Extreme air cooling with TRUE + guide/tips

[Simps]

I - Sorry for my bad english
II - I want to share my system and some guides / tips for aircooled rigs
III - I will share full TRUE mods for best performance



Little guide for aircooling with TRUE

If you are going to build a good aircooled rig, you should consider at least the following...



Case

I - has to be big
II - Alumin. is best
III - has to allow you to put fans on front/back/top/bottom and side

Ideal case for aircooling, will be big, alumin, and will let you place fans on front/back/top/bottom and side. This last part is very important.



Airflow

You have to think about the best airflow design for your case. It will change depending on the case.



Motherboard

I - change stock thermal paste with AS5 or other good stuff
II - add custom heatsinks if needed
III - tie screws hard
IV - Add fans to NB / SB / Mosfets. Don't let them run passive.

This is pretty much what you need to do with your motherboard. Change the NB / SB stock TIM with AS5. Leave the stock thermal pad of the mosfets. You can add custom larger heatsinks to the mobo too. If you have screws on it, make sure they are tied all the way. Push-pins are bad, you need to exchange them if you can. Don't let them run passive, and don't trust them only on the case airflow.



CPU / GPU

I - lap it (280grit, 400grit, 600grit, 1000grit, 1500grit, 2000grit)

CPU and GPU are the most important items to cool. They will go hot. You better lap them for better performance.



TRUE

*** You need to do this with your TRUE, or you will be losing about 2C to 10C on temps.

I - you need to lap it (280grit, 400grit, 600grit, 1000grit, 1500grit, 2000grit)
II - you need to do the washer mod, since when you lap it, material will be gone, and you will have less pressure then you should.
III - you have to push/pull with quality fans (They have to be 75 - 80CMF+ each)
IV - vertical mount for better results

TRUE is good, but only if you respect this steps. A TRUE without lapping will result in about 2C to 4C temps. Washer mod will give you another 1 to 2C. Using high CMF fans with push/pull will give you another 2C to 4C over one fan setup. Mounting it vertically will give you about 1C better temps.

As you can see, someone using the TRUE without lapping, with only one fan, without the washer mod, and horizontly mounted, can be losing about 10C on temps.



Other heatsinks (GPU etc...)

I - you might need to lap it (280grit, 400grit, 600grit, 1000grit, 1500grit, 2000grit)

GPU's will actualy go hotter then your CPU. So you need to use a good custom heatsink here



Case fans for the airflow

I - you will need a lot of them, go with quality and quite stuff (~45CMF - 50CMF / 26dB is good)

You just can't make a good airflow with two or three case fans. This is a fact. Go with quality fans only, and with the noise level you want. For the case airflow, the fans don't need to be monster ones. Just respect your noise level here, and go with at least 45CMF to 50CMF fans.



Thermal paste

I - The only good way to apply it, is with a clean piece of plastic
II - You need to use very little of it for better results on lapped stuff

Forget about credit card for applying AS5. The only good way to do it, is with a clean plastic. You need to use very little AS5, and put a lot of pressure on the plastic when applying it. This might get you another 1C to 2C temp diference. I will show with pictures later.

---------------------------------------------------------



My system

I will share my system, and give some tips.



Case

I am using the coolermaster stacker 832. It is big, alumin, and I can mount fans on top, bottom, front, back and side, and still keep the case with doors closed.

No pic.



Case airflow

http://img523.imageshack.us/img523/2940/airflownf1.jpg



Lapped stuff

I - CPU
II - GPU
III - CPU Heatsink
IV - GPU Heatsink

http://img77.imageshack.us/img77/7871/laplz5.jpg



Motherboard mods

I - AS5 on NB and SB
II - Washer mod
III - Three direct fans on NB and mosfets (heatpipe solution)

No pic.



TRUE mods

I - Lapped
II - Washer mod
III - vertical mount
IV - push/pull with high CMF fans

Best way to fix two high cmf fans in push/pull on a TRUE. They won't move for nothing, and it is very clean.
Look at the 5th pic below.

http://img383.imageshack.us/img383/5549/truefanskh6.jpg

http://img88.imageshack.us/img88/7762/truewasherhn3.jpg

http://img353.imageshack.us/img353/5884/truemounted3vi3.jpg

http://img353.imageshack.us/img353/6559/truemounted4tw0.jpg

http://img513.imageshack.us/img513/52/truemounted2vg8.jpg



TRUE fans

I am using two Panaflos 120x38mm for the push/pull.

http://img116.imageshack.us/img116/1706/pana2bg4.jpg

http://img116.imageshack.us/img116/7292/pana1va3.jpg



AS5

This is how you should appply the AS5 with a plastic layer. You need to use very little AS5, making a very very very small layer.

http://img513.imageshack.us/img513/371/trueas5gi7.jpg

http://img513.imageshack.us/img513/1221/trueas52jo5.jpg



GPU

GPU will get hot, so you better add some custom heatsink to it, and then a lot of fans.

http://img353.imageshack.us/img353/2153/gpu1of4.jpg



Case fans / final look

I - bottom fans

http://img388.imageshack.us/img388/7162/backfansoz2.jpg

II - front fans

No pic.

III - top fans

No pic.

IV - back fans

http://img523.imageshack.us/img523/7644/backfi4.jpg

V - full system

http://img239.imageshack.us/img239/7919/full2ae1.jpg

http://img388.imageshack.us/img388/987/full4is8.jpg

http://img388.imageshack.us/img388/6536/full3mj8.jpg



Temps

This is what I got from a 65nm QX6700 (B3) quad CPU, overclocked to 3.2GHz (400x8) @ 1.3v.
This CPU runs very very very hot. I believe, when overclocked, it would go as close to 200W of heat.
It is a good way to test the system cooling power.
I live in Brasil and it is summer time here, so my room is very hot. Ambient temps were about 27-28C.
NB / SB / MOSFETS temps were very good, so I guess the case flow is doing great.
I am stressing it with the case's door closed.

2h prime95 small ffts.
I used real temp with 100C TjMAX.

http://img504.imageshack.us/img504/2604/temps1gi4.jpg



UPDATE: 19-Jan-2009
3.5GHZ 500FSB 1.43Vcore 65nm B3 QX6700
7h prime blend + 20 loops IBT on max stress
Full-load temps during prime small fft's. 100C TjMAX on Everest.

http://img80.imageshack.us/img80/5244/new50013wu3.jpg



[]'s
Simps

[roofsniper]

you sure love fans don't you? im gonna guess u got 12 in there.

[cabbspapp]

I think he has seventeen, counting the three on top and that one hidden on the very bottom of the case.

[jdmR]

air cooled enough? :P

[demonkevy666]

nice

[cabbspapp]

Did you have to bolt it to the floor to keep it from flying around the room?

[Simps]

Did you have to bolt it to the floor to keep it from flying around the room?

Hehehe no :D

Lot of fans, but only CPU ones are monsters.
Other fans are operating @ 23-24dB only, and I use the computer with the doors closed. So far, I am very happy with the noise and performance.
I have the same results I used to have with a custom watercooling setup.

The secret is the TRUE, but you HAVE to do all the following, to really enjoy this cooler:

-vertical mount
-lap
-washer mod
-push/pull with 75+ CMF fans
-use very little AS5
-have good case cooling / flow

If you do that, you will have close to custom watercooling performance.

On the fan qty, I guess if you are not going to use fans louder then, say 25dB, you can actually use a lot of them, and the noise will still be low.

I will try to post some temps pics later.

[]'s
Simps

[roofsniper]

I think he has seventeen, counting the three on top and that one hidden on the very bottom of the case.

lol i didn't count them all up but i guess if u count the 3 on the ram that seems about right. LOL lots of fans. i have about 10 and i thought i had a lot.

[fatguy1992]

Awesome :up:

What fans are they?

[Simps]

CPU fans are big bad 120x38 panaflos.
Others are 120x25 and 80x25 VIZO FREEZERS with Thermal Control.

http://www.vizo.com.tw/front/product/get_product_detail/63/3

There are some small 60x25 MTEK's too...

[Simps]

hmm I have some comparison now.

current system - TRUE with:

-vertical mount
-lap
-washer mod
-push/pull with 75+ CMF fans
-use very little AS5
-better case cooling / flow

~50C temps



Old system (same clock/settings) but with TRUE with only:

-vertical mount
-lap
-push only with 75CMF fan
-a lot of AS5
-little less case cooling

~58C temps


A gain of ~8C for:

-washer mod on TRUE
-changing from push to push/pull
-using very few AS5
-improving case cooling

[Levish]

i like your thermal paste application method, might want to add that you should clean the plastic bag as well as the heatsink surface area as well

[GregSG]

Wat casing is that?

And wat fans ur using?

[fatguy1992]

And wat fans ur using?


CPU fans are big bad 120x38 panaflos.
Others are 120x25 and 80x25 VIZO FREEZERS with Thermal Control.

http://www.vizo.com.tw/front/product/get_product_detail/63/3

There are some small 60x25 MTEK's too...

He already had answered :p:

[Simps]

I just added some extras and tips to the main post.
Thanks guys.

[]'s
Simps

[GregSG]

Ok thanks...

[ThugsRook]

wheres the hood scoop?

:shakes:

[Simps]

What is a hood scoop?

[Asgard_thor]

holy...

[Asgard_thor]

50C for a overclocked quad is good, what CPU?

[Simps]

CPU is a QX6700 (B3) @ 3.6GHz / 1.44V, which make the results even better I guess... I will post temp pics later.

[spookmineer]

A few myths here... :(



Case

II - Alumin. is best

Why? If airflow and components are the same, a wooden case will give the same results (except for the hard disks if they are hard mounted to the case - the disks will benefit from this because some heat will be dissipated through the chassis).




Case

Ideal case for aircooling, will be big, alumin, and will let you place fans on front/back/top/bottom and side. This last part is very important.
Not in all situations. With low-medium airflow, side fans will mess up the airflow. It all depends on the components used.




TRUE

III - you have to push/pull with quality fans (They have to be 75 - 80CMF+ each)
In this setup, it would be interesting to see if you see a temp drop at all if you only use 1 fan.




Case fans for the airflow

You just can't make a good airflow with two or three case fans. This is a fact.
Heh... Ok, it's a fact... I only have 1 rear case fan, undervolted to 800 RPM (not counting PSU fan, which is in another compartment of the case).
In some cases, efficiency is better than brutal power ;)

[Simps]

spookmineer,



Heh... Ok, it's a fact... I only have 1 rear case fan, undervolted to 800 RPM (not counting PSU fan, which is in another compartment of the case).
In some cases, efficiency is better than brutal power


Fair enough, but I entitled the thread "EXTREME air cooling" for a reason.
When you are going to use a 3.8GHz 65nm quad for 24/7 or a 4.2GHz 45nm quad 24/7, you need to do some more extreme cooling then just putting 1 fan undervolted on the rear of your case. Unless you like to see it running 70-80C daily, and living about 1 year.

Extreme air cooling is for extreme overclockers. No one will do such a thing, and run a moderated overclocked cpu. That is pointless.

I have nothing against people running even passive cooled system (zero noise). I actualy have one too. There is no right or wrong here, it is just choice and situation.

This kinda cooling might be too much or unecessary for your needs. But they are needed for more extreme overclockers. This is xtremesystems.com, not average_joe.com

[]'s
Simps

[spookmineer]

You are right, I forgot about the "Extreme" in the thread title, sorry for that :/

The aluminium case is still pointless. It makes no difference what material the case is made of.

