i was watching my water movement. My pump got disconnected some how, and as soon as i turned the computer on i realize that water in the reservoir is not moving. I went into the BIOS, and see the CPU temperature rising fast. I unplugged the power supply immediately and saved every thing.

So keep in mind when you turn on your computer, just look at your water movement for a second to make sure it is working.
 
annaconda said:
It is Hassle Free also, no worry about cleaning the loop or what if the pump die and you don;t know. You can burn your CPU and all other things in the loop, e.g, Video Card.

It happened to me, but luckily i was watching my water movement. My pump got disconnected some how, and as soon as i turned the computer on i realize that water in the reservoir is not moving. I went into the BIOS, and see the CPU temperature rising fast. I unplugged the power supply immediately and saved every thing.

So keep in mind when you turn on your computer, just look at your water movement for a second to make sure it is working.
Click to expand...

In my expierience, the C2D will always save itself from the brink of thermal destruction because i'm pretty sure it has a wholly internal check for that (even if you turn off that stuff in BIOS). Second, last time my pump didn't start, my C2D idled at 40C instead of 20C...I noticed in OS, and went WTF. It just used the WB as a heatsink, and nothing broke...

Either way though, WC'ing is getting me exactly what I expected, but that's because I haven't really invested enough in it. (TDX + D5 + 1 BIX 120mm), and my case airflow needs redoing (P180). Still, idling at 26C @ 3.1 ain't bad I guess.
 
movax said:
In my expierience, the C2D will always save itself from the brink of thermal destruction because i'm pretty sure it has a wholly internal check for that (even if you turn off that stuff in BIOS). Second, last time my pump didn't start, my C2D idled at 40C instead of 20C...I noticed in OS, and went WTF. It just used the WB as a heatsink, and nothing broke...

Either way though, WC'ing is getting me exactly what I expected, but that's because I haven't really invested enough in it. (TDX + D5 + 1 BIX 120mm), and my case airflow needs redoing (P180). Still, idling at 26C @ 3.1 ain't bad I guess.
Click to expand...

XD.

My girlfriend used my PC to surf around Live Journal while I was in the middle of setting up it's loop. Yes, there was nothing but a blob of AS5 on the chip (no block, NOTHING) for the 20 minutes she was fooling around. She then proceeded to ask me why "Did you spend so much money on such a slow computer?"

The AS5 was bubbling... and the CPU was under maximum thermal throttle, running at around 50mhz.

I also throw on a check in speedfan, to shut down the system if the temperatures rise too high.
 
slackjaw said:
what i want to know is if watercooling the NB is overrated.
Click to expand...

Depends on how you look at that. I'm going to be water cooling my NB when up I upgrade my loop (probably in week or so). My reasons for doing so are as follows:

1. A good air cooler for it is about the same price as a water block
2. I wouldn't need to add another fan to my system if I put it on water (read: less noise). I probably would if I used air cooling.
3. A water block will take up less space that most air cooling heat sinks (so long as you don't count the tubing lol).

Now if you're going to water cool your NB because you think it will be BETTER than similar priced air cooling (~$25-30 USD), you're probably going to be disappointed. While the temps will probably be lower, it's not going to make your computer faster and probably won't allow for much of a higher OC unless you're already in an extreme situation (i.e. phase cooling).

Just my two cents.
 
waterlogged said:
From a Euro perspective I would say that water cooling is for silence. Also the OC too, but the real key to OC is in the components and not the water. :p
Click to expand...

Yep.

4ghz might be worth a million, but doing it at 26dba full load is priceless.
 
Well at the end of the day I enjoy reduced noise and being able to operate at frequencies I could not acheive with air alone.
 
Volkum said:
Depends on how you look at that. I'm going to be water cooling my NB when up I upgrade my loop (probably in week or so). My reasons for doing so are as follows:

1. A good air cooler for it is about the same price as a water block
2. I wouldn't need to add another fan to my system if I put it on water (read: less noise). I probably would if I used air cooling.
3. A water block will take up less space that most air cooling heat sinks (so long as you don't count the tubing lol).

