They're fast, but they'll never even remotely reach what a mid-range desktop PC with dedicated graphics burning 500W is able to do.

A 300W GPU released in 2025 is about 10x M5 perf. The difference is going to be smaller for CPU perf, but also not close.

▲

Aurornis 2 hours ago | parent [-]

> The difference is going to be smaller for CPU perf, but also not close.

This is not true. The recent MacBook Pros are every bit as fast as my Zen 5 desktop for most tasks like compiling.

For GPU there is a difference because both are constrained by thermal and power requirements where the desktop has a big advantage.

For CPU compute, the laptop can actually be faster for single threaded work and comparable for multi threaded work.

Anyone claiming laptop CPUs can’t keep up with desktop CPUs hasn’t been paying attention. The latest laptops are amazing.

▲

chmod775 an hour ago | parent [-]

> The recent MacBook Pros are every bit as fast as my Zen 5 desktop for most tasks like compiling.

Bad example. That's highly parallel, so a higher core-count die is going to destroy the base M5 here.

I don't typically compile Linux on my M5, so I don't really care, but at least online available clang benchmarks put it at roughly half the LOC/s of a 9950X, which released in 2024.

Anything single threaded it should match or even edge ahead though.

It gets for worse for multi threaded perf if you leave behind consumer-grade hardware and compare professional/workhorse level CPUs like EPYC/Threadripper/Xeon to Apple's "pro" lines. That's just a slaughter. They're roughly 3x a 9950X die for these kinds of workloads.

▲

Aurornis an hour ago | parent [-]

> Bad example. That's highly parallel, so a higher core-count die is going to destroy the base M5 here.

The base M5 starts at 10 cores and scales to 18 cores. The performance is similar to high end dekstop consumer CPUs.

> I don't typically compile Linux on my M5, so I don't really care,

If you don't compile large codebases, why do you care then?

I do compile large codebases and I'm speaking from experience with the same codebase on both platforms. Not "LOC/s" benchmarks

	▲	chmod775 6 minutes ago \| parent [-]
		I don't compile Linux or other large C projects on my M5 (why would I). The only thing I have numbers for on both desktop and mobile is your typical JS/TypeScript/webpack shitshow that struggles to keep a Threadripper remotely busy. Might as well do that on the M5. There's a large C++ codebase I need to compile, but it can't compile/run on OSX in the first place, hence the desktop that I use remotely for that. Since it's also kind of a shitshow, that one has really terrible compile times: up to 15 minutes on a high powered Intel ThinkPad I no longer use, ~2 minutes on desktop.