Hacker News new | comments | show | ask | jobs | submit login
Mac Pro: Failure and Future (mondaynote.com)
97 points by kawera on May 8, 2017 | hide | past | web | favorite | 132 comments

They can claim wrong assumptions about GPUs, etc. but the "form over function" of the Mac Pro redesign was obvious from the start. It is even extended to their infuriating web site: there are so many animations, endless scrolling while being fed painfully little information about the product, and no obvious link to something simple and useful like a white paper to communicate the important details about the machine.

They have always assumed that people would be lured in by the interesting design and they make the real details hard to find. That is clueless.

The top of the line product page should have stats front and center.

The current MacPro was designed for one group of professionals: those using FinalCut Pro. Both the software and the dual video cards were tuned to each other, and at the time were a killer combination, faster than any other setup for that workflow.

The two mistakes Apple made (and is referencing there) were that 1) the GPU market would follow down the same path as CPU's were and going with dual (slightly slower) GPUs would win out over single high-temperature GPUs so they would have good replacements to choose from down the line, and 2) other software would adapt to the 2 GPU strategy and FinalCut Pro would be the leader on a new wave of software using GPUs through things like OpenCL.

Neither of these turned out to be true. GPU makers pushed hard into making single-card monsters (and it has worked well), and very few software titles have figured out (or even publicly attempted to figure out) how to make things use the second GPU on the MacPros.

They also bet pretty heavily that the bandwidth of DisplayPort/Thunderbolt would make it possible to expand the capability of the machine through external modules. Not just hard-drives or displays, these were envisioned for external GPUs and more, that perhaps Thunderbolt might become as sensible a design decision as PCI-e. Ports are the new slots!

That didn't work out. There are external GPUs today and they perform surprisingly well, but it's just not the same as having a card you can slot into your machine.

The one thing that the Mac Pro was supposed to do well, it killed at though. Final Cut on that machine can chew through 4K video without issues. The problem is the people that want that machine want more than Final Cut.

Setting aside the dual GPU issue, the Firepro GPUs in the current Mac Pro are just pathetically slow. The mobile GTX1060 will annihilate dual D700s in any reasonable benchmark suite. They were underwhelming in 2013 and they're downright embarrassing in 2017. A four pound ultrabook should not have better GPU performance than a $7000 "professional workstation". The Mac Pro chassis has more than enough thermal performance to cope with a pair of modern GPUs, but Apple have been utterly complacent about upgrades.

This dance has been going on since the first days of PCI.

We go back and forth between dual and single graphics cards every time we get the TPD down far enough that a single card can do something like 1.5x as much as the old cards without melting, because one card without all the data juggling can accomplish a lot more per clock.

This is what I want from Apple.

Laptop: ultralights (MacBook. This can be all soldered), value (MBA, or even the old 13" MB unibody, somewhat upgradable), powerful (13"/15" MBP, ram and ssd upgradable with standard components)

Desktop: all-in-one (slick design), value (smallish, somewhat upgradable), pro (flexible, upgradable)

And the pro models really need some semblance of upgradability, especially desktop pro.

In case anyone is actually reading this. 17mbp too please.

I kept mine going for as long as i could until it was just blatantly too underpowered. I upgraded to the 15in retina(magsafe style, not the new one) and i'm still grumpy about the screen size downgrade.

Where is the 4k 17in MBP? The old one was 1920x1200, it would be a perfectly obvious upgrade

This! I loved my old lunch tray.

+1 My 2006 17" MBP is still used daily (Ironically I'm writing this on an XPS 15 running Fedora 25)

I can understand why people want their Macs to be upgradable, but how is Apple suppose to profit from it? Apple is very much a hardware company, so why would they make a computer that can be upgraded with standard components. Customers would just buy the cheapest option and upgrade with third party stuff.

Sure they are absolutely losing pro customers, but does that really matter when they have the iPhones and high-end laptops? It's only the very high-end professionals that aren't able to make due with an Apple product anymore.

500$ profit from 10 million people beats 2000$ profit from 1 million people. Also, many people don't upgrade, but those who do often make recommendations for those who don't making them a valuable market segment.

I actually was all set to buy a new MBP for around 3k, but the last round's market segmentation pissed me off. They always have one cheap reasonable option and then charge an arm and a leg after that. IMO, they would be better served aiming for a 35% profit margin on everything vs the ridiculous upgrade costs for high end they currently use.

EX: 1TB PCIe-based SSD upgrade from 256GB + $564.00. When 1TB SSD's are running under 300$ and they are saving on the cost of a 256GB SSD. Think about it, if this was a removable drive you would not only save money getting a 3rd party SSD, but also end up with and extra 256GB drive. Which means they would still have a ~40% profit margin if it cost 300$ to upgrade.

Apple is perhaps the most profitable company in the world right now, and yet you believe that they've gotten their pricing model wrong?

You could be right of cause, but it seems a little unlikely doesn't it?

There are huge short vs long term questions when it comes to profit. So, they might be maximizing profit over the next 3-5 years and I am very much an iPad and iPhone customer, but that does not mean their PRO line is optimized for long term profit.

Macs have been upgradeable with standard components for a long time, and they're only just starting to move away from it, so that's a weird argument.

I agree with you that Apple doesn't directly profit from building an upgradeable machine. But from my perspective, upgradability is one of the factors that affect my purchase decision--part of the value for the package, if you will. My current computer is a 2014 MacBook Pro, so I look at it from the laptop perspective.

I am not a power user by any means, so I don't need lots of raw computing power. To me, the values of owning a MBP is design, build quality, good compromise between weight/power, OS X, and longevity. However, longevity is reduced by removing upgradability for RAM and SSD. I don't see MBP as attractive as before, especially with the price premium. I admit that I am cheap, and I try to spend as little as possible. I typically get the lowest spec'd MBP that fits my current need. After a few years, I upgrade RAM and storage (used to be HD, now SSD), and that will last me another few years.