I still wonder though, what actually made for the 8 °C temp drop. You can't draw hard conclusions if you change 4 variables at once.
If you are experimenting more, you will find out that no single case is the same, and that results obtained in other threads/articles might not be the most optimal for you (or others...).
The thing that got me going a bit is that this thread sounds like what you need to do in order to get near optimal results, whereas it actually is a suggestion, bearing in mind the specific hardware used. Other people needing extreme airflow might be better off with different settings/hardware/etc.

Regarding optimal results, for example: I found out that a side case fan really messes up my temps, even with a ducted side case fan (to CPU). Losing the fan altogether got much better results (in my case).
Lowering the rear case fan speed also got me better temps (perhaps because now it is a real flow, and not air chaos in my case, but I'm guessing here).

I'm very interested in your fanless PC btw, maybe you can make a thread about that one as well? :D
[PS I don't have an underpowered PC, I'm a devoted gamer, but I also like quietness and always eager to see what solutions people come up with]

[Simps]

No problem m8.
You are probably right about the case, but I still would like to see some real numbers and comparisons before changing my mind.
I would like to know what contributed more for the ~8C difference too, but I just changed everything in once, I had no patience to do the tests one by one. Its such a pain to mount everything in the case...
About my fanless system (not really, because my psu has one fan), I will try to take some pics for you.

[Vapor]

Try turning off your case fans (or the majority of them)....I'd be surprised if they were contributing to your CPU temps in the slightest. I have 5V Yate Loons as my case fans and my CPU temps are the same as they are in open air. :shrug:

Also, what are you using for CPU temp measurement? 50C from TJMax is crazy cool, especially for 3.6GHz 1.44V :eek:

[Simps]

Try turning off your case fans (or the majority of them)....I'd be surprised if they were contributing to your CPU temps in the slightest. I have 5V Yate Loons as my case fans and my CPU temps are the same as they are in open air. :shrug:

Also, what are you using for CPU temp measurement? 50C from TJMax is crazy cool, especially for 3.6GHz 1.44V :eek:

I would like to do that, but I can't since most of the fans are plugged directly into the power supply molex, and I have tied the connectors with cableties, so it would be a lot of work to mess with that now...

I am using real temp for the temps... but I am using 95C for TjMAX, instead of 100C, intel official stated that TjMAX is 90C for 65nm quads, but I don't believe it. That is why I am using 95C.
With the regular 100C, you can transform that 50C (witch is now, in more extended tests about 55C) to 60C. Still good.

But I will post pics later, and just trying to get a stable max clock now...

[]'s
Simps

[Serpentarius]

this may sound absurd ... for airflow ... you could use Fleming's Right Hand Rule ... to decide the best flow ..

on the contrary of watercooling ... turbulance of water molecules are prefered to breakthough the water surface tension ...
the airflow requires non-turbulance flow to smoothen the air flow .. the best way to have the fastest air flow is to create and maintain a mini vortex .. by using the right hand rule .. vortex appears everywhere .. but maintaining in the casing is hard ..

hopefully you can understand what i'm trying to say ... if you could .. u'll understand why not every front intakes will have the back outgo .. try it :)

[ThugsRook]

oh come on already.

that case is a mess and i guarentee most of those fans do absolutely nothing. you just threw as many fans as possible at it.

looks like nothing more then a n00b in action here.

[Simps]

oh come on already.

that case is a mess and i guarentee most of those fans do absolutely nothing. you just threw as many fans as possible at it.

looks like nothing more then a n00b in action here.


Admin, can you lock this thread?
If I start to answer this kinda post, I will just start a flame war...

[XDMolotov]

Let's get some decent results in here guys :rolleyes:

Redid my TRUE mount yesterday. I've got 2 75CFM Antec Tri-cools pulling air out of the case and 2 of them pulling air in with 2 Yate Loon D12-SM12's in Push/Pull. I've lapped the TRUE (washer mod) and have an E8500 on 3.61 @ 1.16v.

Current temps are 35 idle and 45 load. Ambient is around 25 but it could be higher (don't have a temp monitor right this moment).

