The comparison for the use of 'efficient' is watt and performance to watt. The initial cost is probably double for the Mac Studio, locking in a whole lot of MWh. I see a something like a 60 W difference in average power consumption. So about 15 hours per KWh saved (€ 0,25 per KWh). Say a € 1000 price difference is 60.000 hours of usage. That's nearly 7 years continuous use, or 1500 office weeks or 30 office years.

I hope the future is ARM though. But for general computing, not Apple-specific. Tiny, embedded, silent computers with a low power draw, perhaps with some on the fly computing in the cloud for specific purposes like gaming.

Honestly, I think some of these reviews are just straight up stupid. There are two facts in this review: The Mac uses less power than the PC. Okay, but I have literally no idea why you would care, I have never in my life heard of anyone caring about energy draw of a desktop computer. The second fact is the Mac is quieter than his PC. But not even the reviewer cares - it's not like the reviewer went out and tried to buy a quiet desktop and found the best option was the Mac studio. He takes the Mac studio and then compares it to a generic desktop without even replacing the stock CPU cooler. It's difficult to read this review without thinking "Why are you even talking to us about fan noise if you're so unconcerned about fan noise that you've made literally no effort to minimize fan noise". We could literally say "For the same money as the Mac Studio I can buy this noisy PC and then hire a tradesman to build a false wall in my house to hide the PC so I can never hear it".

> I have never in my life heard of anyone caring about energy draw of a desktop computer

Allow me to introduce myself. I want to be able to leave my computer on all of the time so I absolutely care about power draw, especially with the sharply rising cost of energy.

If your concern is cost, then I suspect you should be focused on the $3,600 price tag rather than the 40W of power draw, which likely costs out to around 0.8cents per hour.

While a valid point, you’re underestimating the increase in power costs in Europe. Last month I paid 0.36 EUR/kWh. Yes, 36 cents per hour.

Yes but the power difference here isn't 1kW, so over an hour it's not 36 cents. At peak usage the difference is 100 watts, so 3.6 cents difference per hour if you're running at benchmark level performance constantly. Realistically even at 0.36EUR/kWh the idle power difference means you're probably spending 1-2 cents per hour. That's really nothing. We're talking about saving about as much as replacing 1 halogen light bulb with an LED bulb.

No offense, but I also don’t care in the slightest about idle power.

The energy consumption also translates directly into cooling requirements == noise.

I'm using a PC with a 5800X and an AMD 6600 XT graphics chip, has like 6 SSDs and 2 HDDs in it.

I also have a meter which measures its power use. It averages between $1.05/month to $1.20/month in power use. I never turn it off.

It's essentially silent, with lots of large, low RPM fans, the exception is in the middle of the night when consolidating data to the HDDs, they make a bunch of noise, but soon I'll upgrade them to flash.

One can build a machine like this for under $1000. The power usage argument for buying a $5000 desktop just won't ever get me. All my AWS instances are ARM, it makes sense for the server and datacenter, but for desktop use the consumer really cares more about price/perf.

Great power metrics but performance seems disappointing? 5600x is generation old mid range CPU.

From what I gather in the post it's not exactly a fair comparison. The Mac Studio is missing 2 E cores and it's running the compile under Rosetta.

I suspect Apple will have much better power metrics for a long time because this is an area where controlling the entire stack (CPU, motherboard, OS) really helps.

But yeah, the performance is nothing to write home about.