Apple's current offerings doesn't fit my need, and I am exploring other alternatives (currently test driving Asus Zenbook 305 running Xubuntu alongside my MBP). I am not bitter, just disappointed since I much prefer Apple laptops.

Also, the lack of reasonably priced SSD upgrade for Retina MBP due to proprietary interface is driving me nuts. I am running out of storage space and can't do anything about it.

Don't forget the halo effect. If the pros leave Apple, the masses may follow.

It's possible to make a computer that's upgradable with non-standard components. People will hate you for it if they realize you're gouging them, but it's a time-honored strategy.

1. Most people wouldn't have the desire or expertise to upgrade with third party stuff. Most people don't want to muck about in the guts of their PC.

2. Apple could sell those parts themselves at a much more reasonable profit margin than they do today.

3. Better to sell the cheapest option (at a good profit margin) than nothing at all.

"It's only the very high-end professionals that aren't able to make due with an Apple product anymore."

This really needs to be qualified, as there are plenty of high end pprofessionals still served well by Apple products.

There's not a lot of choice in the high end for eg. Notebooks. The Surface Pro Performance Base comes close to the MacBook Pro, for example, but not a lot else on a balance of features. If I really need 32GB RAM (For example) I need to lug a back breaking brick from Razer , or deal with the Dell XPS 15 which can't be charged on an airplane.

Well it's a matter of whether Apple wants the workstation market. Do they want it, warts and all? If they do, then they have to adapt. If it's not worth it for them, then just give it up. People are just not gonna buy non-upgradeable workstation desktop. The world has moved on. It's up to Apple to adapt.

For me: mac -> iPhone & iPad

Non-mac PC -> probably Samsung phone & definitely not iPad

Make do, not due, in this instance.

It saddens me that a perfectly valid question is downvoted into grey here.

User replaceable battery.

Home, end, page up, page down, and the power button moved back to not being a keyboard key.

Battery will never happen, they're no longer a singular thing you can replace but a multitude of inter-connected parts, each carefully positioned and glued in place. Why is this a problem? Apple will replace it for a nominal fee should you be experiencing problems, and if it's a dramatic malfunction, like the dreaded bubble battery, they'll often replace it for free.

Because having a user-replaceable battery is much more convenient?

Convenient for who? Apple's explained, over and over, that a removable battery is a no-go in modern laptops, the amount of material required to encapsulate it and dock it eats up tremendous amounts of space.

Most people would rather have a 14 hour battery that's not replaceable than a 6 hour one that is.

I've had a lot of laptops in the last fifteen years and the only time I'd ever removed my battery was to add memory or a hard drive simply because it was in the way. Maybe one in a hundred people actually care about having a replaceable battery.

And forward delete please. I miss it whenever I'm undocked and using the built in.

Fn + Backspace. Not as easy but that's the answer.

I know, I just get used to using the actual delete key when I'm using an external keyboard, so my muscle memory never kicks in fully to use fn+backspace.

That's what external keyboards are for.

I find the built-in style is fine for casual work, even typing documents, but for programming I love my delete key too much.

On the mini keyboard I find myself using ⌘X in place of Fn-Delete, it's a one-handed thing that's more convenient. If you have nothing in your paste buffer that's important it works out well.

I have a laptop which has all these keys though, in a 15" form factor. There's no reason Apple can't put these keys in (and they used to) - they just don't want to.

For whatever reason, Apple prefers to make a single keyboard for their laptops, regardless of the size. Even the 17" had the same keyboard.

Plenty of small laptops with a dedicated delete key. Don't see why Apple couldn't add one.

Probably because the overwhelming majority of people don't really care.

Those that care about a forward delete key are in the minority. They're also the type most likely to have some kind of highly personalized keyboard with very particular Cherry switches and keycaps.

I can't help but feeling there is a trap here, which is what Gassee seems to be getting at. Macs didn't become the default choice of "pros" due to out-and-out speed. You could always build a faster Windows box, and they have always been more configurable. Macs won via the usability factor - whether that was better OS support for certain things, better UI for certain things, or some new app or feature on OS X that made it possible to do a certain task for cheaper on consumer hardware that previously required expensive custom "pro-only" hardware to run. I mean, the whole Avid revolution was that you could do previously esoteric video editing on a Mac vs. buying some horribly expensive custom setup, and the same thing happened with Premiere, and then Final Cut Pro. The whole pattern isn't that Apple used to make expensive (relatively speaking) pro hardware that was faster than everyone else, it was because they democratized some pro function and everyone used it because it was just better.

So why all the excitement for apple to make a "narrow" machine to "cater" to the pro market? That would be against what Apple has historically done. Apple can win the "pro's" over by making iPads or MacBooks able to do pro video editing that previously required a $10k windows/linux tower with dual video cards, not by making their own $10k tower with dual video cards. The whole point may not be that there is a pocket industry of "pros" upset at Apple, but that their function is being disrupted by young grads just out of film school doing their jobs on an iMac.

> Apple can win the "pro's" over by making iPads or MacBooks able to do pro video editing that previously required a $10k windows/linux tower with dual video cards

WAT!?! Imagine editing an 8k, 3D feature film on an iPad Air. The "handheld" device would end up being a 12 lb behemoth with Gilliam-esque cable madness running to a single lightning port and optional (Red) branded oven mitts.

Pro video editing (and FX, hello) requires massive power, storage, and peripherals - the more the better.

One thing I've always been curious about with ultra-resolution stuff like this:

Is there a substantive difference between editing in 8K and in say 720p?

Ie, could I edit my whole movie in downsized resolution, save all the transformations I've applied, then redo them all in 8K on a server someplace for the final product?

You're right; a lot of professional video editing that utilizes high resolution or heavily color corrected footage will first be rough cut with low resolution proxies, only substituting the full quality footage for final assemblies.

Back when REDs first came out, there was a lot of buzz from amateur and student editors who could get 1080/720 proxies of their freshly shot 4K footage almost instantly and hand it off to editors before production even wrapped. This had been the standard for many professionals doing off/online editing for a while, but getting access to that workflow for relatively cheap was the new, cool thing.

