Ever since Apple first announced the move to Apple silicon back in July, I’ve been thinking about it a lot. I’ve been wanting to write about it for a while now, but I held off until there were more concrete details. Well, now that the first Macs with Apple silicon have been announced, I can share my thinking on the subject, and what these new systems might mean for creatives.

New architecture

If you can indulge me geeking out for a minute, the first thing to understand is that Apple silicon is based on a significantly different architecture to the previous intel Macs. For a start the Arm chips use a “RISC” architecture, whereas intel uses “CISC”. I’m not going to get into the differences here. This was discussed to death back in the late 90s when apple was using the PowerPC platform, but things have come a long way since then. Proponents of both architectures argue that their side has inherent benefits, and while previously the CISC architecture of Intel’s X86 platform won out, we’re now seeing a renewed interest in RISC systems, particularly those based on ARM.

Since the announcements, there has been a lot of commentary that ARM chips are just for mobile devices, and are not “real” processors, and have no place in the desktop. But this couldn’t be further from the truth. The fastest super computer in the world) runs on ARM. Fujitsu has one of the fastest processors in the world right now, with it uses in servers and high performance computing platforms, and it is based on the ARM instruction set.

Over the past few months I’ve been looking into this a lot, because I’m curious to understand how Apple’s new systems might scale to the high end, such as a future Mac Pro. Initially I was worried, but after reading about these high end ARM chips, I’m not. The potential is there and the system apparently scales really well, both in terms of multiple cores, and overall performance. One of the key things they kept talking about in the literature I was reading, was the benefits of using unified memory, and I’m happy to see that being used by Apple in the M1.

But the raw processor and ARM instruction set are only part of the picture. What makes Apple’s approach to its now processors unique, is that it can add custom silicon to do whatever it wants, and accelerate parts of MacOS and other functionality, taking the load off the main computing cores. Because it controls the operating system too, these don’t have to be generic acceleration operations, they can be tightly integrated into the OS and this offers huge potential. Apple has made a big deal about the machine learning cores, and these offer dramatic performance improvements for such functions, but there’s other accelerator processes on the chip too. It has a dedicated image processing engine, and a video decoding and encoding accelerator. This offers dramatic improvements when encoding video, based on similar tests using the A14 based iPhone. I can’t wait to see the results when developers take advantage of these new options.

Apple_m1-chip-mac-mini-lg-display-gaming-screen_11102020.jpg

New Macs

The new Macs that Apple released running its new M1 chip are impressive by any standard. Apart from the amazingly increased battery life on the MacBook Air and MacBook Pro, the performance is incredible. The most important thing to remember is that these new Macs are low end computers. They are Apple’s entry level systems, yet they perform as well or in many cases better than the current high end machines.

In single core performance, they are faster than every other intel Mac currently available. Even a Mac Pro. In multi-core tests, an entry level MacBook Air performs better than even a high end MacBook Pro running an i9. This is just nuts.

The only area where performance seems to lag the current high end, is in GPU benchmarks. While the M1 is significantly faster than most integrated graphics, it’s not really close to modern dedicated GPUs. But if you look at the systems these are replacing, they all had integrated graphics, so this will still be a big step up. However, while the GPU performance is dramatically better than their previous counterparts, they still won’t compete with dedicated graphics. Which is probably why we’re only seeing the entry level computers replaced at this stage.

But the thing is, this is only the start. It’s likely that the next round of systems from Apple will probably have either more advanced GPU cores, or a separate dedicated GPU. If this is the entry level, I can only imagine what they can do on the high end. Based on the benchmarks I’ve seen reported on Geekbench (likely from reviewers who have the systems ahead of the release) the GPU performs about as fast as an NVIDIA Geforce 1050 or AMD RX560. But it may not be that clear cut. Most of these specs I could find were for Open CL tests, and I couldn’t find any Metal benchmarks yet. It’s likely that these systems are much better optimised for metal, so until they’re released, it’s hard to tell. But basically, don’t expect stand alone GPU performance on these entry level Macs.

Apple_m1-chip-macbookpro-prodisplay-thunderbolt-connection_11102020.jpg

Downsides

The new Macs sound almost too good to be true, and there are a couple of tradeoffs. The most publicised is the RAM limit. Currently, any of the newly announced M1 Macs can only be configured with up to 16gb of RAM. Also, because of Apple’s use of unified memory, the RAM is directly integrated with the processor chip, so it’s not user upgradeable. This might sound like a big deal, but the same applied for the systems that they are replacing. The previous generation MacBook Air, and entry level MacBook Pro were also only capable of using up to 16gb RAM and it had to be installed at the time of purchase.

I’m sure that when higher end systems are released they will have the ability to use more RAM.

The other disappointment for me is that they can’t use an eGPU. While this isn’t a big deal for most people, it would have made the Mac mini a possible replacement for me for my ageing Mac Pro (in processor performance, it’s nearly 3 times faster than my 12 core 2012 Mac Pro), but without eGPU support, the GPU performance of the M1 just wouldn’t be good enough. But again, these are entry level systems, so I’m perhaps expecting too much. I do hope they bring back eGPU support though.

