Thoughtful post from Marco Arment on the current Mac Pro design, with collected thoughts on where Apple should go with the model currently in the works.
This particular paragraph captured my feelings about the 2013 Mac Pro design:
While minimalism is one aspect of one view of good design, it’s often overused, underconsidered, and misunderstood, resulting in products with surface-level appeal that don’t actually work very well because they were optimized for visual design and minimalism rather than overall real-world usefulness.
And from the wrapup:
There is no single design, no single set of trade-offs, that addresses a large set of pro users: they all want different things, and the only way to serve that with one product line is to have it be extremely versatile and offer a wide variety of configuration options. You can’t do that with a minimalist industrial-design indulgence like the 2013 Mac Pro.
Good read.
Marco should lead the mac Pro design team
not really. despite his popularity within the mac devs echo chamber, his opinions are often misguided.
in this case, the MP’s problem isn’t that they designed it “for minimalism”. they designed it for parallel GPU processing (multiple GPUs rather than one fat GPU), but this trend did not come to be in today’s computing tasks as they predicted it would. Inside the cylinder case the layout is divided into a triangle of equal heat generators — CPU, GPU1, GPU2. Replacing one side of the triangle with a fat GPU throws the thermal dissipation out of whack. They struggled to work with it behind the scenes to find a way until realizing it simply wouldn’t work and a complete re-do would be necessary.
Gruber was a much more nuanced grasp of this and talks about it in his podcast.
In hindsight, I guess…the two paragraphs above seem very obvious. Can we blame Apple for lack of foresight? Or were they blinded by dreams of design accolades?
I’m not sure you read his article.
Neither, I think. @mdelvecchio got it right by noting that Apple bet on parallel GPU architectures and were wrong. Making the bet wasn’t their mistake, that was in taking so long to realize they were wrong and that they needed to start from scratch.
Marco’s also not correct, I think, when he identifies gamers and PC builders as Pro customers. They’re certainly DESKTOP customers but the Mac can’t win those segments since the first doesn’t really care about the OS being used and it’s so easy to build a high powered gaming machine for well under $1000 and the second, being budget conscious, likely aren’t customers for a Pro machine that will probably start at $2500 or more.
Marco tends to meander between being dogmatic and thoughtful. Both the blogpost and his comments on ATP do take variations on this stance. While he is a bit more nuanced later, the first part of his post stands poorly, IMO.
Summarizing Apple’s design as “minimalism” does it injustice. Apple strives to find the next-generation form factor and computing model before its use is here. As part of this, they’re obviously driven to relentlessly downsize as is the rest of computing.
They’ve been shockingly successful. For example, Apple took the lead on eliminating legacy, non-USB ports and now we look back and wonder why everyone else took so long. I bet the same will be said about the 2016 MBP and USB-C, but we’ll see.
The 2013 Mac Pro (I have one on my desk) was a too-risky bet in this regard, and obviously it failed. On that part I think everyone agrees—they took too much risk. But I totally see the vision with this trashcan and why it was appealing back in 2013.
But the point he’s making is that the 2013 MP was designed for too narrow a use case, and he has a fairly convincing list of different kinds of pros and their different needs. The 2013 MP worked for only a small subset. The argument is that it needs to be a machine open to different kinds of needs, capable of different types of loads. I think the argument is sound.
“he has a fairly convincing list of different kinds of pros and their
different needs. The 2013 MP worked for only a small subset.”
When I read Marco’s post, I nodded my head at that part, but I shook my head at his conclusion that the best path is a big tower box into which you can put any kind of thing.
Apple had the right approach in downsizing the MP. The big heavy tower full of empty space to accommodate every conceivable kind of expansion needed to be done away with. Where they went wrong was in saying “well, if you need things that don’t come with this, get an external box from some third party.”
They should have either made expansion boxes themselves, or partnered with 3rd party companies to have guaranteed to work expansion boxes (basically a spinning disk drive box and a PCI slot box) available at launch for the customers who needed such things.
Making it so the MP and the expansion boxes could stack/latch together would have been icing on the cake.
Yes, I agree with much of that part. The high-level perspective, though, lacks any kind of vision and is way too summary-judgmental. “Apple just says no and is minimalistic” is a very clumsy and unimaginative line of thinking.
IMO, if you took something a bit bigger than the 2013 Mac Pro, made sure the thermal design worked, and allowed for more modular storage and CPU/GPU replacement (PCIe?), you could build a pretty amazing machine. It doesn’t have to be a Ye Olde Tower.
Its more than just the dual GPU design. Its also no support for multiple CPUs, no ability to upgrade GPUs in the future, no expansion slots (the combination of the first two meaning no path to 5K displays), minimal internal storage, etc. They shaved away so many useful capabilities that the Mac Pro ended up being a machine that almost no Pro actually wanted.
I agree with much of what Marco wrote but I’d take it further: Apple has leaned too much toward minimalism and flatness in their software designs as well. A bit too much emphasis on looking good rather than being good or put differently, it seems like functionality suffered a bit in order to have a minimalistic design. I say this as a long time Mac user who thinks Ive’s push toward better designed and pleasing looking computers is a great thing, but it seems to me that things have gone a bit too far: too much flatness, too much thinness and in the process, a bit of functionality and “versatility” has been lost.
I’m not what you’d call a pro user. But I love to tinker.
Windows PCs have enabled me to tinker for years. But I hate Windows. If only there was a Mac that allowed me to tinker as such.
Mind you, I have replaced a hard drive in an iMac. But what an emotional roller coaster that was!
It shouldn’t be that hard. And I’m pretty sure I’m not alone in feeling like this.
good read.
i also think, that the macpro failed because it was too expensive while at the same time taking away features. in theory, offering a powerful main computing unit and making it modular via thunderbolt is a sound idea – but they probably didn’t think it all the way through. not enough software supported the dual-gpus (and even apple waited until sierra to officially support eGPUs – even now it’s probably far from “plug and play”), replacing an old mp with a new one required some folks to (re-)buy a lot of now more expensive external hardware (if that type of hw was even available), the hardware was never even slightly updated to address complaints (e.g. no tb3, no optical audio i/o, no new cpu/gpu updates or price adjustments,…) and apple probably could have done more to push their vision (court developers & industry, communicate earlier and more often…).
i think the “we painted ourselves in a thermal corner”-narrative fails to acknowledge that they also could have made it a much nicer, spacier corner to begin with.
It’s hard to know. I guess we’ll see in a year or so. I owned 2 of the old cheese graters long ago. I was excited by the design insofar as access to the innards was amazingly tidy & easy to access. But really, I mainly wanted the engineering effort to go into making the thing expandable and fast. Everything else would have struck me as distraction. It’s somehow different with my iPad and my laptop. I’m honestly not sure there were a lot of people who dreamed about a new pro form factor. I thought the trash-can design was awesome but I wasn’t the target market for that machine. I find Apple’s use of the word “modular” very interesting. If they’d support external GPUs maybe they’ll be right about an iMac with a server/workstation CPU having a very high end appeal.