That's exactly how it's normally done - the offline editor works with low bitrate HD files, then gives the edit list to the online team who re-conform from the full resolution stuff.

There's always an impetus to get higher resolution and quality at the offline stage though, as people get used to improving technology. It wouldn't be outrageous to have a 4k offline now, because people with 4k TVs at home will say "Why can't we? I want to see everything!", even if it doesn't particularly enhance the editor's work.

It would still be done with lower bitrate files, because original 4k material from digital cinema cameras is beyond what most machines are happy with.

I would imagine that color correction, or anything involving pixel masks, would require full resolution. Cutting scenes and stuff might not, but I wouldn't be very surprised if there were cases when not having the full resolution would bite you.

The GP put double-quotes around pro's for a reason. Somebody needing to edit an 8k, 3D feature film is a pro. There are very few of these people compared to the number of people making videos for YouTube or what have you. Apple is a consumer products company today. Devoting a huge amount of resources to a machine that a handful of people working for big production studios is a bit silly.

In part, I think that's Apple's problem right now. 15 years ago, I could imagine Apple somehow building a device that was smaller, more usable and more powerful than any of their rivals. These days, they are just more expensive.

First of all no one is editing 8k right now.

No one is editing 3D right now, except a very tiny portion of the industry who are producing specialist films or blockbuster AAA movies with unlimited budgets.

By the time that those become standard and mass market, hardware will be at a place where laptops and iPads can handle the tasks.

4k is very much the standard right now, and almost any Mac right now can handle 4k editing (some better than others) and perhaps even the iPad Pro can handle it in stride.

So the realities of the market don't hold up to your fantasies of millions of people needing to edit 8k and 3D at the same time.

My 2014 15" rMBP isn't fast enough to edit 4K at acceptable latency (and it doesn't have enough storage, either). Your forecast seems profoundly optimistic.

I work in the world of video by day, but do lots of still image work as a hobby.

Even in the world of photography, there is no Mac that edits/filters/transforms/renders 30-40-megapixel Raw images without very noticeable latency. For fluidity and efficiency in what I do, I'd gladly take 10x performance over what exists today on the Mac.

Announcing streamed apps with more power than you could fit in a box! Apple approved, gce tested, NSA blue box

The app still has to reside on a box somewhere. Centralizing it in the cloud is a little absurd since only one user (the editor) would use it.

Depending on the studio, content is centralized, but for smaller studios, cloud-based storage is prohibitively expensive since they work with raw 8k/4k video files.

Contact gce for fatter pipe. Business line for those bubble pushers.

All my stuff is Apple hardware. But I went from an extremely happy client to an "okay" client: because of convenience, I don't want to start messing with Android (and having Google monitoring my like a little baby duck) and going back to Windows (hey some dissapointing "the grass looks greener on the other side"-stories) or Linux, where still, if your eye wants something, there's so many quite horrible inconsistent ugly things happening.

So Apple is still the best choice for people like me who don't totally want to break their current workflow, simply because there's no convincing great alternative out there. But the credits, or forgiveness of all this is dwindling fastly. If I would've bought this new MBP for 3500 EUR from my own money, instead of it being a company laptop - rather than my personal one --, I would feel seriously cheated -- and probably would've gone for a Lenovo instead.

Apple, under Jobs, made plenty of mistakes, but in the last years, he was pretty much spot on.

In my eyes, Tim Cook, makes one mistake after the other. And it's not just the Mac Pro, but also the Mac Mini (which is a great computer-format), the audio jacks on the iPhone, the removing of the magsafe, dongle-life, etc. So there's plenty of reasons to say basically: let's bail from this mess. The problem is that the alternatives are not so much obviously better.

GrumpyGamer's posts when compiling Thimbleweed Park for Windows were a pretty good reminder of staying away from Windows. ;)

The most interesting part of Gasee's piece was the increased ASP of the Mac line. Since Macs are 7% of the of market by unit sales, and its ASP is now almost 3x the rest of the market, it's nearing 20% of of industry revenues, and and probably 60% of pc hardware profits.

I'm a critic of the poor pace of mac updates, but Apple is making far more profits in iPhones and Macs (and overall) now than when Jobs was in charge. It's market position is far stronger as well.

> and having Google monitoring my like a little baby duck

So instead, you prefer Apple monitoring you like a little baby duck?

I'm not particularly an Apple fan, but Apple isn't an advertising company fundamentally premised on profiling users. The business is far more aligned with user privacy than Google's, inherently.

That said, Google seem to have some smarts about defending against high-level attacks, though also some very noteable failures. Remember, John Podesta's email phishing occurred on a Gmail account.

And it's Gmail which secures the Traitor Puppet Fascist Donald John Trump's Twitter account password recovery.

Purely subjective, but I'd say there is a difference in comfort levels when looking at companies where you paid cash in exchange for a product, and companies where they give you free stuff, and then eyeball you for the rest of your life, trying to find ways to extract money from you.

Yes, there difference being that in the former case you get to pay twice - once in cash, the second time in personal profiling.

Also, it is possible - and not all that hard - to run Android without any proprietary (closed) Google software. The same can not be said about iOS.

That is not what Apple says though.


>We don’t build a profile based on your email content or web browsing habits to sell to advertisers. We don’t “monetize” the information you store on your iPhone or in iCloud. And we don’t read your email or your messages to get information to market to you. Our software and services are designed to make our devices better. Plain and simple.

Thing is, I don't trust Apple, just as I don't trust other companies (Google included, which is why I don't have Google proprietary stuff on my Android devices). I trust them to the extent that their current business plan does not centre around 'monetising' profile data but that's about it. I fully expect they did in fact design their systems to make it possible to eventually 'monetise' such data.

Do I have proof for this, other than the fact that they've shown to be less than fully truthful about their data retention policies? No, I do not, but I do know that it is more rule than exception for commercial entities to eventually renege on their promises of personal privacy, probably because that huge 'big data' carrot on that stick is just too juicy to be ignored.

Given Apple's enormous cash reserve they pose the additional threat of being able to outright buy large industrial players in lucrative fields which would allow them to use the collected data to target their customers without needing to 'sell to advertisers'.

If Apple really believe that they can replace MacBook Pros with iPads, I think they will be sorely disappointed. Apart from vim/emacs hackers and sysops that can get by with just a shell and keyboard, just about everybody needs a touch pad or a mouse to be productive, and more importantly, a proper desktop OS. Yes, app/context switching will become even faster, iPad screens can be made bigger, and long battery life is alluring; but how am I supposed to mix music without proper peripheral support, edit pro video without external storage mechanisms (or even file system access), or manipulate pictures without fine mouse/touchpad control? How will I develop apps without XCode or Android Studio, debug IE CSS bugs without a VM, or write actually useful scripts without a fully blown unix shell or rubygems etc?

> I think they will be sorely disappointed. Apart from vim/emacs hackers and sysops that can get by with just a shell and keyboard

Hell what? Not even that works. Apart from the curios* I've seen no live use of anyone using an iPad instead of a MacBook. From custom window management to the simple fact that a hinge between the screen and the keyboard is just the perfect HID with a TUI.

* yes it does work and some people seem to like it (to each his own) but every attempt I've seen just ends up turning an iPad into a frankenlaptop just to end up claiming "hey I work on an iPad, iPads are fine!", which they are, but I'd rather jack a car up with a jack rather than with a beam and a log.

It all depends on what you do.

Executive and field workforce people love the ipad. When I stood up our mobile team, a group of field auditors literally sent us cookies and brownies when we trashed their corporate crapbooks with iPads and VDI -- despite significant UI issues.

As SSO has become better with iOS, people have gotten even happier. The instant ok and lower friction (yet secure) connectivity experience is huge.

The biggest issue is legacy windows apps. But people come up with solutions for that as well. In one case, a guy wrote an access front end for a Windows workflow with giant buttons that were easy to click. Ugly as sin? Yes. But it worked.

Sorry I was not clear (I phrased it badly and rereadingit in light of your comment shows how ambiguous mine was), I was solely talking about development tasks and such, notably on the terminal. I fully appreciate how an iPad (or the tablet form factor in general, including Surface) can be useful and appreciated in various use cases.

"or even file system access"

That right there tells you the iPad will never be a pro machine or replacement for a laptop or anything but a toy. I bought my sister one a couple years back. I still have no idea how to play movies on it in non Apple-approved formats. Yes, I could research it. But this "it just works" machine is one now that requires research to play a movie. From what I could tell, each app will do it differently. Yeah, that's progress all right. Progress on a toy, a toy that by design (of its OS) will never be anything more.

> I still have no idea how to play movies on it in non Apple-approved formats.

You download VLC from the App Store and use that. Not much different from Mac or Windows, both of which require you to download VLC or another third party application to get a wider range of video formats.

(And iOS does have what amounts to an accessible filesystem these days; it's called iCloud Drive.)

iCloud Drive is a filesystem? As in, it can be used without an iCloud account, without Internet access, and without copying anything off the device to a remote server? Because if not, it's not a filesystem.

The irony in this piece (which Gassée is classy enough to only point to obliquely) is that what pros are clamoring for is for an Apple that's less like Steve Jobs' Apple and more like Jean-Louis Gassée's.

The trashcan Mac Pro came out after Jobs died, but in spirit it's a classically Jobsian product: fascinated with aesthetics and style, disinterested in performance and expandability. It's striking to look at, but required a whole bunch of increasingly painful compromises on the inside in order to make that visual statement possible.

Which makes it a lot like the original Mac, which was also a Jobsian product. Its revolutionary industrial design makes the 1984 Mac a wonderful object to contemplate even today; it's like the Platonic ideal of the term "personal computer." But the original Mac was also hobbled in important ways to fit Jobs' vision; it came with just 128K of RAM (low even by the standards of the time), and no mechanism was provided for adding expansion cards or other upgrades inside the enclosure. (It took two years for Apple to come out with a version that included a SCSI port!)

What the pros want, by contrast, is a Mac that puts functionality over aesthetics; it's OK by them if it's a big ugly box, if that big ugly box comes with cutting edge parts and can be stuffed full of RAM and drives and expansion cards. Which sounds a lot like Gassée's baby when he was in charge of product design at Apple, the Mac II: https://en.wikipedia.org/wiki/Macintosh_II

The Mac II discarded the beautiful lunchbox form-factor of the original Mac completely, replacing it with... a big ugly box. But it was a big ugly box that was twice as fast as the lunchbox Mac, supported true-color graphics and had plenty of room for expansion, so the pros of the time loved it. The Mac II was the start of the long line of high-performance Macs that would make the Mac the go-to computer for people in creative industries.

So when pro users complain about there not being a Mac for them anymore, this is what they're talking about. They want a modern Mac II. They want a Mac that's less Steve Jobs and more Jean-Louis Gassée.

Great point. The Mac Pro was pretty obviously this decade's G4 Cube.

Ironically they didn't learn their lesson and thermal failure was still the Achilles heel.

In their defense, Apple hits home runs with those kind of products. I worked in a computer store when the original iMac came out... the serious computer people thought it was a joke, but it sold like hot cakes.

> In their defense, Apple hits home runs with those kind of products. I worked in a computer store when the original iMac came out... the serious computer people thought it was a joke, but it sold like hot cakes.

What "kind of product" do you mean, that includes the G4 Cube and Trashcan Pro (interesting but expensive failures), and the original iMac (decent mass-market computer that was incredibly successful)?

I mean products that break the mold.

The bondi blue iMac was an incredible gamble. It was bondi blue, looked like a hard boiled egg sliced in half, lacked a floppy drive, and was incompatible with every Apple accessory/peripheral on the market.

But it hit the mark and was a hit.

> In their defense, Apple hits home runs with those kind of products.

Apple makes plenty of turkeys; it's just when they score a hit, they hit it out of the park.

