Apple Silicon lands: Only Catalyst can accelerate the adoption of these new Macs

There’s a big difference between getting developers to write apps for a mobile platform that sells in huge volumes versus a completely unproven platform with no users.

Apple Silicon: Should you buy into M1? Not so fast

At its third event in as many months, Apple announced the first new models of Mac computers -- a new Macbook Air, a Mac Mini, and a 13-inch MacBook Pro running on the Apple Silicon M1 processor, the System on a Chip (SoC) that will enable the transition from Intel to Arm-based architecture.

The M1 is an in-house designed chip, integrating 16 billion transistors on its wafer. The 8-core chip utilizes four large cores, four high-efficiency cores, up to eight GPU cores, and a 16-core neural engine for machine learning tasks. The entire chip employs a unified memory architecture design and essentially is the big brother to the six-core A14 Bionic used in the new iPhone 12 and iPad Air. 

Apple says it is the most powerful processor and the most power-efficient chip it has ever built. But just how fast is it?

I'd rather not speculate on these things because, frankly, I feel the actual horsepower of these chips is the least important aspect of the entire architectural transition. We expect the M1 to be quite fast -- I have no doubt Cupertino has designed chips that not only match the raw performance of the current-generation Intel Macs but exceed them entirely.

The real challenge that Apple faces with these first-generation Apple Silicon Macs will be the availability of native applications that exploit the new architecture, which will compel the existing userbase to switch to the new machines from the systems they already have.

For early adopters, the first generation Apple Silicon-based systems will certainly have no lack of software to run on them. That's because Big Sur, aka MacOS 11, has been provided with multiple emulation and runtime capabilities that address several types of end-user scenarios and will permit virtually all of the App Store applications for iOS and the Mac, respectively, to function.

The first is Rosetta, which enables x86-based Mac apps to be translated to Arm-based instructions on the fly. At WWDC, over the summer, Apple publicly demonstrated some large creative and business apps, as well as resource-intensive games running on this layer. However, there hasn't been a lot of information on how well the vast majority of emulated applications run. 

Only developers who have received the Device Transition Kit (DTL) know for sure --  under nondisclosure  -- how the software they are porting functions on the A12Z-based systems. Nobody has been allowed to benchmark these legacy apps on Apple Silicon for fear of reprisal.

Developers willing to discuss this tell me most of the apps, with few exceptions, now run fine on Rosetta, but they don't necessarily run faster. That's fine for software support purposes as we wait for the balance of apps to be ported to native code, but it's not a compelling reason to switch to the new machines if you've got perfectly good Intel systems running your Mac workloads today.

Similarly, Big Sur also can run iPhone and iPad apps natively. But the App Store for iOS and iPadOS is largely consumer-focused. While some business iPad apps exist, such as Microsoft's own 365, or Adobe's Creative Cloud, most of them aren't nearly as robust as their Mac desktop counterparts written for x86.

There's no doubt that thousands of iPad apps are going to work out of the box on Apple Silicon from day one. That's a simple matter of the individual developers clicking on a check button to publish them to the Mac App Store and doing a relatively small amount of testing to make sure they work as they are supposed to, unaltered. 

But those are not going to be a substitute for the full-blown desktop apps, many of which are written using Apple's Cocoa framework and Objective-C in XCode. Those will take a much longer time to port to native 64-bit Arm code, depending on which libraries are used to build them on the Intel platform. I include Apple's own Pro Apps Bundle (Final Cut Pro X, Logic Pro X, Motion 5, Compressor 4, Mainstage 3) in this group as well.

Additionally, running iPad apps in a window will hardly take advantage of all the new system hardware or screen real estate. Sure, we want to see them because there are many iPad and iPhone apps that can fill functionality gaps, and yes, we need them for transitional purposes. But it's not the accelerant that the new Macs will require, at least to attract a large userbase immediately.

So while we wait for these big, complex Cocoa apps to be ported, what do we need to hold us over? Well, that would be Catalyst.

Catalyst is a set of APIs (Application Programming Interfaces) developers will use to rapidly port iPad apps to Apple Silicon and fully take advantage of the new machines' new capabilities. That includes all of the native user interface components of macOS, the increased numbers of threads and cores, the expanded memory, all of that. 

Not only does it work for Apple Silicon Mac apps, but it also can be used to produce Intel Mac apps for systems that can run Big Sur. That's good news for those of us who own late-model Mac Intel systems. Apple already demonstrated this with their News App, Apple Maps, Stocks, Voice Memos, Music, and also Messages, which is built into Big Sur. Twitter has also used it to port its iOS and iPad app to the Mac desktop.

Long-term, Catalyst will be the unifying API in which developers will write all apps for Apple devices. It provides a framework that allows an app for iOS, iPadOS, and MacOS to be created from a single unified codebase. That is potentially huge because it allows developers to consolidate all of their efforts instead of having parallel development teams that may not have codebases in sync with each other on features, bug fixes, etc.

All of this sounds great, in theory -- turn an iPad app into a full-blown Mac app with a minimum of fuss. So, where are these apps?

Well, right now, short of Twitter and Apple's own stuff built into beta versions of Big Sur, there aren't many of them yet. Now, it's possible that thousands of developers with (and without) DTK systems are furiously porting their iPad applications to Apple Silicon/Intel Big Sur apps. Still, we can't know just how many Catalyst apps there will be on the new Apple Silicon Macs. They just haven't shown up in the Mac App Store on Big Sur yet.

Apple really needs to marshal its forces to get developers excited and motivated for the transition. Yes, the change to Apple Silicon is exciting in principle because it's a major shift and allows the company to do some truly exciting things with its hardware. But there's a big difference between getting developers to write apps for iPhones and iPads that sell in huge volumes and exist in a population of over 700 million devices versus a completely unproven platform with no users yet. Even the Intel Mac only has maybe a 100 million systems in operation, and a good portion of them cannot run Big Sur, either.

After years of speculation, the age of the Brave New Mac is finally upon us. But will it be a big bang of a launch, with a large population of older systems displaced by the new machines in the first year or two of use, or is it going to be a slow burn, with native apps coming in slow dribs and drabs? Is Catalyst going to be what truly places the Apple Silicon Mac into prime time? Talk Back and Let Me Know.