Ordering a new cooling setup, help me build it!

Cooling Setup Complete, need fan control help.

EDIT: Parts purchased and installed, need help with fan controlling, see last post.

Okay, so I run an Antec 300 case with an Nvidia GTX275, AMD Phenom II 955 Black Edition Quad Core Processor, a 600W OCZ PSU, 4GB RAM, and the two basic case fans...already you're thinking wow lol, because my cooling setup sucks. One fan sounds off-balance too . I just got a whopping $500 check to use on just this, and I'm looking for a kind of conservative but not TOO, cooling setup. I have low-level fan profiles set up with Rivatuner. I run my GTX275 OC'd, and experimented with very detailed overclocking of my processor with varying wattages, and got good performance, but soon realized that I was reading the wrong temps and my CPU was reaching temps I wasn't comfortable with. Playing games like HL:2, Garry's Mod, BF:Bad Co generally around 4x AA 4xAnisotrophic @ 1920x1080 native res.

Here's a basic overview of all my stats:




I'm not looking to spend all $500. Antec 300's take one 140mm top fan, two 120mm front fans, one 120mm side fan, and one 120mm rear fan. Five fans. I'd prefer to not run a fan controller and have everything automated in something like SpeedFan or Rivatuner based on temps. No watercooling. Pretty sure I've found a good heatsink, and I'm not sure that I necessarily need a side fan if it's going to interrupt the air flow. Two front fans would be running intake and everything else outtake.

Here's the heatsink I like:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835887011

Dimensions look good and I've read a review mentioning a great fit in the Antec 300 (measure myself, clearance is about 6"-6.5" (WITHOUT a side fan).
Would it be possible to replace the fan with something like a 120mm 4000 RPM Delta fan?

Main top fan:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835608017

4x 120mm fans?:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835200021
(Again, open to suggestions, I've got the money but don't want it to be too loud.)

And thermal paste:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835186020
(Anything better?)

Have to go to bed since I've got work in the morning, I'll add to this thread tomorrow. Please post your thoughts and I'll be sure and give any more information you want to know. Thanks everyone.

Dude just get a corsair H50. you are going to have to spend over $500 if you want anything that cools better than the H50 and the H50 is $80. or go with a dual 120mm eco system, about $120. my H50 is sitting on my AMD 965 at 4.01ghz and my temps never get higher than 55c and that is with the smoldering Arizona summer. oh and that is also with a Voltage of 1.48.

4GHZ said:

you are going to have to spend over $500 if you want anything that cools better than the H50 and the H50 is $80.

Click to expand...

Umm, thats probably a slight exaggeration.

In fact, this cooler could be considered to edge out the H50 in these tests, and its cheaper than the H50. However, if the H50 is using cool outside air it likely has an advantage depending on how good your case airflow is.

nh-d14, nuff said.

Always more helpful to support your claims. Here's some good support for it:
http://www.legitreviews.com/article/1212/6/

"nuff said" from an internet person I don't know doesn't always count for much, but if they give me a good reason to believe them its a lot of help.

Its almost 2 am and I spent the whole day writing and editing, my logic was I've posted links enough, if he's intersted google or the search feature on this forum will more than clear it up. but thanks , its good that there's at least one person here who's not as lazy as me

I.M.O.G. said:

Umm, thats probably a slight exaggeration.

In fact, this cooler could be considered to edge out the H50 in these tests, and its cheaper than the H50. However, if the H50 is using cool outside air it likely has an advantage depending on how good your case airflow is.

Click to expand...

its not much of an exaggeration. i have a $300 coolermaster Aquagate MAX CPU liquid cooling and it only stays cooler in idle. and i also said those temps are in the smoldering arizona heat. in the winter here my temps never even hit 50c at 4.01ghz in load. maybe past $300 would beat the H50 but not much, the H70 would do even better. and I use 2 coolermaster Excalibur for my radiator fans. and there is no air cool system that beats the H50, the v10 is the only one that come close from what i have seen.

coolermaster aircoolers =

$300 prebuilt stuff does not count. For $300, I can build a real watercooling loop...you can't compare on a case by case basis, you have to compare the best for each price point, given, the cheaper the cooler the better performance per dollar, so watercooling's payout in low temps is limited compared to going from stock to aftermarket air, but the performance per dollar of the h50 is quite low. I will give you that the h50 is good for tight setups, but if you have the room big air beats it.

http://www.tomshardware.com/reviews/h50-fort120-cogage,2401-6.html

that's against an old tier 1 cooler, the fort120, it was on par, a current review shows the h50 getting beat by a full 7c by the d14

I guess it depends what you consider "much"...

Both the review links I already posted show the H50 getting beat by top-end heatsinks. The H50 is roughly equivalent to high end air in all reputable testing, and top end air coolers are comparable in price.