Another limitation that some seem to have missed is that you can only use one external monitor with the laptops. Previously you could use two 4k monitors, but with the M1 based systems you can only use one (up to 6K) external display (in conjunction with the internal display). This may be a limit for some people, but again, these are supposed to be entry level computers, so I’m not sure how much of a deal breaker this will be given the target market of these computers. I have no doubt that future high end systems won’t have this imitation.

Cost performance

In its launch presentation, Apple talked a lot about performance per watt, but I think a more interesting metric should be performance per dollar (or euro). I’m not going to get into comparisons with PCs but the new M1 Macs offer a level of computing performance for the cost that is far better than the previous systems. To get the processing power of the new entry level MacBook Air in a previous Mac laptop you would have to spend around $3k. Now you can get it for a third of that price. I think this is pretty significant, and it really bodes well for the future of the platform to become more affordable.

Optimisation and Software

One of the reasons that I am so excited about this move is that it makes the Mac unique again. As it is, the Mac is mostly PC hardware with a different operating system, and while this had its advantages, its led to a lot of cross platform software not bothering to optimise, or take advantage of the Mac.

With the M1 and future Apple silicon, the systems will be significantly different, and there will be big features of the Mac hardware and software that will allow the Apple’s platforms to stand out again.

In the days of the G5 Apple really showcased the differences of the Mac with software like Final Cut Pro, Shake, and Aperture and the whole Apple FCP suite. This was originally created to showcase the capabilities of the system. That kind of went away when Apple lost interest in the pro apps after the switch to intel. I hope that someone, whether Apple or a third party, finds the new capabilities interesting enough to once again create something truly unique on the Mac and create powerful Mac only apps that really take advantage of the Apple Silicon hardware.

The power of these computers will only really be realised when companies release optimised software for them. Already some big names either have versions out, or are coming shortly. Maxon has already release Cinema 4d and Blackmagic Design has a beta version of Davinci resolve. Of course all of Apple’s apps are ported. Affinity has converted its software, and Iridient has released a version of Iridient Developer supporting the M1 Macs. Microsoft has a beta of its office suite, and there are lots more applications from indie developers already released.

There’s no word yet form Capture One, and Autodesk hasn’t set a schedule for Maya as far as I’m aware (I may have missed it if they did).

Which of course leaves Adobe. Any veterans of previous transitions know that Adobe has always been a laggard with these things and this looks likely to be the same. They mentioned that Lightroom and Photoshop are coming, but this won’t be till next year. As for the rest of the creative suite, who knows. There are real possibilities for Adobe here if they take advantage of them. The ML core could be great for their new AI based filters in Photoshop, After Effects and others, but that is if Adobe actually takes advantage of them. In the past they haven’t been great for optimising their software for the specific capabilities of the Mac, so I won’t hold my breath.

But this opens up a whole range of possibilities for other developers who want to make Mac specific apps. Skylum Luminar AI for example has the potential to really fly on these systems if designed to utilise the machine learning hardware (and I have no reason to doubt that it is). We’re also seeing some other developers commit to making use of the hardware, such as Pixelmator Pro and others.

As I said at the start of this section, the new hardware and software capabilities have great potential if someone chooses to take advantage of them, and I guess only time will tell, but there is such potential there, that I’m optimistic about the future.

Apple_M1-chip-MacBookPro-user-on-xcode-app_11102020.jpg

Should you buy an M1 Mac right now?

This is a tough question and it really depends on what your planned use is. At the moment, we have no idea how much of a performance hit Rosetta2 will take on applications such as Lightroom, Photoshop, After Effects etc. While the processing potential is there, it’s not clear how much of an issue it will be. There may also be compatibility problems with software, plug-ins etc.

Having said that, if you had been planning to get a MacBook Air or 13” MacBook Pro, these new Macs should be much faster than the previous ones, so in the long run, they well definitely be better value. But for now, maybe wait a few weeks until they get into the hands of early users to see if there is any issues with the software you might need to run, especially those running under rosetta.

The bigger conundrum is if you are considering a higher end system like the 16”. Given that the entry level MacBook Air now outperforms this, you might want to wait if you can until Apple comes out with its higher end M1X (?) based systems, probably next year. Of course if you need it now, it will still be a good computer, but you may well be kicking yourself this time next year.

Images courtesy of Apple PR

Help Support the Blog

This blog is a passion of mine and I work to bring you my own take on photography, both tutorials and tips, as well as inspiration from my own art. Running all this isn’t cheap, and so If you like what you see and you want to help keep this all going, I support this work, and my YouTube channel entirely either via sales of my eBooks and presets/styles, or the kind support of my readers. (I also have some affiliate ads).

Patreon

If you like what you see here and you find this useful, then you can help support this blog and help me keep making great content like this by supporting me on Patreon for as little as $1 a month. There are a number of options available with different rewards, such as behind the scenes content, special Patreon only videos and more. Check out my Patreon Page for more details, and a big thanks to everyone already supporting this blog on Patreon.

Buy me a coffee!

If you’d rather not use Patreon, but still want to say thanks or help, then you can feed my caffeine habit and buy me a coffee via PayPal with a one off donation to my PayPal tip jar.