8 comments

[ 1.9 ms ] story [ 24.9 ms ] thread
Apple is going full cost cutting on this gen. Can't see how this is going to help with their sales number next quarter, their bestseller refresh (M3 MBA) is scheduled for spring/summer 24. Pros are either going to hold on their M2 Pro MBP, and M1 users are going to look for deals on the M2 Pro MBPs. M3 Pro is gonna tank hard.
This is hardly full cost cutting. M3 and M3 Max are still good improvements over their M2 counterparts. Full cost cutting would have them nerfing their entire line to target M1 upgrades only.

Clearly, Apple has decided that it has mispositioned the Pro line and decided to take the hit on adjustment now. Yes, it won't help their M3 sales, but this will probably be better for them in the long run.

Apple wants to keep their SKU chart simple. If they keep it to only three models, and considering computer purchasing habits (ie, your high end will be quite flexible on pricing), then it makes sense to have "basic/anchor", "get people who want to splurge" and then "take everything you can from people who can actually quantitatively justify the price for performance".

"Pro" isn't primarily for professionals. It just cannot be in a 3 SKU system for a product as complex as a CPU plus GPU soc.

It looks like the M2 Pro is much closer to being a Max than the M3 Pro.

I was delaying purchases until we knew in case Apple caught wind of something in the market we weren't expecting and innovated (it's been a pretty exciting year, after all), but it looks like they're back to 2017-19 "penny saved" tactics.

M3 chips the build on 3nm fabrication which is most expensive at this time. Apple hardly does cost cutting, they always go getting edge because they know they can charge premium on all their products.
I’m more optimistic for the m3, well at least barring the economy tanking, that would put a damper on sales for sure.

But Mac’s have a long life, apple’s comparison for upgrades was the m1. But there’s still people on intel Macs. I’m finally replacing a 2017 mbp.

But do I understand it correctly that you get a barely faster chip with a lower TDP?
Bit odd to shorten the release cycle and end up with that outcome
It feels weird to not include some sort of ML inference tasks when benchmarking modern compute hardware. The GPU capabilities are a major factor of the M3 upgrade it seems?