Jobs was a truck user in his famous analogy. As CEO, he clearly didn't do much design or programming or have any true pro needs, but if you see photos of his home office, what do you see? A big honking Mac Pro (and even a scale model on his desk!).

Jobs didn't need a Mac Pro and a huge external monitor, and could have clearly used a laptop or iMac. The fact that he used the company's pro computer in his home office tells you a lot about him and the kinds of computers he liked.

The Mac Pro is very much a Johnny Ive design and idea (and not updating it feels like something that a numbers obsessed CEO like Cook would do). I'm not convinced at all Jobs would have let the Mac Pro flounder like this.

I truly believe that if Steve Jobs were alive today, the Mac Pro would be routinely updated.

Don't give Gassée too much credit. It was under his leadership that Apple decided to overprice the Mac and chase after revenue and they let MS and Windows take over -- almost killing Apple.

> a classically Jobsian product: fascinated with aesthetics and style, disinterested in performance and expandability.

That's the description for a Jobsian consumer product.

All the previous Mac Pro designs (and Xserve designs!) were greenlighted under Jobs, and it's obvious that tons of thought was put into their expandability and maintainability.

The obvious example is the PowerMac G5 with its screwless design (before that was at-all common); but looking further back, you've got things like the PowerMac G3, where most of the computer lived on a slide-out tray (https://www.ifixit.com/Teardown/PowerMac+G3+All-In-One+Teard...).

The G3 AIO was obviously an "aesthetic" product (however people might feel about the aesthetics of said "Molar Mac") but it was also built to serve the education market, not the consumer market, so it had maintainability as a goal.

> But the original Mac was also hobbled in important ways to fit Jobs' vision; it came with just 128K of RAM (low even by the standards of the time)

This was about cost savings! Same strategy as Nintendo: cheap parts + lots of value-add in integration and software = huge profit margin. (But, unlike modern Nintendo, early Apple did this out of necessity—a risky investment into a new product line when you don't have much capital is even riskier if you have to make minimum purchase orders for large quantities of expensive parts. May as well design the first one to be low-end, and then, if you can't sell it, repurpose the low-end parts for something else, like, say, disk drive controllers.)

> no mechanism was provided for adding expansion cards or other upgrades inside the enclosure

I'm not sure why they did this for the Mac 128K, but I'd guess it was simply that there wasn't room on the motherboard, with the motherboard constrained by the tiny case (where the form-factor of the case was part of the value-prop of the Macintosh product: it fit on your desk without stealing your desk.)

Look at the 128K's motherboard (https://www.ifixit.com/Teardown/Macintosh+128K+Teardown/2142... )—there's no room there to add anything.

> It took two years for Apple to come out with a version that included a SCSI port!

The Macintosh 128k was essentially meant to be an all-in-one, classy "8-bit micro" (at only a slightly higher price than one, once you factor in the price of the included color computer monitor in the 1980s.) It's competitors were Amiga, Amstrad, Commodore, etc. Nobody else in that world used SCSI ports either.

Apple basically noticed the reaction creative professionals had to their not-originally-intended-for-creative-professionals product, and then started adding "pro" features to it (while still never thinking of the Macintosh line as for professionals, until the popularity of Macs pushed them to eliminate their other lines.)

> The Mac II was the start of the long line of high-performance Macs that would make the Mac the go-to computer for people in creative industries.

Note that the Macintosh SE continued this trend, and yet returned to the lunch-box form-factor. The difference was that, by then, the chips on the motherboard had miniaturized enough to let them stick some expansion sockets on it as well.

It's well-documented that at the time of the original Mac, Jobs was vehemently opposed to internal upgrades. Here's the story of how Jobs nixed a proposed slot for the original Mac: http://www.folklore.org/StoryView.py?story=Diagnostic_Port.t...

That said, he definitely came around after his stint at NeXT. Just look at the PowerMac G3/G4, with the single hinge (zero screws) fold out motherboard. Still the best computer case design ever IMHO. https://d3nevzfk7ii3be.cloudfront.net/igi/Nm4mRELLtcYI4YYI.m...

> Considering the continued decline of classical PCs, I don’t think Apple can go all the way and design an Ax processor for high-end Macs.

I think this is wrong. This is what the benchmarks say:

- Apple A10 Fusion @ 2.35 GHz [1] / Geekbench 4 single-core: 3500 (typically) - that's 1489 points per GHz

- Intel Core i7 7700 @ 4.20 GHz turbo / Geekbench 4 single-core: 5400 (typically) [2] - that's 1286 points per GHz

- Intel Core i3 7100U @ 2.40 GHz (no turbo) / Geekbench 4 single-core: 3150 (typically) - that's 1312 points per GHz

In clock-for-clock performance, Apple already is competitive with the highest end Intel performance. It's just a question on whether they can scale this up to a desktop chip with an appropriate amount of cores, clockspeed and I/O.

Before you protest, Geekbench 4 is one of the few benchmarks that tests individual core speed while being fairly independent of I/O performance. Of course any desktop processor will easily outclass a mobile chip when it comes to I/O. It's not a perfect comparison for sure, but I think it shows that it's quite conceivable Apple will use their own chips in their next Mac Pro.

[1]: This fails to take into account any potential 'turbo' functionality the A10 Fusion has - there is so little known about it that we don't if that exists or how much it boosts.

[2]: I chose the non-K variety because the 7700K results on Geekbench typically seem to vary wildly due to its overclockability.

Apple imagined a world where everybody did all their work on iPads and the laptop was a relic from a forgotten age.

We might get there one day, but NEVER while Apple maintains closed-garden approach to iOS.

Apple allowed its 'Pro' lines (less so MacPro and more so MacBook Pro) to start catering to its biggest buyers - consumers who had lots of money and simply wanted 'the best'. And back in 2012-2013, the 'Pro' lines were really well regarded amongst professional users which re-inforced in the consumers mind that they were the best.

Apple focussed on thinner and slicker looking devices and added gimmicks such as touchbar and cylindrical cases.

