You're absolutely right, a well designed water cooled device is quiter and has better cooling than pure air. I've even seen very large machines that were water (and antifreeze mix) cooled, you really can't get better unless you go with something like freon. (gas/heat exchange)
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Chrome overheated my M1
- Thread starter slooksterPSV
- Start date
- Sort by reaction score
I used to own a PowerMac G5 with dual PowerPC processors that had a liquid cooling system (PowerMac G5 2.7Ghz dual processor). Right from Apple's factory. It was quieter than the lower-powered models with traditional air cooling.
The fluid used was some kind of antifreeze... https://everymac.com/systems/apple/powermac_g5/faq/powermac-g5-liquid-cooling-info-leaks-issues.html
My machine never developed any such leaks, fortunately.
The volume of fluid is much greater than the volume of a solid heatsink; the fluid can also be cooled over a much larger radiator -- much bigger than any metallic heatsink you could put on the processor. The larger the radiator, the slower (quieter) the fans blowing air across it need to be.
A well-designed liquid cooling system for a PC can be much quieter than an air-cooled PC. Not sure the mass-market all-in-one liquid coolers the gamer kids use count.
Only one of the G5 PowerMacs utilized liquid cooling (I've had a bunch of those, the air cooled ones as well). The reason is, they couldn't get enough air through the case to cool the CPU without using louder fans (spin them higher). The reason for this is that Apple always designed their systems with form over function in mind. They did the same with a trashcan-MacPro and ran into thermal problems.
What people usually don't do is check how much heat their source is generating and buy a properly specced cooling solution. They just slap something onto it and hope it works. The next step is to crank up the fans, because the heat sink doesn't have enough surface area. So why not start there and buy a large enough air cooling system? Instead of comparing a stock cooler to a 3x120 radiator which makes no sense at all. Funny that people never compare a top air-cooler to a single 120 radiator. Speaking of comparing apples to oranges.
You can also use a 3'x3'x1' custom radiator and passively cool a Intel/AMD/GPU desktop system. If such a heat-sink would exists and you could actually mount it somehow, you could do the same. Obviously these need to be placed outside the case where liquid cooling comes in handy.
I'm running a liquid cooling system in my non-Mac home office setup. The reason is not that air-cooling wouldn't work, the reason is the back of the case is placed close to a wall. That doesn't allow to suck cool air in on the front (it's a mesh case) and blow it out the back. I'd actually have to redirect the airflow to the top and blow it out there (I'm talking about the majority here, as some air will always go out top). So moving the air in through the front and blow it out on top, makes a liquid cooling the better choice in this case. If the case is placed further away from a wall, air-cooling works just as well.
If anyone would like to play around with this, plenty of software systems exist that allow full simulation. Matlab has a very simple introduction ignoring many aspects, but it's a good place to start: https://www.mathworks.com/help/physmod/hydro/ug/simple-cpu-cooling-system.html
You're never cooling with pure air. You always have some form of base attached to a heat plate moving the heat to the sink and then cool the surface area. Same is true for liquid cooling, you're moving the air away from a source to a radiator and use air there. Depending on the quality of your air-cooler the heat pipes transferring the heat from the base to the sink are filled with liquid or gas, the rest is a function of surface area of the sink and airflow.
See above.
Of course. and I agree air can be mostly as good as a liquid PC cooling system, but it certainly wont be near as quit.
Three models did. The original 2.5GHz dual, the updated 2.7GHz dual, and the quad-processor 2.5GHz.
One key benefit you're not addressing, however, is a liquid-cooled system can move the surface area of the heatsink (radiator, in this case) to an area that can more easily support high volume airflow, regardless of where the processor is physically. Because of that, you can use larger but slower fans to move just as much air at a lower speed (thus quieter) to cool to a similar degree.
I think we're arguing the same point. Yes, air cooling can be just as effective when done right. But it's sometimes more efficient to use liquid than to attempt to get proper airflow in a fixed rectangular PC case with fixed component locations.
Those were always the top models which replaced each other. Apple had two smaller models available in the lineup back then. So it was always entry, mid and top tier and the top used liquid cooling, while the other were in air. The 2.5GHz version also had an upgraded liquid cooling system, as they had quite a bit of trouble with the first version. It was a nice system back then, but from a heat and energy consumption point of view, they were horrible machines.
Yes, I did a few posts back. If you can hide aways a full rig, including external radiators, liquid cooling certainly can make it a lot easier to run things more silent to the point of passive cooling. A friend of mine had a fanless gaming rig, but the radiator was sitting right next to the PC case and was much bigger than the case itself.
Noctua NH-D15 provides better performance in both temperature and noise (at 100% fan speed it is quieter than almost every SP focussed fan in existence running at 25%) than any commercial water cooling setups with a single radiator that can still fit inside a case.
Looking to existing Apple products is also a place for solace.
I live in a warm but hilly part of California. I've had many iPhones overheat because I would put them on my dash to overcome poor cell reception but leaving them in bright sunlight.
Even though the reason they overheated was only partially because of the internal processing, the device still protected itself from overheating using the same method that would happen if it was under heavy thermal load.
I live in a warm but hilly part of California. I've had many iPhones overheat because I would put them on my dash to overcome poor cell reception but leaving them in bright sunlight.
Even though the reason they overheated was only partially because of the internal processing, the device still protected itself from overheating using the same method that would happen if it was under heavy thermal load.
91C? My 6-Core 10th Gen i7 is just getting started at 91C
Read As: You'll be just fine.
Heat transfer is different than removal. You still need the massive space for a radiator and fans, so they aren't smaller, just easier to place in different areas.
A large heatsink can easily be quieter because a slow fan speed is much quieter than any water pump made.
And I'm not sure how they can be more energy efficient when you need a pump plus fans compared to just fans.
The chips never failed, it was the lead free solder they've used. By that definition, everything using lead free solder is failing, which is of course not the case. Reballing the BGAs with "proper" solder did the trick.
That was a different issue though. Two actually. Nvidia had a bad run of GPUs ~13 years ago which affected most of the laptop makers, and ended up with repair/replacement programs.
There was also a faulty solder issue in a batch of MBPs about ten years ago as @GrumpyCoder mentions - wrong type of solder used for the expected temperatures to be encountered.
So neither of those applies to the OP's concern nor do they apply to CPU temperatures and throttling to remain within the specificied operating temperature range.
Last edited: