Do water cooled memory blocks really work?

[Airbrushkid]

I am thinking about getting a MIPS RAM Freezer like this




And like to hear from anyone that has used a memory water block. Are they really effective? I am using a OZ ram cooler right now. And from the looks of the MIPS ram freezer it doesn't seem like it could do it. With a Fan cooler you have air moving thru the ram but with the water cooler water only moves in the block that sits on top of the ram. Hope to hear from you guys on this matter.

[Ketzer7]

I don't think there's ever been a controlled test comparing the MIPS block to any of the RAM air coolers and measuring their effectiveness, unless someone has done their own personal thing..

Part of me thinks more of the usefulness of the MIPS block is to be consistent in going all the way with water cooling in the setup or for purposes of removing any noise that might be generated by the RAM air coolers. Don't know yet about the newer 60mm-based RAM coolers, but the older 40mm-based Corsair air cooler was pretty loud, especially at full bore.

[Stealth42o]

I know I read on here somewhere that it may help with temps, but does not help with over clocking.

[gmat]

Don't know yet about the newer 60mm-based RAM coolers, but the older 40mm-based Corsair air cooler was pretty loud, especially at full bore.

I have the 60mm one, and it's loud as hell unless turned down almost all the way (mine's running on 4.5V now) and even then, they are far from quiet.

[Serpentarius]

I am thinking about getting a MIPS RAM Freezer like this




And like to hear from anyone that has used a memory water block. Are they really effective? I am using a OZ ram cooler right now. And from the looks of the MIPS ram freezer it doesn't seem like it could do it. With a Fan cooler you have air moving thru the ram but with the water cooler water only moves in the block that sits on top of the ram. Hope to hear from you guys on this matter.

it helps ... but depends on your RAM heat output ...

some benchmarks revealed the heat output is as high as >50°C, but some heat output is low as 27°C

well since lower temps gives better o/c bandwidth especially on ddr3 (assumed), but the only troubling is whether is worth cooling it (+max squeezing extra bandwidth) or getting another RAM stick
http://www.techpowerup.com/reviews/M...Freezer/4.html

[Humminn55]

I have the 60mm one, and it's loud as hell unless turned down almost all the way (mine's running on 4.5V now) and even then, they are far from quiet.

That's why I use one of the Antec Spot Coolers on low. Nice and silent.

[Humminn55]

it helps ... but depends on your RAM heat output ...

some benchmarks revealed the heat output is as high as >50°C, but some heat output is low as 27°C

well since lower temps gives better o/c bandwidth especially on ddr3 (assumed), but the only troubling is whether is worth cooling it (+max squeezing extra bandwidth) or getting another RAM stick
http://www.techpowerup.com/reviews/M...Freezer/4.html

1. I don't think you can make the lower temps=more bandwidth assumption at all without any proof.

2. Those high temps were on DDR2 memory at 2.1V, not at the 1.6V DDR3 is typically run at. Even the DDR3 they tested, run at 1.7V, was only heating to low-to-mid 30'sC. It took 1.8V or greater to approach 40C, and that was without any sort active cooling.

[DeathWalking]

1. I don't think you can make the lower temps=more bandwidth assumption at all without any proof.

The maximum stable clock speed of an IC is a function of heat, voltage, and overclocking capability. Heat is also a function of voltage, but also of rate of heat dissipation. As heat decreases, the voltage required also decreases, but as heat decreases, the voltage required also decreases. It follows that lowering heat decreases the voltage required. Therefore, decreasing heat will increase the overclockability of a part, assuming sufficient headroom exists within the physical limitation of the part.

[Dreddy71]

... As heat decreases, the voltage required also decreases, but as heat decreases, the voltage required also decreases...

lol. i think that may not be exactly what you ment to say... your just repeating yourself lol. well, im pretty sure this is supposed to be:

As heat increases, the voltage required also increases, but as heat decreases, the voltage required also decreases

i think...

[damtachoa]

you don't have to think too much. just use it as decorating your wc rig.

[SoulsCollective]

"Do water cooled memory blocks really work"?

Well that depends entirely on what you mean by work. Do they allow the passage through of coolant to cool your RAM? Yes. Do they assist in obtaining a higher OC? No. Do they look cool? Yes.

/thread

[Utnorris]

It probably will not allow for more overclocking headroom, however most use it to eliminate fan noise and for looks.

[Airbrushkid]

From looking at all the pictures I have found the memory water block thing doesn't cool the memory. It cools the top of the right angle piece of metal. It doesn't make it down to the chips.

[naokaji]

Unless you have very little airflow and a very high oc on the ram it simply won't run hot enough to justify any extra cooling, but if you ever held a mips block in your hand you will still wan't to cover every chip you can find with a mips block.

[Kenjiwing]

Its purely aesthetic.

[Zeus]

Well, judging by the warmth i can feel when i touch the top fins on my Dominator GT modules, i think replacing those fins with a waterblock will help to dissipate this warmth.

This is more due to the nature of the heatspreader of the Dominators, for other modules the contact with the chips isn't good enough to justify the hassle of mounting a waterblock on top imho.

Also keep in mind you're adding more restriction to the loop. It looks nice though.

[JoeBar]

For me it's only for bling...

[Vinas]

For me it's only for bling...

Agree. Although my old ballistix tracers would get really hot to the touch. So much so I had to add a spot cooler, and that was on my old antec 900 (good air flow).

So depends on the ram. I doubt I'd ever be bothered with W/C the ram though TBH. The dominators I use now have the famous air cooler.

[wez]

For good looks and to get rid of some fan noise

[DeathWalking]

lol. i think that may not be exactly what you ment to say... your just repeating yourself lol. well, im pretty sure this is supposed to be:

As heat increases, the voltage required also increases, but as heat decreases, the voltage required also decreases

i think...

Good call. Yes I msesed up