Suddenly we find that the MacPro hasn't been updated in years (yes Mr Schiller CAN'T INNOVATE YOUR ASS indeed) and the MacBook Pro can handle the same amount of memory as it did over 3 years ago (battery you say Mr Schiller? Well most 'pro' users will sacrifice some battery life for performance and those that don't want to do that could have gone for regular MacBooks).

The Pro lines aren't so well regarded anymore and Apple will watch slowly as those rich consumers who propped up sales dwindle. Maybe Mr Cook and co will wake up and actually realise they should rather just drop the pro line than consumerize it.

iPads could have done so well in the pro space if they'd supported proper integration with Macs and could be used as extension devices rather than exclusively as replacement devices - but no - Apple had to try will a market that simply isn't there into existence.

To do away with the walled garden approach would require a robust, scalable solution to the malware and spamware problem.

For most non-pro users, a walled garden is awesome. You can actually use a computer without it becoming progressively more and more infested with malware and without it "rotting" from badly written installers, etc. For most users closed is a feature, not a bug. Closed means it works. Closed means you don't have to do your own IT.

A real solution to these problems that isn't based around walled gardens and app whitelisting would require very innovative solutions to major problems in application and operating system security. These aren't the kinds of problems that get solved overnight, and unfortunately I don't see very many people working on it.

Ultimately I think what's going to happen is more pragmatic: the computer market will fully bifurcate into one kind of device and OS for power users and developers and another for everyone else.

> battery you say Mr Schiller? Well most 'pro' users will sacrifice some battery life for performance and those that don't want to do that could have gone for regular MacBooks

My understanding is that Apple expect everyone who uses an MBP "for work" to actually be using it as the portable sidekick to a desk-bound workstation, rather than as their primary computer. They're supposed to be the cutter you launch off the side of your ship, not the ship itself. The Mac Pro (or often the iMac) is supposed to be the ship.

You might think that the lack of updates to the Mac Pro line belies that argument—but I'd say it's completely possible to have a strategy and then fail in tactical execution in a way that makes it look like you don't have that strategy.

> Apple expect everyone who uses an MBP "for work" to actually be using it as the portable sidekick to a desk-bound workstation

If that's the case, how are you supposed to do it? Specifically, how do you keep your two accounts, files, bookmarks etc in sync between the two, especially 'live' so you can just pick up your laptop and continue?

I'd actually very much like to do this, so Apple's expectation or not it would be great to achieve.

Well, looking at what Apple defaults for people these days, you've got your Desktop and Documents living in iCloud, which is (I assume) where Apple expects you to save pretty much all your "working state." Bookmark (and tab) sync happens through iCloud as well, if you're using Safari.

Window state doesn't sync, but then, it wouldn't make much sense for it to, workflow+performance-wise. The cutter is not the ship; it doesn't have a 50-gun battery, nor room to take a 1000-person crew aboard it. You're expected to multitask on the desktop, and then single-task on the laptop (picture an executive VP with a fancy watch whose time is Very Valuable: if they're working at all while not at their desk, it's because there's one very important/urgent thing they need to do.)

As such, it makes much more sense to save your GarageBand project, or your PSD file, to iCloud on your desktop, and then re-open just that one thing on your laptop, rather than attempting to squash your whole workflow around as if you'll keep interacting with it the same way at the destination. (Though, if you're desperate to have that particular experience, Back To My Mac exists.)

(I do imagine an OS could pull off something like the Firefox UI experiment of "tab groups", with native app windows, by combining a ubiquitous cloud store with all apps being wired to save+restore window-state, and requiring that any apps installed in a "session group" get installed on all the computers in the group. You could then grab out individual pieces of your workflow and un-hibernate them onto any computer you liked, work for a while, then put them back away. But that'd require a radically different style of window management than we have today.)

My main items are:

* Chrome, with window state / open tabs

* A VM

* Various documents, downloaded files etc

ie, really, syncing much of the profile's files and app state.

iCloud seems suitable if you want everything going up to the cloud and back, which for two computers sitting next to each other most of the time seems overkill. I'd love an automatic local file sync and something similar to Continuity that worked Mac <-> Mac, not iDevice <-> Mac.

> iPads could have done so well in the pro space if they'd supported proper integration with Macs and could be used as extension devices rather than exclusively as replacement devices

And indeed, the places I see iPads used most heavily are exactly those. You see it heavily in the audio industry: Logic Remote is really really good and all the other similar tools (Lemur, TouchOSC, etc.) have huge uptake. I'm in the process of building a video/CG overlay system for TV/streaming[1] and the best way to use the backend management system I've found while I'm trying to concentrate on something else is my old iPad 2.

[1] https://github.com/bossmodecg

That looks really neat! will try to follow it on GH!

Apple's problem is more subtle. The MacPro was a good choice for a tiny market niche - pro video editors and pro musicians.

This is a market in which a $5k 10 core PC is considered an entry level machine. 3D houses and some music studios have racks of server-grade hardware that is far outside the reach of most mortals.

There is a bigger market of users who like to label themselves pro, but are actually prosumer. It includes smaller music studios (home or not), design studios, serious hobbyists and dabblers, smaller video houses, academic music composers, and so on.

Also many software developers.

These users are not in the Hollywood league, and they typically have a limited budget.

The old cheesegrater MacPro was a good fit for this market. It was also good enough, and expandable enough, to get some interest from the no-compromise pros.

The new MacPro is too expensive for the prosumers, and poor value - and underpowered - for the no-compromise pros.

I'm not yet convinced Apple understands this. Jobs understood it because his links with Pixar meant he could see first hand what Hollywood pros were using.

I don't think the current management is as open to creative input from creative professionals at all levels.

If I'm right, the 2018 Mac Pros will be open like PCs, but they still won't truly satisfy either type of user.

I'd beg to differ on the point of Pro-Video editors and musicians seeing as I've contracted for a few TV shows and am a hobbyist musician etc. It was a mediocre choice, if not poor choice:

Generally a TV Show workflow (since FCP X killed its user base) an Avid centric base. Editors grab all their footage off a central server to edit the down converts (1080p). SSD is great for a boot drive but you'll need hefty internal storage. Now you need to buy an expensive external solution. Also for same goes for Digital Musicians or recording, although the storage demands are quite a bit less.

