Benchmarks surface to support Apple's claims about M3

Earlier this week, Apple debuted three new chipsets, along with new Macs powered by them. It’s the M3, M3 Pro and M3 Max chipsets, all of which are about 20% faster their predecessors. Now it’s one thing to say that on stage, with Apple’s vague graphs, it’s another to see this performance in real-life.

Thankfully, benchmark scores have started to surface already. There’s a new benchmark result on Geekbench which show that the base M3 chip had a single-core score of around 3,000 while the multi-core was roughly 11,700. Now if you compare that to the base M2 chip, which had 2,570 for single-core and 9,600 for multi-core, that comes out to being about 16% faster on single-core and right at 20% for multi-core. Backing up Apple’s claims that the M3 is 20% faster than M2.

Now when we compare this to the M1, it’s closer to about 50% faster, according to these benchmarks. That’s not quite what Apple promised, but that is still pretty good for three years of silicon. Something to keep in mind here is that benchmark scores are not always the same, and these Macs are likely not on final software yet, since they don’t go on sale until Tuesday.

Apple compared to M1 and Intel because of customers

During and after Apple’s ‘Scary Fast’ event, there was a lot of talk about why Apple decided to compare the new M3 chips to its M1 and the Intel-based Macs, versus M2. Saying that it was all because the numbers look bigger that way. And while that might be true, I believe there was a bigger reason for this.

It’s said that about 50% of Mac users are still on an Intel-based Mac. The last Intel-based Mac was announced over four years ago. And the M1 was first announced in 2020, so those with those Macs are more likely to be getting ready to upgrade versus those on an M2 Mac. Unlike Intel, AMD and even Qualcomm, Apple is not selling these chips to their partners. They are selling directly to the consumer. So while others will compare it to their previous generation, Apple comparing to older chipsets makes a lot more sense. Target those that are almost ready to upgrade, versus those that bought a MacBook Pro in the last 9 months.