I’m a developer and Apple’s Silicon MacBook Pro is the fastest notebook… I’m not going to buy
So let’s cut to the chase. Geekbench test scores from a “MacBookAir10,1,” reveals a single-core score of 1687 and a multi-core score of 7433. The 8-core processor was clocked at 3.2 GHz.
Which has Apple’s new ASi M1 chip blowing past all mobile Macs, all current Mac mini configurations and even a healthy portion of the iMac desktop specs.
And that includes the late-2019 MacBook Pro with the Intel Core i9–9980HK processor clocked at 2.4 GHz.
So their lowest-end MacBook Air with ASi beats the eight-core Intel Core i9 used in the high-end 16-inch MacBook Pro.
My current machine.
Think about that for a minute.
Those are some impressive specifications. And the new ASi 13" MacBook Pro and Mac mini models with better thermal management should let them maintain that pace even longer than that of the fan-less MacBook Air.
They should be blazingly fast.
But I’m still not going to buy one.
Let’s find out why.
Let’s say I was in the market for the new 13" MacBook Pro.
According to Apple’s website, the maximum I can spec that machine is with 16GB of RAM and a 2TB SSD. Whereas my current 16" MBP has 64GB of RAM and 4TB of SSD storage.
The biggest limitation is lack of RAM.
Then again, as I mentioned in an earlier article there’s a YouTube video where an editor is using LumaFusion on a 4th Generation iPad Pro to edit a 4K video. He was playing back multiple 4K streams and doing effects and color grading in real time. The final 4K render produced by the software was also just a few seconds off from fully rendering in real time.
This was on an iPad with just 6GB of RAM, and it wasn’t even breathing hard.
So does a MacBook Pro with ASi need 64GB of RAM? Is a direct comparison between ASi and Intel chip performance even possible? Or valid?
Apple’s SOC designs and tight software integration with its hardware has always let it get by with less RAM, typically half that required by its Android breathern.
In Androids case, this is partially explained by Android’s memory management and garbage-collection architecture used in the JVM, where still-active objects are copied and moved from one section of the heap to another. This basically means that you need twice the space (e.g. the old space and the new space) to perform the same tasks.
Another example is that bitmaps and textures can be handed directly from the CPU to the GPU cores without needing extra space on the GPU side, which in turn means that the space needed by an internal GPU is reduced, which in turn frees up more of the shared system memory. (This was also alluded to by Craig Federighi during the product introduction.)
So it’s entirely possible that Apple’s new MacBook Pros, with their new unified memory management system and much, much faster SSD storage simply don’t need as much RAM.
But since that RAM isn’t upgradable, I think I’d want to be positive of that.
Again, my current MacBook Pro has 4TB of SSD storage, whereas the new M1-based machines all tap out at 2TB.
I could get by with less, I suppose… but I don’t want to. It’s nice to know that I have about a TB and a half of space available if and when I need it. And cutting back to 2TB would mean that once more I need start playing games with storage management and deciding what lives on my main drive and what lives on a backup.
And again, I just got past worrying about that.
The new M1 machines have two Thunderbolt ports instead of the four I have now. And while I don’t typically use all four at once I’ve done so between massive reorganization projects and having only two would mean I’d have to start thinking about docks and so forth.
And once more, I’d just gotten past that.
The new Air and MacBook Pro use the same 720p camera used in the current generation of machines… and it’s time for Apple to do better.
Apple states that quality is improved using the M1’s machine learning capabilities and I believe them.
It’s just that I keep thinking about what those same capabilities could do with a better camera to work with.
Limitations: Boot Camp
As of today, Apple Silicon Macs preclude the capability of running Windows in Boot Camp or via virtualization with Parallels.
And then there’s the whole x86-Docker thing.
This set of problems doesn’t affect me, but it could be an issue for some.
This one is more personal, and not quite in the same vein as the other limitations, but the thing is, if I get a new machine I’d kind of like it to be a new machine, and not just a new chip in the same industrial design that I’ve lived with for almost a decade now.
The Real Reason
So there are a lot of good reasons as to why I’m not going to buy a new Apple notebook.
But it’s not the main reason.
In an earlier article discussing Apple Silicon performance I included a quote from Apple’s Craig Federighi where he was discussing the A12Z chip used in the Developer Transition Kits.
“It’s not a basis on which to judge future Macs, of course, but it gives you a sense of what our silicon team can do when they’re not even trying,” Federighi continued. “And they’re going to be trying.”
And, to me, that’s the number one reason why I’m not buying one of Apple’s new Silicon M1-based Macs.
This is what they can do with a low-end chip today.
And I want to see just how far they can take it tomorrow.
Come on Apple. Blow my mind.