The dual GPU setup was under-utilized and the consumer grade cards are far better for render assistance bang-for-buck. So much so that it was better just to upgrade GPUs with gaming cards. Notably, the Fire GLs are underperfomers for After Effects. Also Musicians get almost zero benefit from dual GPUs. It's unnecessary cost.

Since in Hollywood TV, you are contract you, you bring your own hardware. When USB 3.0 took off, most of my editor friends just grabbed USB 3.0 cards and popped them into the computers. Mixing guys tended to have a lot invested in PCIe audio interfaces for ProTools. A few friends had MPEG4 encoder accelerators which were great for quick encodes for dailies. I'm not sure what fantasy world Apple was in that small size of the Mac Pro 2013 mattered but post-guys set up shop for weeks at a time, and its easier to cart one heavy tower followed by monitors than a small computer with a box of dongles and misc items.

The friends I still keep in touch with in LA mostly have switched to Windows instead of going for the Mac Pro 2013. More bang for the book (time is money) and more performant. Anyhow, other than that I agree. just rerelease the 2012 with new innards and call it a day.

Or, in other words, not the Minimum Viable User:


When I look around at the pro market that need fast machines I see people working with 3D/CAD, video and sometimes sound and pictures. At least here in The Netherlands almost all those people are working with Windows or Linux based software. So they already get a new video card once in a while and add memory and disk space when needed.

I just think that the pro market for Apple became so small because normal computers are getting fast enough for most pro people. For example photo editing works fine on a Apple laptop even with very high resolutions and a lot of layers.

So I can imagine that Apple just doesn't need a pro desktop anymore because the market became too small.

Apple just pulled a self fulfilling prophecy.

By ignoring the pro market demand for faster turnaround and bleeding edge performance they pushed everyone jumping ship faster; now that the market for the mac pro shrunk to a point the line is unprofitable the circle is complete.

meanwhile the trillion dollar render/graphics/cad industry is gonna need something to keep up with the always increasing demand, and will simply look elsewhere.

The ultra high-end market has always been turbulent. Look at Silicon Graphics. They dominated for a good stretch, but then a wave of less expensive windows based machines (like Integraph) followed by commodity graphics-card powered PC boxes completely decimated that market.

During that entire time Apple was always positioned in a niche. For a time it was desktop publishing, video editing, photography, audio. A lot of that had to do with loyalty of the developer community, who produced killer apps that only really ran on Apple hardware, as well as appealing to a customer base that deeply preferred Mac OS.

But it's worth keeping in mind that the Mac Pro wasn't really ever dominate in certain high-end spaces like 3D. I mean, I'm pretty sure I remember hearing that the ID team at Apple was for a long time using non-apple hardware to run their CAD stuff until one day Ive and Jobs were like, 'this is unacceptable'.

The render/graphics/cad industry is not 1.5 percent of world GDP.

Maybe he's including feature films, game development, aerospace, and anything that requires bleeding edge tech.

If the market for pro workstations is more than a couple billion, I'd be pretty damn surprised.

Realistically, that is probably the case, but people who are using macOS professionally are still willing to spend money on hardware with more performance, 'it can never be too fast' etc. My old 2010 Pro compiles node.js about 4X faster than a newer i7 iMac (and uses at least that much more electricity to do it, so that may not seem like a huge win, necessarily). It seems clear that the old Pro's flexibility and upgradability are essentially at odds with Apple's general direction towards soldered and glued products that are only extensible via Thunderbolt; I would be happy to see them move back towards more user-serviceable hardware, but I am not holding my breath. The question of whether that makes sense for their business is trickier, they do have deep pockets, so unless they lost money on the new Pro, it would at least provide bragging rights.

I still use a Mac for audio work, but it's a Hackintosh. Last real Mac (aside from a Macbook) was a G5 Mac Pro. Which, was the last good Mac Pro ever made.

As far as development work, I run it all on servers. Whether it's something cloud based, or my own Linux box at home, A low powered laptop SSHFS'd into a server is my preferred way of working.

> was a G5 Mac Pro. Which, was the last good Mac Pro ever made.

Which wasn't even a Mac Pro. Interesting.

When I look around at the pro market that need fast machines I see people working with 3D/CAD, video and sometimes sound and pictures

That's the market Apple have historically singled out, but it isn't the only "pro" segment. Another big one is software developers, who care a lot about compile and/or test time (and maybe having enough RAM to run a bunch of VMs...) -- and this apparent was recognised in the April meeting.

"Three senior execs sit down with five reliable (meaning no clickbait)"

Not that these aren't great journalists, but Buzzfeed, Tech Crunch, and Mashable are no strangers to clickbait.

While I can see that the can-like current design of the Mac Pro was a honest mistake in predicting the PC market, I am baffled how long it took them to recognize the mistake and take no action. They must have noticed it quite recently, so they did not care for the mac pro for years.

The mac pro isn't important for Apple measured by the money it makes by its sales. It is important for keeping the mac line alive. It is the machine for all the people writing the software for iOS devices, and of course for all the mac hardware which does create the revenue for Apple. Also, it keeps the professionals on the mac platform. Even if Apple would sell the mac pro at a loss, it would be worth it for its role in the environment. It plays the same role for Apple, as all those photographers at the sports event for Canon and Nikon. There isn't much money to be made from them, but by their reputation, all the equipment is sold to the big public.

In the mid-2000 years, the first sign of macs becoming popular was that at conferences, 80% would have an Apple laptop. They were leading the wave. But today, their market share is still high, yet, one can read more and more blog posts on how to switch to Linux or build a Hackintosh. I guess this is, what got the message at least partially to the Apple management.

I am writing this on a 5k iMac, which I quite like (except: no video input, what were they thinking??). In a year or two I would be interested in upgrading to a mac pro, if its a reasonably versatile device. I don't need too much upgradability, but I would expect to be able to order it with a current high-end graphics card and proper internal storage. Otherwise, I might consider going back to Linux eventually.