Benchmarks below were done with very cold ambient temps, 5-10C :p:

[Metroid]

First thanks very much Simps for your hard working. The way I see is that 4x3.6Ghz Q6700 65nm is almost the same thing compared to a 4x3.6Ghz Nehalem 45nm concerning the temperature. I wonder by how much my Zalman NT9700 can hold Nehalem at 4.0Ghz.



oh come on already.

that case is a mess and i guarentee most of those fans do absolutely nothing. you just threw as many fans as possible at it.

looks like nothing more then a n00b in action here.

Have you seen the date Simps joined here at XS?

Yes was in 2002. So respect his opinion, next time you come here do me a favour, if you want to contradict his results, first do a research or tests then come back and show us the outcome.

Simps spent a good amount of time hoping to guide users who are still using air cooling systems.

[dan7777]

Thanks alot simps great thread enjoyed reading through ive got a q6600 at 3.8 1.56vcore idles 36 load is bit high 66c =67 how can i lower temps anyone ? gpu is 8800gts 320mb upgrading any day to gtx 260-216sp thanks in advance for your help. dan

[TMM]

In this setup, it would be interesting to see if you see a temp drop at all if you only use 1 fan.
I run a TRUE120 with a very low speed fan (Nexus Real Silent), and i was only seeing 1-2C difference with two fans as opposed to one. The difference will be even smaller with high speed fans.

With a 38mm Panaflo (higher static pressure) there is going to be even less advantage from using two fans.

[Simps]

Thanks Metroid, thanks dan.
Added two more pics to first post.

[Simps]

Added on the first post.

This is what I got from a 65nm QX6700 (B3) quad CPU, overclocked to 3.2GHz (400x8) @ 1.3v.
This CPU runs very very very hot. I believe, when overclocked, it would go as close to 200W of heat.
It is a good way to test the system cooling power.
I live in Brasil and it is summer time here, so my room is very hot. Ambient temps were about 27-28C.
NB / SB / MOSFETS temps were very good, so I guess the case flow is doing great.
I am stressing it with the case's door closed.

2h prime95 small ffts.
I used real temp with 100C TjMAX.

I will post pics of 3.6GHz later too.

http://img504.imageshack.us/img504/2604/temps1gi4.jpg

[neo mike]

Come on bro, thats wayyyyyyyyy overkill on the fans, you dont need half those fans, you have air flowing that dont even know what direction to go.:shrug:

[Simps]

Come on bro, thats wayyyyyyyyy overkill on the fans, you dont need half those fans, you have air flowing that dont even know what direction to go.:shrug:

Noted. Thread is entittled "extreme air" for a reason.
And those extra case fans ARE making a difference, since my old setup had about 4 less case fans, and I can see up to 2-3C drop on my SB / NB and MOSFETs temps. And for that to be happening, it has to be related to the extra fans.

Cheers,
Simps

[Jamesrt2004]

how did you mount those 2 fans on the true?

-noob q-

[jdmR]

i wanted to do a push pull on my s1283 but i wouldnt know how to mount it unless i use zip ties or something. so instead, i put a big ultra kaze to pull the air out. for my hs, i have a ultra kaze intake then a sflex to push the air through and then another ultra kaze exhaust.

[Simps]

how did you mount those 2 fans on the true?

-noob q-

http://img383.imageshack.us/img383/5549/truefanskh6.jpg

http://img513.imageshack.us/img513/52/truemounted2vg8.jpg

[Jamesrt2004]

ahhhh very clever ! ill get on the case now :D


just need to lap and add the pull fan!

[X4er0]

Did you have to bolt it to the floor to keep it from flying around the room?

