Cooler Master Master Liquid ML240L RGB Review

[Guest]

We test Cooler Master's ML240L RGB closed-loop CPU cooler.

Cooler Master Master Liquid ML240L RGB Review : Read more

 

[princemoga]

I just wonder what would happen if I swap my Masterliquid 240 pro fans out and get their corsair sp120 fans. Would be there any performance loss?

 

[Crashman]

I just wonder what would happen if I swap my Masterliquid 240 pro fans out and get their corsair sp120 fans. Would be there any performance loss?

Try it and let us know?

 

[damnation12]

FPI = FINS PER INCH NOT FOLDS PER INCH

 

[n0ns3ns3]

I just wonder what would happen if I swap my Masterliquid 240 pro fans out and get their corsair sp120 fans. Would be there any performance loss?

probably no significant change to any direction.
a pair of NF-F12 or NF-A12x15 would make a positive impact.
on the other hand, putting 50-60 worth of fans on another wannabe liquid cooler crap is not a good idea anyway.

 

[n0ns3ns3]

Is it me missing something or Tom's team is not including normalized noise results ?
i mean chose some level of noise around 40db and test how good the coolers compare at this noise level. much more important in day to day usage than 50 and 100 percent of completely different fans.

 

[rubix_1011]

FPI = FINS PER INCH NOT FOLDS PER INCH

Technically, FPI is 'folds per inch', because the radiator fins are actually folded copper or aluminum soldered between the radiator tubes, like an accordion, not individual 'fins'. However, both are generally accepted when discussing liquid cooling radiators.

Is it me missing something or Tom's team is not including normalized noise results ?
i mean chose some level of noise around 40db and test how good the coolers compare at this noise level. much more important in day to day usage than 50 and 100 percent of completely different fans.

Also, for normalized noise results - not all coolers reach normalized 40dB at load. I just tested a cooler that even at full fan speeds registered just barely over 30dB. Noise level is a byproduct of cooling performance; both pump and fan RPM, both of which are easily controlled by the end user, typically by fan curves, not decibels.

 

[rubix_1011]

I just wonder what would happen if I swap my Masterliquid 240 pro fans out and get their corsair sp120 fans. Would be there any performance loss?

probably no significant change to any direction.
a pair of NF-F12 or NF-A12x15 would make a positive impact.
on the other hand, putting 50-60 worth of fans on another wannabe liquid cooler crap is not a good idea anyway.

The cooler performed about as well as I assumed it would, especially for the cost ($70) but the real draw is the RGB lighting and motherboard integration of this control. There comes a point in diminishing returns - you can add better fans to a cheaper cooler, but by that time, it would have been better off to spend the money upfront on something higher-performing.

 

[n0ns3ns3]

^^ that's exactly my point

 

[n0ns3ns3]

Also, for normalized noise results - not all coolers reach normalized 40dB at load. I just tested a cooler that even at full fan speeds registered just barely over 30dB. Noise level is a byproduct of cooling performance; both pump and fan RPM, both of which are easily controlled by the end user, typically by fan curves, not decibels.

That's true.
but no one stops from normalizing to 30 and 40 which are both comfortable.
I mean 100% fans on some coolers (H100 for example) are sound like a jet engine and no one sane would use it on a daily basis at 100%. IMHO, for the majority of users it's important to know performance on a "comfortable" noise levels regardless of the fan speed.
And yes, I know that there are many coolers that do not really reach 40db - Swiftech H2x0 X2, Arctic Freezer, Noctua 14 and 15 etc ...

 

[rubix_1011]

That is the unfortunate reality with most boxed AIO/closed loop coolers...they only offer a cooling potential with the fans being the only variable since the flow rate and radiator are already defined.

The bigger issue is that most of the marketing around these is the lack of definition of these coolers vs. actual water cooling - they are touted as 'liquid cooling' and to the uninformed buyer, any form of 'liquid cooling' must be great, because anytime you hear 'liquid cooling' it is always in the company of 'excellent cooling performance'. However, there is not a benchmark for comparison...and on purpose.

 

[rubix_1011]

Also, for normalized noise results - not all coolers reach normalized 40dB at load. I just tested a cooler that even at full fan speeds registered just barely over 30dB. Noise level is a byproduct of cooling performance; both pump and fan RPM, both of which are easily controlled by the end user, typically by fan curves, not decibels.

That's true.
but no one stops from normalizing to 30 and 40 which are both comfortable.
I mean 100% fans on some coolers (H100 for example) are sound like a jet engine and no one sane would use it on a daily basis at 100%. IMHO, for the majority of users it's important to know performance on a "comfortable" noise levels regardless of the fan speed.
And yes, I know that there are many coolers that do not really reach 40db - Swiftech H2x0 X2, Arctic Freezer, Noctua 14 and 15 etc ...