Now if you're going to water cool your NB because you think it will be BETTER than similar priced air cooling (~$25-30 USD), you're probably going to be disappointed. While the temps will probably be lower, it's not going to make your computer faster and probably won't allow for much of a higher OC unless you're already in an extreme situation (i.e. phase cooling).

Just my two cents.
Click to expand...

It also depends on the NB. For example nforce 6 nb is a furnace, it will help you a lot better then cooling a p35nb
 
I'll never go back to air cooling. I enjoy pushing my computer and all I hear is the buzz of my monitor. I could care less who thinks it's overhyped and who doesn't.
 
I went watercooling not too long ago. I got tired of my vantec tornado whining so loud for a few years already. Just decided to finally go water and i did. Its a lot more cooler now and a lot less louder. Well worth it if you ask me :). The noise went from a jet taking off to a super silent humming. Now i can overclock without having to deal with a really loud fan lol :cool:
 
what i want to know is if watercooling the NB is overrated.
Click to expand...

P965 and newer Intel MCHs ? Hell no if you are a heavy overclocker - say FSB over 425MHz, if not redo the heatsink/heatpipe thermals and just slap a fan on it.

In a year it will not be needed as much (at all ? ) due to the onboard memory controllers.

It is my impression EVERYTHING is over hyped, cya, going to go turn off everything electronic and read a real , made from processed plant fibre, bound in preserved animal skin, book and escape this madness.
 
BillParrish said:
...going to go turn off everything electronic and read a real , made from processed plant fibre, bound in preserved animal skin, book and escape this madness.
Click to expand...

I have several books on government from the mid 1800's that I find to be great reads. :D When my grand kids read them it will be over 200 years old. :eek:
 
The author of that article is an idiot with his head in the sand if he thinks that high end air beats water cooling. Water can do it cooler and quieter. It's just more much more expensive and difficult to setup usually. You can spend $100 for a low end custom but decent water cooling setup for your CPU or you can go all out and spend $300ish to cool your CPU and GPU with top of the line components for maximum performance.
 
Did you guys not realize those kits are like super cheap compared to what "Normal" water cooling would look like? I mean those things have wimpy pumps, wimpy fans. Yeah they are quiet but they are not designed to be super cool like the high end systems. Whoever thinks watercooling is hard...I don't see that at all. I have been watercooling for around 3-4 years now and if I have to upgrade something I just get a new waterblock with it, drain the system, change it out and I'm back up and running. I just changed my setup last night to make room for my 1600 watt Ultra X3 (FRIGGEN AWESEOME) and it was super easy. One thing that makes it a lot easier is getting some quick disconnects that close off when disconnected. I got 2 sets from McMaster Carr and I can disconnect the mobo/vidoe and remove it on the tray without draining the system.

Water can absorb heat 30X better then air...this is not an opinion it is a fact. The surface area on a radiator is about 100X more then a typical tower air cooler due to how radiators are made. The physics pretty much speak for themselves, 30X better heat transfer to 100X more surface area. A fan that is barely spinning will move enough air to cool a lot of wattage due to the huge surface area increase. The units that were tested are a joke, they are meant for the kids that want to be cool and brag to their friends that they now have water cooling but really don't have a clue what real water cooling is about. Setting up a water cooling setup is not hard, just takes some planning. I may be a little biased on this because I have a UFO-U2 case that is big enough to stick my head in when I need to see things and work on stuff...but still its not hard and its a lot more enjoyable afterwards and you really feel like you have accomplished something. As far as cost, yeah I guess it costs more in the short term .If you get a water block like that apogee GT that supports all the major chipsets and will most likely have new plates for new boards that come out then figure it this way. You have 1 block that will fit on like 8 boards be it AMD or Intel. You'd need at least 4-5 air coolers that could do that, some will mount on multiple boards but not all 8. So will say 4 air coolers at $50 each, that right there is $200. Once you have the radiator, the pump, the res and all the tubing the only things you will ever change are the water blocks and most are made to fit many designs. The costs are minimal when you look at for the long term. Not sure about you guys but I upgrade about once a year, the pump lasts 5 years so I wont need to replace that but I would need possibly 5 heat sinks...Bottom line, the cost is not that much!
 
