This is FUD. PPC universal binaries stayed around forever, and that was back when storage was much more expensive.
The big reason PPC seemed to die so fast was that Intel rocketed so far ahead of them in performance that everyone switched as soon as they could. A few years after the switch the speed difference was like 3-4X or more. It's very unlikely to be such a large speed difference today, though ARM Macs could see longer battery life and less heat.
I'm betting there will be a huge perceived speed boost simply because Apple will have full control over the device's stack. I'm sure they'll find ways to better optimize the OS for the hardware.
What I /am/ curious to see is if the CPU options shrink per device - i.e. will there be one A-series SoC for "pro-level" computers and one for "consumer", will there be different clock speeds, etc. We'll have to wait until they announce hardware - I'm assuming at their September/Fall event.
Intel may maintain a performance lead on certain tasks that take advantage of things like AVX, though Apple may be able to make up with more cores. AFAIK ARM has no equivalent to the more advanced vector instruction sets of X64 chips.
I do expect ARM to have an instant heat and battery life advantage. It may be sizable if you combine a more efficient chips with smaller process nodes. I expect a bigger gap here than in raw speed.
Even at a fancypants place like where I work, where they can definitely afford to replace your machine early and often, we have 36 month lifespans for our laptops. I'd expect "universal" amd64+aarch64 binaries to be available for almost all relevant software for at least 5 years from now. This article is not thinking clearly
The thing to be critical of Apple for is the vague and mysterious guidance re roadmap. That's ante for playing with Macs. After years of defective laptops, it's ridiculous that we need to read tea leaves. As a commercial user, do I wait for broken 2018/19 keyboards, or replace with a potentially obsolete device? We wont know until Apple controls the CPU pipeline.
I loved the switch to Intel, it made virtualization on par with Windows and Linux, and lately it's easy, for example, to load games from Steam. I'm glad I got a 2019 MacBook Pro, I hope to use it for many years, then I'll have some decisions to make.
13 comments
[ 2.9 ms ] story [ 39.7 ms ] threadThe big reason PPC seemed to die so fast was that Intel rocketed so far ahead of them in performance that everyone switched as soon as they could. A few years after the switch the speed difference was like 3-4X or more. It's very unlikely to be such a large speed difference today, though ARM Macs could see longer battery life and less heat.
What I /am/ curious to see is if the CPU options shrink per device - i.e. will there be one A-series SoC for "pro-level" computers and one for "consumer", will there be different clock speeds, etc. We'll have to wait until they announce hardware - I'm assuming at their September/Fall event.
I do expect ARM to have an instant heat and battery life advantage. It may be sizable if you combine a more efficient chips with smaller process nodes. I expect a bigger gap here than in raw speed.
Clickbait *ftfy
The thing to be critical of Apple for is the vague and mysterious guidance re roadmap. That's ante for playing with Macs. After years of defective laptops, it's ridiculous that we need to read tea leaves. As a commercial user, do I wait for broken 2018/19 keyboards, or replace with a potentially obsolete device? We wont know until Apple controls the CPU pipeline.
Also, Apple has announced that iOS/iPad apps will run on ARM-based Macs without any developer intervention.
Buying an Intel-based Mac now means missing out on forwards compatibility.