Then we're looking at variables of 'what is normal', 'who is the average user' and 'what is too loud, what is acceptable'? These are overly generalized terms that mean different things to different people and therefore, no real way to quantify them. There are a great many people that choose coolers only from turning in the lowest testing temperatures, bar none. Heck, I ran dual Swiftech MCR320's with 6x 3000RPM Ultra Kaze fans on them for a few years. With all of them running at 100%, it was loud, but I had a great cooling delta.

 

[Crashman]

Also, for normalized noise results - not all coolers reach normalized 40dB at load. I just tested a cooler that even at full fan speeds registered just barely over 30dB. Noise level is a byproduct of cooling performance; both pump and fan RPM, both of which are easily controlled by the end user, typically by fan curves, not decibels.

That's true.
but no one stops from normalizing to 30 and 40 which are both comfortable.
I mean 100% fans on some coolers (H100 for example) are sound like a jet engine and no one sane would use it on a daily basis at 100%. IMHO, for the majority of users it's important to know performance on a "comfortable" noise levels regardless of the fan speed.
And yes, I know that there are many coolers that do not really reach 40db - Swiftech H2x0 X2, Arctic Freezer, Noctua 14 and 15 etc ...

We got some formulas from another reader who wanted it normalized at a given temperature. Ideally we'd do a dozen or two test points to create a reasonably accurate curve, then chart those in every possible way correct?

 

[n0ns3ns3]

We got some formulas from another reader who wanted it normalized at a given temperature. Ideally we'd do a dozen or two test points to create a reasonably accurate curve, then chart those in every possible way correct?

If you had infinite resources, it would be great. Also fan efficiency chart as a function of RPM would be great. Max thermal load (or thermal capacity or heat dissipation) at few delta T(ambient - liquid or component temp) would be also interesting.
In reality, there are only few points of noise level that are important.
We all agree that the maximum performance is important so what a cooler performs like @100% RPM and at what noise level is kinda must.
But other point of 50% is mostly meaningless.
Back to the sound, 30db is an ambient noise of a quite room. 40db is not annoyingly loud (though depends on tone).
So why not to take any (or both) of these noise levels and put them into the chart ? I honestly doubt it would take more than 30-40 mins per single noise level on a 240-280 AiO liquid cooler and much less on an air cooler.
If doing a chart from max performance towards 0 RPM, it would take may be 10 minutes per level. so even doing absurdly fine measures of noise and temps at 5% RPM steps, it will take 2-3 hours of automated test cycle ?

 

[Crashman]

We got some formulas from another reader who wanted it normalized at a given temperature. Ideally we'd do a dozen or two test points to create a reasonably accurate curve, then chart those in every possible way correct?

If you had infinite resources, it would be great. Also fan efficiency chart as a function of RPM would be great. Max thermal load (or thermal capacity or heat dissipation) at few delta T(ambient - liquid or component temp) would be also interesting.
In reality, there are only few points of noise level that are important.
We all agree that the maximum performance is important so what a cooler performs like @100% RPM and at what noise level is kinda must.
But other point of 50% is mostly meaningless.
Back to the sound, 30db is an ambient noise of a quite room. 40db is not annoyingly loud (though depends on tone).
So why not to take any (or both) of these noise levels and put them into the chart ? I honestly doubt it would take more than 30-40 mins per single noise level on a 240-280 AiO liquid cooler and much less on an air cooler.
If doing a chart from max performance towards 0 RPM, it would take may be 10 minutes per level. so even doing absurdly fine measures of noise and temps at 5% RPM steps, it will take 2-3 hours of automated test cycle ?

I can ask him if he wants to start collecting 40db temperature and noise data as a replacement for the 50% fan data. Both allow readers to create a mental cooling-to-noise slope, but the 40db numbers might be more useful directly. Another option would be "the lowest speed that keeps the CPU under 100 °C", plus full speed. Heck, that might even be more useful than either of the other two options.

 

[n0ns3ns3]

well, anything that allows to compare coolers directly beside max performance would be useful
though 100C is hard as users have different ambient. also when they put it into a case, a fan can play a big role depending on location (not all AiOs come with SP optimized fans)

 

[Crashman]

well, anything that allows to compare coolers directly beside max performance would be useful
though 100C is hard as users have different ambient. also when they put it into a case, a fan can play a big role depending on location (not all AiOs come with SP optimized fans)

So, maybe 75 °C over ambient