Is this thread still going?

Depending on how you look at it, it is correct "IF" you meet the biased pre-conditions of the article.

What the article isn't saying is that the kits it is presenting as representing watercooling are "low end" kits (1x120mm rad = low end). Compare low end watercooling to high end air and of course the conclusion will be that low end WC isn't worth it.

Therefore this is biased reporting and I'm surprised Anandtech has articles like this on their site as it just makes them look silly.
 
Instead of a new thread, I'm going to resurrect this one. I'm begining to wonder if my next build shouldn't forego the WC. I really did it initially just for the geeky dweebness of setting up a WC setup.

Now, with the newer generations of chips running so much cooler and with heat being less of an issue, I'm really seeing no benefit to a WC setup. Even in terms of noise, you can get big old honking quiet fans that will be nearly silent and which will still keep temps well under control on newer processors. I think the 45nm chips are changing the picture even more.

In my opinion WC was barely a practical thing to do over the last few years, and now I think it's totally irrational in almost (but not quite) all cases. What really changed my mind was my recent HTPC build. E6420 running at 3.0GHz with almost 0 effort in a tiny Antec Fusion case with an SI-128 and a big slow fan. Peak temp under orthos is 47-48C. If I can do this silent build with air in a small, cramped case with no clearance I think my water cooling days are over.
 
jvrobert, good luck pulling the same feat with a quad core ;)

I had my computer with a SI-128 and a Silverstone FM121 at max speed (110 cfm) and it was unable to keep heat in check over 3.2 GHz so watercooling is my only option since my case cannot take a tower-style heatsink.

If you can manage to keep under 50C at full load with aircooling, watercooling may be overrated but for those who cannot do this without blowing your ears, watercooling is a great option. Another thing in favor of watercooling is if you pick the right parts when you first build it, you can ride on the same parts for a long time (with only the cpu block to change if the manufactuer doesn't release a new mounting plate) while a heatsink might not work so cost is not a issue there.

In the end, aircooling and watercooling has their own share of tradeoffs and only you will determine what is the best for your needs, depending on your budget.

 
I've had experiences with both air and water cooling.

The biggest disadvantage to watercooling is the ability to change parts quickly in and out. You have to fully drain a system usually before you can remove a part that is being watercooled. Because it takes a lot of time and effort to plan a watercooling rig, changing it is (not impossible) but a little cumbersome at times.

Air cooling on the other hand is easier to change things out but definitely noisier. The biggest advantage of watercooling is the fact that your noise level will sound more like a monotone air noise rather than different pitches of fans going at different decibels.

Watercooling also can maintain very stable temperatures as oppose to air cooling that relies heavily on the ambient temperature of the things around it.

Would I recommend my friend's 80 year old grandmother to watercool? no.. probably not even his 25 year old sister who goes online to check FB. However, just like cars, if you're into tweaking and tinkering with it, watercooling is a funy hobby and not overhyped by any means.
 
Wshmaster0 said:
I've had experiences with both air and water cooling.

The biggest disadvantage to watercooling is the ability to change parts quickly in and out. You have to fully drain a system usually before you can remove a part that is being watercooled. Because it takes a lot of time and effort to plan a watercooling rig, changing it is (not impossible) but a little cumbersome at times.

Air cooling on the other hand is easier to change things out but definitely noisier. The biggest advantage of watercooling is the fact that your noise level will sound more like a monotone air noise rather than different pitches of fans going at different decibels.

Watercooling also can maintain very stable temperatures as oppose to air cooling that relies heavily on the ambient temperature of the things around it.