HAHA :rofl: HAHA i was just thinking the same, and als "won't is just get knocked over to the ground because of the wind blasting against the MB"

It's the big bad wolf blowing your PC away.... :ROTF:

ps: i have lots of fans too, but you win hands down

[Simps]

UPDATE: 19-Jan-2009
3.5GHZ 500FSB 1.43Vcore 65nm B3 QX6700
7h prime blend + 20 loops IBT on max stress
Full-load temps during prime small fft's. 100C TjMAX on Everest.

http://img80.imageshack.us/img80/5244/new50013wu3.jpg



[]'s
Simps

[tiro_uspsss]

Come on bro, thats wayyyyyyyyy overkill on the fans, you dont need half those fans, you have air flowing that dont even know what direction to go.:shrug:


oh come on already.

that case is a mess and i guarentee most of those fans do absolutely nothing. you just threw as many fans as possible at it.

looks like nothing more then a n00b in action here.

geez u 2 sound like :censored: :censored: :censored:
stfu & gtfo


nice results Simps :up:

[Demo]

Simps

Thanks for sharing how you mounted the dual fans, Im going to try that route some day when me new case and fans arrive.

This is what i get now with 30-31c ambient, with 95-95-102-102 TjMax settings.

single FM122 at about 1k rpm

http://img93.imageshack.us/img93/1662/cpuzmx1.jpg
http://img93.imageshack.us/img93/177/rtdb7.jpg

Lapped but no washer mod yet.

[valiant]

What is a hood scoop?

DANG!! thats alot of fans!! Kudos to you. Looks like alot of work. Good job

Here is what a hood scoop looks like :D
http://i117.photobucket.com/albums/o79/jherr1/L05.jpg

[Simps]

geez u 2 sound like :censored: :censored: :censored:
stfu & gtfo


nice results Simps :up:

Thanks, I will update soon, with higher clock and volts too.




Simps

Thanks for sharing how you mounted the dual fans, Im going to try that route some day when me new case and fans arrive.

This is what i get now with 30-31c ambient, with 95-95-102-102 TjMax settings.

single FM122 at about 1k rpm

Lapped but no washer mod yet.

Nice, I hope you can drop another 1C or 2C by doind the whasher mod. It won't drop anymore then that. Now, if you go into push-pull with 80CMF fans + whaser mod, then you can expect a 5-7C drop or more.

I am not sure you can do that with your TjMAX. I think you just need to leave them at 100C for 4 cores.




DANG!! thats alot of fans!! Kudos to you. Looks like alot of work. Good job

Here is what a hood scoop looks like :D

Holy Moly!

[]'s
Simps

[Demo]

There was a little guide on the RT thread on how to calibrate for your TjMax if you wish to discard what intel said about 90c, at at 200Mhz FSB and 6x multi with any cooler the CPU should idle about ~8c above ambient and 95TjMax gives me around 9-10c above ambient and 90TjMax results in lower or = to ambient.

[Simps]

There was a little guide on the RT thread on how to calibrate for your TjMax if you wish to discard what intel said about 90c, at at 200Mhz FSB and 6x multi with any cooler the CPU should idle about ~8c above ambient and 95TjMax gives me around 9-10c above ambient and 90TjMax results in lower or = to ambient.

I wouldn't trust that. No way.
I would just use 100C TjMAX for all 4 cores on 65nm quads. I also don't believe 90C or 95C is right for 65nm quads.

[Demo]

51c is extremely impressive for a B3 at 1.296v loaded... you sure you dont have clock modulation enabled ?

[Simps]

51c is extremely impressive for a B3 at 1.296v loaded... you sure you dont have clock modulation enabled ?

No, don't have it.

[Ozzfest05]

ok if this is EXTREME air cooling for Extreme clocks why you only at 3.5 I had my setup which I might add was whisper quite the q6600 @ 3.8 24/7 push pull true with some Sflexs I consider that a decent clock considering 24/7 definetely not extreme I wasnt going for that I was going for stability, show us extreme put that q6700 to 4.2 and lets see if that mess of fans and noise warrant the temps

[Simps]

ok if this is EXTREME air cooling for Extreme clocks...

I stop reading here. I never said this is for extreme clocks. Where did I say that, can you show? My quad needs 1.43v for 3.5GHz and 1.45v for 3.6GHz. It is an old B3 quad, and for 3.8GHz it would need 1.55v or even more. I am not going to push those volts on it. Making it 3.8GHz or 4.0GHz stable for 24/7 is much more dependent on having a good chip, then the cooling itself. Also B3's do get 8-10C hotter then G0's on same volts, so... My chip is no good, and I said, it would need 1.55v to 3.8GHz and I am not pusing those volts.

These is extreme air cooling for extreme good temps. If you want to compare the temps, you can just clock your quad at my clocks and volts, and then compare temps. This is about temps, not clocks. Best luck on the next time.

[]'s
Simps

[Ozzfest05]

I never claimed you said it was for extreme clocks more so just assumed with a unit making that much noise and looking gumbled with uneccessary fans you would be clocking this chip to warrant all the added troubles you went through to put a fan in every inch of the case I meant no offense but if you want to get upset that is your priority more fans dont always equal better temps

cheers

[Simps]

I never claimed you said it was for extreme clocks more so just assumed with a unit making that much noise...


This time I stop reading here. Lots of fans don't mean the system is noisy. You can have 10 very quiet fans together, and they will never ever sound as loud as a loud fan. My system is very quiet. Most of my fans are 25dB, and some are 28dB. CPU fans are 35dB Panaflo's, but they are undervolted for 24/7 and probably running 28dB too. I could even lower the noise if I want too, but there is no reason, since case is closed, the system is VERY quiet.

Looks like you keep making the wrong assumptions all the way hum?
First it was that my thread was about high clocking, and now, that my system is noisy...

Well, looks like you have a lot to learn...

Cheers!

[Vapor]

Underhanded sarcasm and aggressiveness will not be tolerated :rolleyes:

Everyone has a different definition of quiet...some can tolerate a 3000RPM or a system with a bunch of 2000RPM fans while others think their system with 5x 800RPM fans makes too much noise.

[Vapor]

Oh and FWIW, 15x 25dB fans is equivalent to just under 37dB....which is a far stretch from quiet according to most here. :S

[Simps]

You said it all Vapor. Acceptable noise levels will vary from one to another. I can link you to some forums about 100% passive cooled systems, where for those guys, even a 800RPM fan will be labeled "noisy fan". They won't tolerate fans.

I also can link you a lot of guys, that runs highspeed 120mm fans, 44dB or so, and they are fine with it.

I am no where near those two extremes. In fact, before the highspeed 120mm fans, you have the mediumspeed ones (2000rpm, 38dB's).

I am at the lowspeed 120mm fans, 1600 - 1700RPM or so.

And my system is, for me or course, VERY quiet. I doubt it reachs more then 35dB (I have fans at 35dB so I can compare). But then again, this opinion will vary from one to another. You post that you are using 800RPM fans on the 100% passive cooled systems, and they will flame you. It is all personal opinion. Aside from the extremes (very noisy and 100% passive), I am on a very quiet system :)

And I guess the cooling power of the system, is very good, as pics show.
You can't just talk about noise, without talking about performance. For me, it means nothing to have a zero noise system, with mediocre performance :) Also, a very noisy system with insane performance is stupid for me. I believe my system has a good deal between noise and performance, and that is where I choose to be on the "noise x performance" chart :P I don't see what is wrong with that? Maybe people should respect my preference, maybe that is what is wrong with this :)

Cheers!

Thanks for your contribution!

[Ozzfest05]

I may have came off a bit aggressive I apologize I posted at work after having 2 1/2 years tenure they gave me a joke shift at best, to work for the next year, I get sunday and tuesday off one day I work four hours the next 2 10 hours then another 4 and another 10 hour. I was a bit on edge I can assure you I understand a balance between noise and performance having two 133 cfm fans on my single stage my 24/7 rig is h20 cooled to keep the other people in this house happy they dont have to listen to my rocket ship all day only when im benching.