Jean-Louis Gassée unpacks last month’s Apple Mac Pro confessional, seasoned with Microsoft’s new Surface devices and recent financial results.
Lots to enjoy here, but one particular conjectural question stands out for me:
Apple’s Developer’s Conference (WWDC) will be held in four weeks. We’ll be treated to the usual discussion of whether the iPad is a laptop replacement, of course, but Mac talk could prove to be even more interesting. In particular, will Apple announce an ARM-based Mac?
That seems far-fetched to me, though rolling their own Mac processors certainly seems a logical path for Apple to take in pursuit of owning the entire stack. Given how small (in terms of number of units as well as revenue) a slice the Mac Pro represents, seems to me (and I’m far from an expert) it’d be hard for Apple to recoup their investment on a custom Mac-targeted ARM chip.
That said, read Jean-Louis’ take here. Always thoughtful.
Apple changing their chip is one thing but getting Adobe and others to roll out new software is another. I remember it wasn’t quick and painless in the past.
Just no.
No one really knows what Apple’s ARM chips are capable of except for Apple. Apple underclocks their chips.
For a moment, I thought the headline above was a link to a Business Insider analyst predicting the inevitable inadequacy of any 2018 Mac Pro. Whew.
I usually like Gasee’s insight, but woof, this was a pretty jumbled entry.
Am I wrong to assume all macOS apps need to be written and compiled in X-Code?
If that is the case, Apple can “simply” make future versions of X-Code secretly compile macOS apps that run on both intel and A-series chips for a few years. Then Apple can make the switch as soon as they track all pro apps, even the ones sold outside of the Mac App Store (Office, Adobe CS, etc.), are compiled with A-series chip support.
Oh the ARM chip discussion again. OK interesting – how many software developers would be willing to add an additional compile target to their deployment song-and-dance? I’m not seeing this happening.
Technically, XCode isn’t required. Practically, I’d guess most developers use it.
I’d imagine a fair number of simpler apps could be handled the way you describe, but it isn’t that easy for everyone.
Virtualization is a big one. You can’t run VMWare on ARM. Lots of people rely on having Windows in a VM to do a few, important tasks. Java is another – lots of developers use Macs. The Intel JRE is the best, fastest, most well tested Java Runtime. I imagine there’s something for ARM, but I doubt its nearly as good. There’s probably a bunch of other stuff like this.
There’s no way they can enable some ‘secret’ compilation. And it’s not just flip a switch – there’s always a ton of testing needed.
Changing CPU architectures is not in the least a trivial thing do to do and there’s no compelling reason to move to ARM.
This is a point people never explore. Every developer of Mac software would have to develop against two different platforms – Mac/Intel and Mac/ARM. Most simply won’t.
Hmmm. Then we probably go back to universal binaries that is a possibility. Can you imagine the impact that can make on bandwidth consumption? I also think it would be radically different this time – as in much more bandwidth – than the last time around when Apple switched to Intel chips.
Right now, there are systems running macOS on ARM in Apple’s R&D labs. This is a certainty.
Developers could build new fat binaries pretty easily. The debugging work would be minimal for many apps. It is just another arch for Clang to output. And one that is already exceedingly well-tested and supported.
The question is: “When, where and how does it make sense for Apple to release ARM-based Macs?”
I think it makes most sense at the low end. I think they could undercut the price of current low-end Macs by a few hundred dollars by using their own chips without giving up any performance.
Is that compelling to Apple? I don’t think so. What strategic advantage is there to splitting the Mac line between “consumer ARM machines” and “pro Intel systems”?
I think it makes more sense to diversify the iPad line. Maybe create some kind of docking experience that takes advantage of increasing ARM horsepower by augmenting the user experience? The iPad is the future of computing, according to Apple, but maybe it needs to incorporate some current capabilities of computers. Like the ability to use a pointing device. Or drive multiple screens. Or run XCode.
Bandwidth isn’t the issue… testing is. You’d have to test against both architectures and for what? Macs exist mostly for their power. ARM chips don’t provide more power. They’re energy efficient, but that’s less of an issue on Macs and not an issue at all on the desktops.
The move to Intel was made because it was clear that PowerPC chops were not going to keep up and thus the Mac would continually fall behind in the power game. Moving to Intel meant that any Mac with a given chipset was equivalent in power to PCs using that chip.
I need Windows…not happily, not very often and it doesn’t have to be particularly fast. If Apple were to change processors, I wonder if they could pop in a low-power Intel chip just so I can run Window on the side.
<
blockquote>Google is paying 97$ per hour! Work for few hours and have longer with friends & family! !mj197:
On tuesday I got a great new Land Rover Range Rover from having earned $8752 this last four weeks.. Its the most-financialy rewarding I’ve had.. It sounds unbelievable but you wont forgive yourself if you don’t check it
!mj197:
➽➽
➽➽;➽➽ http://GoogleFinancialJobsCash197HomeCloudGetPay$97Hour… ★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★✫★★:::::!mj197:….,……
What on earth does investment in a Mac-targeted ARM chip have to do with income from the Mac Pro?
I think that when we see an ARM Mac, it will probably be a portable entry level model, not the Mac Pro.