99% of programmers don't need a machine with liquid cooling. Even if they are targeting high performance hardware, development work can be run from a more modest machine.
99% of programmers don't need a machine with liquid cooling. Even if they are targeting high performance hardware, development work can be run from a more modest machine.
Does he need that much power, though? I certainly wouldn't, but for what he is doing I can see it. He's running 12 containers in VirtualBox. Avoiding VirtualBox would help a lot, but depending on what he's doing, he may not be able to (and on Mac, I think you have no choice -- he may not realise that he'll get tons better performance without it). On top of that he's doing Closure (presumably with power hungry development toys) and lots of video conferencing. I often pair program remotely with video conferencing and it can eat half of your CPU(s). Mumble and Tmux is a much better way to go, but again setting up and getting used to the tooling is not trivial.
$2K and building your own machine vs doing some pretty serious work trying to lighten the development environment -- I can see an argument for it, even if it's not the way I would choose to go.
When it comes to these things, I can't find anything but anecdotal evidence. Even the most respected reviewers usually have a sample size of one.
One fan is a lot quieter than 2-3 fans + pump.
Something like Watercool MO-RA3. This can dissipate about 500W heat fanless. (4x that with full fans) Of course the room might get hot after hours of constant use.
Combine with fanless (or watercooled) PSUs and very quiet pumps for maximum silence.
It takes a lot of static pressure to force air through a radiator, and such radiators are definitionally mounted near some kind of air intake/outlet so there is usually very little noise dampening available. Also, AIOs usually have the cheapest parts available (pumps/fans) and are lower quality than you would get from a (much more expensive) custom setup.
There are quite a few benchmarks and real world comparisons on YouTube that seem to indicate significant improvement in some workloads.