How could I make my computer colder for overclocking and less noisy?

[BonzaiDuck]

Negative pressure in the extreme wouldn't be a vacuum, as in space; but as in a vacuum cleaner, or the intake side of a jet engine.

But fewer molecules grabbing heat, with low pressure? Interesting to think about it . . .

"Intake side of a jet engine." Hmmm . . .

 

[MongGrel]

If you're depending on a qualified "technician" or a friend who is computer-building "savvy," maybe water-cooling isn't for you. I think you could do better than what you now have by selecting either an AiO cooler or top-end heatpipe cooler like a Noctua NH-U14S, NH-D14 or NH-D15.

One of us here -- check the "CPUs and OC'ing" forum -- is setting up a Haswell-E system, and has managed so far to overclock it to 4.4 Ghz with a D14 cooler. Since the Haswell's have higher TDP (thermal-design-power) than your Ivy-Bridge "E" system, I think you could do just as well -- maybe better.

If you don't feel so confident meddling with your hardware, you could either let your tech-guy add the water-cooling, let your tech guy add a better heatpipe-cooler, or proceed cautiously and slowly on your own to build that confidence.

Nothing to be ashamed of. I have an old college friend (class of '70) who dropped out and later enrolled in a "computer-science / information systems" program. She's deathly afraid of touching internal computer components, and I don't think that's changed much.

A lot of us have deferred using custom water-cooling because of the risks cited by the Puget-systems author, but there are risks with everything. It IS important to understand that you could install cooling capacity that keeps severe load temperatures below 45C, and you still wouldn't get better overclocking results. Instead, you want cooling capacity that will keep temperatures low enough so that temperatures don't require higher voltage (leading to higher temperatures), and so that temperatures don't create electrical noise causing instability. There is a happy medium.

I think it's like Goldilocks and the three bears with the porridge. You don't want "too big;" don't want "too little." You just want it to be "just right."

But ask AigoMorla -- our water-cooling guru. You will have to do "maintenance." Part of the fascination with water-cooling is the "bling."

As for fans -- I get by (no -- I excel!) with only four. I don't have a noise problem. But I had to pursue some tedious work to squeeze the maximum out of my D14 cooler. You don't sound like the type of user who wants to do such things.

I wouldn't discourage you from BECOMING that type of user, though.

+1 But I have a big one with about over 10 fans, it's been pretty modified though and is very quiet.

I just don't like the idea of the extra maintenance mainly, and have even thought of some off the wall water cooling things over time.

D14 works fine here.

And it's still quieter than the HTPC I built with less fans in the bedroom, but as I have the sound turned up in there while watching movies etc I haven't worried about buying replacements for it, I used some of the old ones that I had around that had been replaced for it.

 

[Cerb]

But fewer molecules grabbing heat, with low pressure? Interesting to think about it . . .

In a computer case, you aren't varying the pressure by enough for the difference in total air molecules to be enough to matter, I don't think, and only barely enough to be measurable.

Positive and negative pressure are matters of differences inside the case and outside the case, that cause air to move where and how it does. Does the air seek equilibrium by going out all the little crevices and vents, or by coming in through them?

Perfectly neutral pressure should do the same good as positive, but is hard to reach and maintain.

 

[BonzaiDuck]

In a computer case, you aren't varying the pressure by enough for the difference in total air molecules to be enough to matter, I don't think, and only barely enough to be measurable.

Positive and negative pressure are matters of differences inside the case and outside the case, that cause air to move where and how it does. Does the air seek equilibrium by going out all the little crevices and vents, or by coming in through them?

Perfectly neutral pressure should do the same good as positive, but is hard to reach and maintain.

I can only say with my own solution (2x 200mm intakes) that you can feel the leakage where there are crevices or vents. Both the exhaust + CPU and the chassis fans (the 200's) ramp up to higher RPMs under full load. The Exhaust + CPU are limited to 80+%, because no additional speed and airflow seems to make a difference in cooling, but does make a difference in the noise level. But the signs of constant pressure can be felt in the leaks at either idle or load.

 

[schmuckley]

Volenti cooler
http://www.overclock.net/t/1392108/guide-horizontal-volenti-coolers/0_100
It is the most silent and coolest, bar none.
I run a bong cooler.
PS: I don't need no stinking win8 key

 

[ClockHound]

Volenti cooler
http://www.overclock.net/t/1392108/guide-horizontal-volenti-coolers/0_100
It is the most silent and coolest, bar none.

Those are both cool and terrifying at the same time. Used to call them swamp coolers, but Volenti or Bong is a way cooler name.

Some clever company should make an AIO Volenti. ;-)

Already have the slogans for them: "We sweat the details."

or

"Overclocking got you hot? Sweat it out!"

I'm here all week....

 

[BonzaiDuck]

Volenti cooler
http://www.overclock.net/t/1392108/guide-horizontal-volenti-coolers/0_100
It is the most silent and coolest, bar none.
I run a bong cooler.
PS: I don't need no stinking win8 key

This is obviously a superior solution, even to the traditional bong.

But it's a big -- BIG -- investment in man/person/hours. I understand how the basic bong works, but I'm a bit puzzled about this. Is it a closed system? Is the evaporation accomplished somehow inside a sealed chamber? Looks like a lot of work with Lexan panels, leak-testing -- the whole enchilada.

 

[Raskolnikov]

Meanwhile in OP's world, the Primo is still unavailable everywhere. :\

Someone at NCIX is helping me build a liquid-cooled environment for it; did I make a mistake by stating a preference for a single-loop system? From what I read, dual loop systems are more aesthetically pleasing, but don't really impact performance that much.

 

[Aikouka]

Someone at NCIX is helping me build a liquid-cooled environment for it; did I make a mistake by stating a preference for a single-loop system? From what I read, dual loop systems are more aesthetically pleasing, but don't really impact performance that much.

Whether or not they're aesthetically pleasing probably depends more on the case's spaciousness than anything. With a multi-loop setup, you need more room for another pump/top and reservoir. With a single loop, you only need one of those combinations (keep in mind that a fair number of people go for multiple pumps (redundancy) per loop), which should cut down on the space required.

If I were to build a setup again, I would still go with a single loop with two pumps.

 

[Raskolnikov]

He asked me nothing about pumps. I mentioned I wanted the x61 Kraken as heatsink (for the CPU). So I'm guessing that the loop is for the GPUs. Or maybe I'm way off base.

Summarised correspondence:

Him: Are we cooling the GPU's and the CPU in one loop or are we making two loops, one for the CPU and one for the two GPU's?
Are we also cooling the motherboard? That board specifically had a waterblock designed for it.

Lastly... if you want to use the Kraken... it would be better to design a custom loop than use an off-the-shelf closed loop cooler.

Me:
I'm not really knowledgeable about cooling. Maybe something resembling a relatively common non-overkill setup? Is it too early to to give a more or less precise price disparity between all the aforementioned options?

Him: The cost will be way different if we are doing 1 or 2 loops. I will provide you some options then.