It's a good cooler and I'm not knocking it - it's just not going to make you a sandwhich when you get up in the morning, walk your dog, or beat any other cooler which costs less than 100s of dollars more than it.

im sorry but i cant belive that for a second, no offence. my AMD which is at 4.01ghz with a 1.48 volt never even hit 55c and a room temp of 81f. they must of done this out side in 120f degree temps. there is no way that the H50 on an i7 920 at 3.5ghz would ever hit 69c. these are about as accurate as toms hardware GPU reviews. or they used a defective H50, or they used the 1 stock fan that it came with. if you use the push pull method on the h50, with high end fans and use arctic silver, you could put that 920 at 4ghz and never hit 65c

well i cant seem to disply photos on this site but look at this link.

this is my computer temps and specs:
and CPUZ: http://valid.canardpc.com/show_oc.php?id=1331955

Well just remember if you decide to get the H50 you can forget about any RMA or even a response from Corsair. I ended up taking mine back after only a few days and got my money back. For the first day it did decent then after that the temp kept rising. Two emails for an RMA got no response and I'm not the only one that had an issue and tried to get an RMA go to their forum and look at how many threads they locked as soon as someone posted about it. By the third day that thing had my cpu at STOCK hitting the same temps the stock cooler was getting (intel i7 920). @ 24C ambient.

if you use the same fans on the d14 it will outperform the h50 at low speeds, haven't seen an apples to apples at high speeds, but from what I've seen the d14 has more headroom

http://www.silentpcreview.com/article1042-page6.html - they used a nexus fan, which is just a rebranded low speed yate loon, which is a highly respected brand for its radiator performance (look up yate loon in the wc section, you'll see they come highly recommended) and the d14 still beat it. SPCR is highly regarded (if you find an issue with their methodology I'd like to know what it is because I'm continually impressed by their standards and thoroughness), as are the review sites that IMOG posted. The h50 is NOT a bad cooler, it performs with the top air coolers on the market, but it is not the best thing out there. It performs in tight spaces and high case temps (I feel like i'm repeating myself, lol), but if you have the room, a good tier 1 air solution gmv

I also think that someone needs to stand up for tom's here, they've been slipping recently and definitely have an intel/nvidia bias, but the data they gather is perfectly valid, take what you want from the written text, their test bench is solid and produces number in line with others...

I am not saying the H50 is the best cooler, I know it’s not. But I know for a fact that there is no Air cooled system on the market that beats the H50. And my HAF 932 has plenty of space for a huge cooling system.
Besides if you are here in Arizona of you get an air cooled system on a gaming rig you will regret it. I have gone through some of zalmans most high end air coolers and here in the summer they are totally useless, they are about as good as the stock cooler.

you should acctually see a youtube vid i made on my rig. look how i have it configured, and the H50 is an exhaust.

Please, before you go making these rediculous statements, read the stuff others are posting. It's not the best thing out there, it doesn't perform with custom water, it performs with tier 1 air, but its at the bottom of tier 1 air in terms of performance. It doesn't matter whether your air or water, since you're being cooled by air in the end (its the water being cooled by the air or the cu/al being cooled by the air, your temps are limited by the ambients, which are the same regardless. If you truly think this statement is wrong, I can get into newtonian cooling and the basics of heat transfer, but I don't feel that that worthwhile unless absolutely necessary. The h50 has value in that you can run hot air from your cpu directly out the case, but I feel like that benefit is minimal with most of today's cases being setup the way they are with an exhaust sucking air right off the cpu sink. The true advantage I see to the h50 is if you can use it as an intake so that it's ambient temps are your room temps not your case temps, which typically means ruining your case airflow or reversing your case airflow since the tubing on it is so short.

Now what statement did I make that was ridicules? I know all about heat transfer and the laws behind it, I have been studying for more than a year in my classes at DeVry University.

But you have to keep a few things in mind. None of what I said was ridicules; it is all true to the fullest. Liquid cools much faster than metal does; reason being that even a simple liquid cooling system in the hottest of AZ days could totally out perform the best air cooled systems. Water does not transfer heat well. An air-cooled system can only stay as cold as the air around it, liquid cooling systems on the other hand have more to it than air. You have the water thinning in the radiator along with that water also being insulated, which in turns cools the water naturally, and when fans an equipped it improves on that. Then it goes through the pipes further cooling the water, then hitting the copper base and recycling.

Study this basic equation and you will understand more: T(t) = TA + (TH-TA) e-kt
Also this equation on insulation now that I think about it: H = A(Thot-Tcold)/R

actually most of your comment makes no sense at all. you said "water does not transfer heat well" which is 100% wrong because water does transfer heat better than air because the molecules are closer together, because water is more dense than air. But the main thing to remember is that most water cooling setups are limited by the air temp, just like a "normal" heatsink. The reason is that your radiator is the only place where energy is being removed from the water, and the radiator is being cooled by air. So if you are in a 90 deg F room, then the coolest you can get the water is 90 deg F. At the same time, if you are in a 68 deg F room, then you can only get the water as low as 68 deg F.

So in reality, a very good heatpipe heatsink will only be a few degrees warmer than an expensive water cooling system. And a cheap watercooling setup will be about the same as a regular heatpipe heatsink.

4GHZ said:

Now what statement did I make that was ridicules?

Click to expand...

This statement for one:

4GHZ said:

<snip>But I know for a fact that there is no Air cooled system on the market that beats the H50.<snip>

Click to expand...

...and pretty much everything in this paragraph:

4GHZ said:

But you have to keep a few things in mind. None of what I said was ridicules; it is all true to the fullest. Liquid cools much faster than metal does; reason being that even a simple liquid cooling system in the hottest of AZ days could totally out perform the best air cooled systems. Water does not transfer heat well. An air-cooled system can only stay as cold as the air around it, liquid cooling systems on the other hand have more to it than air. You have the water thinning in the radiator along with that water also being insulated, which in turns cools the water naturally, and when fans an equipped it improves on that. Then it goes through the pipes further cooling the water, then hitting the copper base and recycling.

Click to expand...

I don't really know where to start. I'll just begin typing and hope some it makes sense.

First, read the legit reviews link IMOG posted. It proves precisely the opposite of your contention that no air cooled solution beats the H50. To be blunt - the facts that you "know" are wrong.

The H50 will not perform better than the best air cooled solutions.Liquid cooling is affected by ambient temperatures just a smuch as air cooling is. The difference with a strong water loop (and the reason they get better temperatures) is that the surface area available to run ambient air across is increased enough to pull the water temperature (and by extension, the CPU temperature) closer to room temperature than an air cooler could.

Where your argument runs into problems is that the H50's surface area is not that much greater (or even less) than strong air coolers. In addition to that, no one knows what's under the copper that touches the CPU (that I've seen), so we don't even know how well the block portion of the combo pulls heat away from the CPU compared to a CPU block like the Swiftech Apogee XT or EK Supreme HF.

As far as the paragraph quoted above, I'm sorry to say it's not accurate. I've already explained why these simple liquid solutions aren't necessarily better, so moving on - Water transfers heat better than almost any liquid out there. Water has a "higher specific heat capacity than any other common substance" (quoted from here). As such, it does a superb job at removing heat in a water loop.

To be honest, I have no earthly idea what you're talking about with water thinning and being insulated. Wouldn't insulating water make it retain the heat? Then talk of cooling it naturally and fans improving on it...no idea there. Lastly, tubing does not cool water in any way, shape or form.

Sorry if this sounds a bit harsh, that's not the intent. However, these things you are stating as 'facts' are just wrong and the record needs to be set straight.

I also should note this isn't to say anything at all bad about the H50. It has a niche and is pretty much equal to a strong air cooler. It may even cool as good as some near the top, but with a little less noise. It does have a strong reason for existence. That reason is decidedly not that it beats every air cooler on the market, nor does it compare even remotely close to a good water loop.

Allright guys, calm down, I'm getting the NH-D14. Read it fit in a mini-Antec P180 tower so I'm pretty sure it'll be fine in my 300. 4GHZ, I have no intent of OC'ing to those extremes anyways. They're both great I'm sure, and the difference is negligible so I'm going with the best bet.

On to my next questions...

This for a solid 120mm fan?
http://www.newegg.com/Product/Produ...004&cm_re=noctua_120mm-_-35-608-004-_-Product

EDIT: Actually probably this one, read a review.
http://www.newegg.com/Product/Product.aspx?Item=N82E16835608009

I'd be ordering three or four, depending on what I figure out about room for the side fan. Two for front intake, one for rear exhaust, and possibly one side fan. Getting a Noctua 140mm for the top.

And this next part is sort of noobish. The Antec fans in my case only connect to the PSU which is normal. No mobo connection anywhere. When changing fan speeds in rivatuner, am I only changing the speed of the CPU fan? I want to be able to have full control over all of the case fans. Does the mobo do this already by varying wattages to the PSU? Are case fans supposed to be linked to the mobo? Also, check out my fan speeds from SpeedFan in my first post. One of them is 15 rpm, what's up with that?

Also, I do plan on lapping my CPU before I install this new heatsink.

It will control fan speed by dropping the voltage/raising the voltage. As far as being about to control the fans from the MB you would have to look at how many fan pin outs are offered on your board. On my wifes board I can hook 2 additional fans and control them from the MB, yet on my board I think I can only hook one additional fan and control it from the MB so its up to the MB. You would have to look for a fan controller to set the fans to what you need in order to control the rest. The wifes Antec 1200 fans do not impress me at all and I ordered a couple of Scythe Gentle Typhoon 120 1850rpm fans yesterday, I intend to replace all 5 of the 120's in her case before its over.