Would I recommend my friend's 80 year old grandmother to watercool? no.. probably not even his 25 year old sister who goes online to check FB. However, just like cars, if you're into tweaking and tinkering with it, watercooling is a funy hobby and not overhyped by any means.
Click to expand...

Exactly, watercooling is only for those who like to tinker with the computer and ooze the most performance or the most silence out of a computer. If you just want something which works without any fuss, watercooling is not for you.

It's mostly a hobby like cars enthusiasts who love to tinker to improve the performance or handling. If all you do with the car is to go from A to B, then modding the car is not for you as well.

 
jvrobert said:
In my opinion WC was barely a practical thing to do over the last few years, and now I think it's totally irrational in almost (but not quite) all cases. What really changed my mind was my recent HTPC build. E6420 running at 3.0GHz with almost 0 effort in a tiny Antec Fusion case with an SI-128 and a big slow fan. Peak temp under orthos is 47-48C. If I can do this silent build with air in a small, cramped case with no clearance I think my water cooling days are over.
Click to expand...

I agreed, but as some one say good luck with Quad Core. My friend it is changing also, look at 45nm Quad which can overclock to 4.6ghz on air.
 
Despite the Anandtech anti-watercooling hype, I went ahead and bought a H20 Compact 120 kit yesterday.

My ThermalRight Ultra 120 Extreme, paired with a 1600 RPM Scythe SFF-F, wasn't keeping the temps on my B3 Q6600 where I wanted them. At 3.2Ghz/1.30v, I idled at 48-50 on all four cores, and hit the mid-70's on all four under OCCT or Prime. I'd tried reseating the damn thing dozens of times, tried all four orientations, used the washer on the base trick to get better contact, and wasted almost a tube of AS5 in the process.

I basically did everything BUT lap it, and there was no way I was going to spend an hour or more in agony (early Rheumatoid arthritis) to make up for bad design/production.

Installed the Swiftech unit inside the case, with the supplied Ceramique and my old Scythe in push/1000 RPM, fired the unit up, checked to make sure it was cooling, and then went and watched TV for an hour. Came back, and my idles were 40, 35, 35, 36.

Ran OCCT for half an hour, came back and checked the graphs...my hottest core never went over 62c, and the other three ranged from 57c-61c.

Folks, that's better than a 10c average drop across the board, with a "less effective" TIM, and a single little 120 radiator.
 
Croak said:
Folks, that's better than a 10c average drop across the board, with a "less effective" TIM, and a single little 120 radiator.
Click to expand...

But anandtech says watercooling is overhyped, so you must just have a fluke result. Sorry.
 
Arcygenical said:
But anandtech says watercooling is overhyped, so you must just have a fluke result. Sorry.
Click to expand...

Damn, it must be fluke day. Since the H20 120 Compact was doing such a freakishly good job at 1.30v, I decided to do something I had no luck with using the TRUE 120.

Bumped up the voltage to 1.52 in BIOS, giving me a solid 1.50 actual. Cranked up the fan speed on the radiator to 1600 RPM. Set the bus to 400, and the multi to 9.

Fired up and running at 3.6Ghz, idling at ~48, a half hour of OCCT gave me 75c max.

That's hotter than I like, especially with a B3, and it's surely overloading that poor 120.1 rad, but it's impressive because my system would go to 90c in a matter of seconds using the Ultra Extreme at those settings and fail Prime 95 almost instantly if it didn't lock up, so I never tried for 3.6Ghz again.

My only complaint is that the pump/block combo paired with my flimsy CM 832 has some vibration issues, which is causing some annoying noises.
 
I think it depends on what you're cooling and what you're going for.

For me, silence was my main goal, and it was acheived. As a secondary accomplishment, that 8800gt that's been cooking my case now runs nice and cool... which has helped out the rest of my case temps greatly.

For just cooling a normal CPU though, an air cooler will probably be only a few degree's different, for alot less money, and alot less hassle.
 
You must log in or register to reply here.
Back
Top