"The A10 processor inside my iPhone is no less powerful than the Intel processor inside my MacBook"

[citation needed]

The first has results from Geekbench finding that "A10 Fusion has roughly the same single-core processor performance as Intel's top sixth-generation Core m processor".

The second references Matt Mariska's misleading tweet: "Geekbench has the iPhone 7 beating the $6,500 12-core Mac Pro in single-thread." - Never mind that it's a $6,500 12-core Mac from over four years ago. This one gets regurgitated over and over.

The third still illustrates performance edging close to the Core M series of processors.

Assuming he has a Macbook Air that's probably true but only because the A10 in an iPhone is performing at the sweet spot on its frequency-voltage curve while the Core iX in a MacBook air is severely undervolted to get to the Air's target power. If Intel were to design a core from scratch for that power envelope they could do better. But Apple has done very well in getting comparable performance to Intel despite Intel's process advantage.

I suspect the author was being exact and referring to the product just called 'MacBook', rather than referring vaguely to a MacBook Air.

(I can't back that up, except to note that Apple fans are usually fairly scrupulous about referring to their MacBook Pro, iPad Air, iPhone 6S etc specifically rather than lumping them together).

In any case the current MacBook is quite similar to the MacBook Air and the rest of your comment probably stands.

If you scroll down on this page, there's a comparison between the A9X (which is slower than the A10) and the Core M used in the MacBook:


Not really surprising since the Core M is ostensibly designed as a tablet processor.

My guess is that the author read this https://www.theverge.com/2016/9/16/12939310/iphone-7-a10-fus...

Perhaps without understanding the full context.

The author of the article is this guy: https://en.wikipedia.org/wiki/Jean-Louis_Gass%C3%A9e

I could be wrong, but I imagine he still knows a thing or two about CPU architecture and performance.

No, you're right, I missed that. He either has an old MacBook though, or is being a little cheeky.

I've been using the new Mac Pro since 2015 and it's served me will without the need for upgrade... except for one thing. The primary disk being only 256GB causes me all kinds of issues. Everything wants to install and save there and there's no simple way to offload stuff to a secondary disk, so I'm constantly moving, symlinking, etc to keep enough space free so that my computer doesn't crash. Very annoying.

This could, of course, be solved with software.

Check out this company that built their whole infrastructure around the Mac Pro. What a waste of capital.


I mean, they seemed happy with the results right?

That's right

Generally speaking, I really like OSX/macOS. I am starting to grow a disdain towards Windows, with it's insistence on changing the default browser to IE + Bing on every update.

Linux is great, but it just doesn't do it superficially for me. I can make it look great for screenshots, but as soon as I click a button or engage with the DE, the facade falls apart.

macOS/OSX seems to be like a Linux distro with a very well developed desktop manager.

The only problem is, I need to buy a Mac. Who the hell can afford that, holy sh*it

I am starting to grow a disdain towards Windows, with it's insistence on changing the default browser to IE + Bing on every update.

This did not happen when I installed Creators Update a few weeks ago. Firefox was still my default browser after updating.

I like my Mac Pro. It's more powerful than any MacBook I've used and gets stuff done.

I find all the bad press and hate confusing. It's just a computer.

These are the two sources of confusion, and they are completely valid. Neither you "liking" your Mac Pro, nor it being "just a computer" address these completely valid concerns.

1.) Why is Apple not keeping this system in line with the state of technology? If it's targeted at professionals, why have they left their only option in that department 5 years behind in technology?

2.) Are they planning on making the next iteration upgradeable? For some time now, Apple has used third party silicon, interchangeable with that of any other system. Intel, AMD, nVidia. Professionals on any other platform can spend a couple hundred bucks and modernize their computer for 2-3 years. Isn't it therefore possible that the lack of sales in Mac Pro division aren't because of lack of interest or a small professional market, but because of a bad idea to make a proprietary desktop tower which locks down upgradeability?

It's very expensive and ~five years old at this point.

Nope, it wasn't cheap. But as an artifact, and me being a Computer Engineer, I'm impressed with it as a piece of hardware and design.

I was impressed when I first saw it as well. However since they say it can't be upgraded, I'd consider it a failure.

This is about Mac Pro, which is the spaceship looking like server hardware from Apple, not Macbook Pro, which is a notebook.

Why are there so many articles, mostly condescending or blatantly speculative, about all this Apple? I can't, for the life of me, understand this.

They do what they do and that's it. Only Apple understands what Apple does. No one else. How is that not clear?

Sorry for the rant.

You are on a discusssion forum. Not a "welp, everyone does what they do for the reasons that they do 'em, let's move on" forum =)

People want Apple to be successful because they like(d) their products.

I want Apple to be successful because all of their competitors' products have annoying quirks and reliability issues. Ultimately, I'd love to see a company actually try to compete with Apple on quality but it doesn't seem likely to happen.

I want some companies to be successful but I won't be writing about them to death and speculate and pretend I know better, writing articles "teaching them" what they should focus on and etc.

Really? Do you think it's reasonable the amount of people that seems to devote themselves to this?

Because entire industries have built themselves around MacOS.

If industries like video and graphic design (and possibly even app development?) are forced to switch to Windows to maintain pro-level capabilities, what's going to motivate them to stick with iPhone?

The ~82 million people in America alone with iPhones as of 2015.

[0] https://www.statista.com/statistics/232790/forecast-of-apple...

Do you think it’s weird for an ex-CEO to have thoughts and opinions about a company he used to run?

He's not ex-CEO; he was the exec heading the mac division.

That was more than 25 years ago; any company would change a fair amount in that time.

Jean Louis-Gasse is the former CEO of Be Inc., and CATC.

So, no, not of Apple, but of two other technology companies.

huh, my mistake

I enjoyed the article.

What I can't understand is why people feel the need to make this kind of comment. Do you want me to explain why I enjoyed it? Or do you get some personal satisfaction from publishing your bewilderment?

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact