Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Exactly! I'm thinking the same thing, I tried that time with external hardware with a Mac Pro 6.1, but it looked like the Borg had assimilated my desk. The Mac Pro tower is the best solution for me, but not just for neatness and aesthetics - but also for speed, the Mac Studio can't have a RAID 0 array with 4 SSDs inside, and on the Mac Pro it's easy-peasy.

Apple silicon only does software RAID anyway


The studio can do the same RAID 0 array over thunderbolt as AS Mac Pro. You just can’t put the disks inside the same box
 
I don’t think they ever made printers/scanners, at least not in the OSX/MacOS era, possible they did so in the 80’s or 90’s.

Apple helped invent / create the desktop publishing industry by releasing the first (somewhat) affordable and in its quality for the time amazing "LaserWriter" laser printer. This was a cooperation between Canon and Apple.
For years Apple offered a series of superb laser printer products. Long before anyone else really entered the market. Outside Canon.

Before that they offered dot matrix printers.
They also offered a few ink jet printers in the lower end market segment.

While I agree that it would be nice to have Apple printers and Apple WiFi routers again, I do think that both of these are no longer "key" macOS elements in this day and age.

Printers have become niche as people send PDFs via email, even electronically signed ones, and corporations usually print via their photocopier machines.

WiFi will also soon be a thing of the past as WiFi will slowly merge with 5G/6G built-in phone modems.

Local backups on the other hand will likely be the last to disappear. At least in my opinion.
At least not until Apple offers 50+TB iCloud backups for $5.99 a month. Which will likely not happen for a few decades. If ever.
 
  • Like
Reactions: G5isAlive
Pretty sure 3.5" is obsolete tech in Apple's view.

Apple's strategy - like it or loathe it - is based on external expansion using USB3/4/Thunderbolt - or network attached storage, which is why the higher-end Macs have insane external I/O bandwidth (including 10GB Ethernet as standard on the Studio, which is fairly unusual).

They're not making hard drives obsolete - just saying that they belong in external (or probably remote) boxes.

As it is now, I find the Mac Studio not enough of an upgrade over the MBP.
The Mx Max Studio isn't meant to be an upgrade over a Mx Mac MBP - it's an alternative that offers better thermals and more connectivity than a MacBook if you don't need mobility. Meanwhile, the Mx Ultra Studio has a more powerful SoC that would melt a MBP.

So how does Apple suppose users install and use Time Machine?
On an external hard drive or NAS with hard drives. Nothing obsolete about hard drives - people are making that up. TM on an external HD is plug'n'play. Most decent NAS boxes support TimeMachine over SMB.

The only thing that Apple have made "obsolete" is the old, proprietary Apple file-sharing protocol used by TimeCapsules, which has been replaced by the (effectively) industry standard SMB/CIFS protocol, after years of warnings. Networked TimeMachine backups now run over SMB and is supported by the ubiquitous Samba implementation of SMB - or directly on a USB or TB-attached HD.

For starters, using an external HD means you can periodically swap TM drives to maintain multiple backups (a singke backup is no backup).

So if your Mac comes with a 4TB SSD how does Apple suggest one create a 16TB Time Machine backup?

On a 16TB external or networked hard drive. You can easily get 28GB external HDs, but if you're managing that sort of data you could probably use a decent NAS with multi-disc arrays.

If you need 4TB of working storage then you've probably outgrown Time Machine anyway - why are we talking about a point-and-drool consumer backup tool in a thread about $4000+ "pro" systems?

Even with TM it's easy to exclude folders and volumes from the backup, so you could just use TM for your system/app settings, email and admin documents, and find a more efficient backup strategy for whatever is taking up the other 3.5TB of your 4TB system.
 
Time Machine is a key macOS component that is now seemingly 100% reliant on third party products
It's reliant on industry standard products and protocols that are available from a multitude of sources.
Pretty much any external HD enclosure and many NAS boxes with open-source software will work perfectly with MacOS and TM.

You do get why that makes things better, right? That relying on proprietary black-box Apple peripherals is why all of those TimeCapsules are now landfill bound, while many contemporary third party NAS boxes, or PC based home servers, can simply be switched over from Netatalk to to SMB?

It's far more important that Macs work well with standards-based products than for Apple to produce their own closed version of every type of peripheral. People expect to use Macs at work, where they can't choose what WiFi router or shared storage they need to connect with. At home, many, many people have the WiFi router supplied & configured by their ISP (...and there are at least 4 different types of router-to-broadband interface to juggle).

While I agree that it would be nice to have Apple printers and Apple WiFi routers again, I do think that both of these are no longer "key" macOS elements in this day and age.
Nostalgia isn't what it used to be.

Mac-compatible printers - either Apple or third party - used to cost an arm and a leg c.f. PC-oriented printers.
Ditto ADB mice, keyboards etc.
Macs at work were tricky to connect to PC-oriented email and file sharing system without expensive (and often half-baked) software and/or persuading PC-obsessed IT departments to install server-side Appletalk-compatable services.
Macs had SCSI interfaces for external drives, scanners and CD-ROMs - that was standard but largely "server class" and expensive c.f. PC options.

Things got immensely better when the iMac came out with USB (which also catalysed the PC industry to start supporting those odd rectangular ports that had been appearing on PC hardware) and within a few years we had a far wider choice of cross-platform printers, scanners, external drives, mice, keyboards... The flipside of that was there was no longer much point in Apple or specialist Apple manufacturers making such products unless they could distinguish themselves in some way.
 
  • Like
Reactions: eldho
The Mx Max Studio isn't meant to be an upgrade over a Mx Mac MBP - it's an alternative that offers better thermals and more connectivity than a MacBook if you don't need mobility. Meanwhile, the Mx Ultra Studio has a more powerful SoC that would melt a MBP.

The Mac Pro was not only about the CPU. That is the main misunderstanding that people have when they constantly bring up a Mac Studio as a "replacement".
Not everyone cares about the M Ultra chip. Whether it melts a MacBook Pro or not.

The Mac Pro had up to 4 internal HD bays and PCI expansion options - all in a quiet, well cooled package that provided enough power for all these extensions.
That was a very neat, clean and quiet solution. One which is now missing in Apple's hardware lineup.

It's not about the fastest chip.

For starters, using an external HD means you can periodically swap TM drives to maintain multiple backups (a singke backup is no backup).

This is not a feature unique to external HDs!
There are plenty of HD bay solutions out there, which Apple could easily adopt if they so choose, that allow for easy swapping of 3.5" SATA HDs. No screwdriver needed. Some vendors also sell external carry or storage cases for "raw" 3.5" HDs, that can be used for storing backups off-site.

There is no need to buy external HDs.
You can do exactly the same thing with "raw" swappable SATA drives.

You do get why that makes things better, right? That relying on proprietary black-box Apple peripherals is why all of those TimeCapsules are now landfill bound, while many contemporary third party NAS boxes, or PC based home servers, can simply be switched over from Netatalk to to SMB?

Oranges and Apples get mixed up here.
Apple being Apple and creating proprietary systems is one thing.
But the Mac Pro internal HD bays were never "Apple proprietary", nor were their PCI extensions.
Those were all industry standard.
Any SATA drive, many SSD drives, and many PCI cards could be used "as is". The Mac Pro had nothing to do with the ill-advised proprietary standard Time Capsule products.


why are we talking about a point-and-drool consumer backup tool in a thread about $4000+ "pro" systems?

Because it is the internal HD bays and PCI expansions, all contained in a single cooled and power providing package, that the $4000+ "pro" system offered - and that is now missing from Apple's product lineup.

The Mac Studio is just not a replacement, in my opinion.
Apple forcing its customers into a noisy fan, messy cable, unsightly power brick, ugly external expansion box chaos, is not the same thing.
It may provide technically similar features. But it is much noisier, messy, and in most cases terribly ugly to look at.

There are people who do care about their desktop environment.
Not everyone has a "I-don't-care-about-mess-and-fan-noise / anything-goes-as-long-as-it-works" attitude.

It just really puzzles me that Apple, a company which prides itself on product design, forces this kind of mess unto their customers from day 1 now. Their whole insistence on good product design becomes somewhat hollow and a bit unbelievable if at the same time they know that their customers's desktop setup will now always be noisy and look messy as they no longer provide a complete, well designed solution anymore.

Maybe not everyone - but some customers do care about these details.
 
Last edited:
It just really puzzles me that Apple, a company which prides itself on product design, forces this kind of mess unto their customers from day 1 now.

Apple are constantly prioritising showroom appeal at the expense of real-world practicality. Hence all the iMac and Mac mini docks that put USB ports and SD card readers in an accessible position at the front; laptops with just a couple of USB-C ports, that need a dongle for HDMI / USB / Ethernet etc. They couldn’t give a **** about your desk, as long as their product shots look good.

The reason they don’t make a tower is because it doesn’t fit their hardware architecture, for the reasons discussed above. Not because they don’t understand the value of internal expansion. As for HDDs - if you’ve got the cash for a Mac Pro, you can likely afford a 4-bay NAS. I’d rather locate the drives away from my desk anyway, as they aren’t exactly quiet.
 
Last edited:
  • Like
Reactions: Charlie Bonesx
WiFi will also soon be a thing of the past as WiFi will slowly merge with 5G/6G built-in phone modems.

Soon? Dunno about that. In any case, there will be wireless network capability that families and organizations can manage as a security domain separate from the wider (WAN) world.
 
The Mac Pro had up to 4 internal HD bays and PCI expansion options - all in a quiet, well cooled package that provided enough power for all these extensions.
That was a very neat, clean and quiet solution. One which is now missing in Apple's hardware lineup.
The M2 Ultra Mac Pro did not sell well, though. At least that is what we are told. On paper, it is a real powerhouse, but, we don't seem to hear that much about it. I assume that it would have sold better if it hadn't cost so much, but still, it should have sold better to "pro" users if that is what people really wanted.
 
it should have sold better to "pro" users if that is what people really wanted.

Pros seemed happy to spend thousands on MPX GPUs the previous generation. They were a major selling point at release. It’s not surprising that removing that capability made the MP a less attractive proposition.
 
  • Like
Reactions: Dreamail
Apple silicon only does software RAID anyway

The studio can do the same RAID 0 array over thunderbolt as AS Mac Pro. You just can’t put the disks inside the same box
I know it's software-based, but that has a minimal negative effect, the difference in read/write speed is double compared to the default Mac Studio setup.

Also, I'm kind of skeptical that an external RAID 0 box would work as fast as something connected directly to the PCIe bus. And how much is a Thunderbolt case that can support 4 x NVMe slots that can be configured as RAID 0?
 
I think the 2013 Mac Pro was the death kneel to the Mac Pro line, the 6 year delay was a huge pain for people who relied on Mac Pro's (or previous tower Macs) to make a living (especially prosumers and professionals). Apple had a horrible strategy for 2019 Mac Pro, seemed like they didn't learn anything from the 2013 debacle except maybe the form factor. Yeah they added all these PCIe slots back with a tower, but they made these custom MPX modules which never got any upgrades (and had custom onboard power slots which never got used outside of Apple produced cards) and then they stopped adding AMD drivers past RDNA 2 (we are at RDNA 4 now) and completely dumping NVIDIA driver support...and then Apple silicon came and completely killed it. The 2019 Mac Pro had horrible pricing too, with configurations reaching $30,000 which is insane. I get that Intel Xeon CPUs were very expensive (28c was $7,000 OEM btw) but outside of that the base model was $6000 or something and very underpowered. If you compare that to the 5,1 Mac Pro for example, it was highly upgradeable by anyone and it never got that expensive. They really went "too professional" with the 7,1....and then the inability to add GPUs or RAM with the M2 Mac Pro also killed it. And the lack of PCIe lanes (32 compared to 64 with Intel Xeon)...this meant that only about 2 16x PCIe cards could be utilized, so the PCIe slots didn't matter at this point.

Really a botched job by Apple, or maybe on purpose. The Mac Studio makes more sense for these users, but the lack of addon cards and going back to external thunderbolt enclosures for professionals (such as musicians, production houses, etc) really pissed people off but people who are still on Macs will go this route as there are rackmount options for Mac Studio and external thunderbolt PCIe enclosures.

Anyway, what's done is done, Mac Pro is dead.
 
The Mx Max Studio isn't meant to be an upgrade over a Mx Mac MBP - it's an alternative that offers better thermals and more connectivity than a MacBook if you don't need mobility. Meanwhile, the Mx Ultra Studio has a more powerful SoC that would melt a MBP.

My point was that a Mac Studio with a few M.2 slots would make a very strong proposition for me, give it a clear edge over the MBP.

It would be a modern-day Mac Pro replacement for me.
 
I think the 2013 Mac Pro was the death kneel to the Mac Pro line, the 6 year delay was a huge pain for people who relied on Mac Pro's (or previous tower Macs) to make a living (especially prosumers and professionals).

The 2013 model was the death knell for the Mac Pro because it indicated Apple really didn't want to build a tower (in the tradition dating back to the PowerMac). They could only justify building and stocking a high end desktop if it took up little space in the storeroom. Thermal corner or not, they could have revised it with a newer Xeon / chipset, with TB3 and increased RAM and SSD speeds. Perhaps even got AMD to configure GCN 3.0 GPUs with appropriate TDPs. But the 2013 MP was probably already losing Apple money and they didn't want to spend a penny more on it. Computer-indifferent beancounter Tim Cook probably took some convincing to not just can the whole line. If advance feedback on the iMac Pro had been more positive, I expect he would have.

Apple had a horrible strategy for 2019 Mac Pro, seemed like they didn't learn anything from the 2013 debacle except maybe the form factor. Yeah they added all these PCIe slots back with a tower, but they made these custom MPX modules which never got any upgrades (and had custom onboard power slots which never got used outside of Apple produced cards) and then they stopped adding AMD drivers past RDNA 2 (we are at RDNA 4 now) and completely dumping NVIDIA driver support...and then Apple silicon came and completely killed it.

MPX was a reasonable solution and would likely have continued for a few generations. But from a standing start in April 2017, the 2019 MP only hit the market at the end of December 2019. Meanwhile, Apple A-series performance was a consistent upward line on a graph, and despite being recently dominant, Intel couldn't stop falling over their own feet. Basically, the 2019 MP was likely released later than hoped, and the Apple Silicon transition likely occurred earlier than predicted.

No Mac had included an Nvidia GPU since 2015, so that was nothing new.

The 2019 Mac Pro had horrible pricing too, with configurations reaching $30,000 which is insane. I get that Intel Xeon CPUs were very expensive (28c was $7,000 OEM btw) but outside of that the base model was $6000 or something and very underpowered. If you compare that to the 5,1 Mac Pro for example, it was highly upgradeable by anyone and it never got that expensive. They really went "too professional" with the 7,1....and then the inability to add GPUs or RAM with the M2 Mac Pro also killed it. And the lack of PCIe lanes (32 compared to 64 with Intel Xeon)...this meant that only about 2 16x PCIe cards could be utilized, so the PCIe slots didn't matter at this point.

The MP likely struggled to make money for Apple. They tried radically downsizing it, but that got a lukewarm reception. They tried getting rid of it altogether, by giving the iMac a Xeon, but perhaps got pushback when previewing it to key customers. So they went back to a tower, but decided their only option for profitability was to go upmarket, doubling its price. They'd lose some traditional sales of course, but these would be offset by high-spec configs for big media companies.

Really a botched job by Apple, or maybe on purpose. The Mac Studio makes more sense for these users, but the lack of addon cards and going back to external thunderbolt enclosures for professionals (such as musicians, production houses, etc) really pissed people off but people who are still on Macs will go this route as there are rackmount options for Mac Studio and external thunderbolt PCIe enclosures.

Apple Silicon effectively precludes slots. The Studio is using a repurposed laptop SoC or two, which don't have many spare PCIe lanes. Further, AS and macOS are all-in on unified memory, meaning dGPU support would be a massive pain to implement. A traditional slotbox would be possible if Apple really put their back into it, but the M2 Ultra MP was as good as they could get whilst keeping their costs / effort reasonable. Apple knew it wasn't inspiring, but it did the job to round out the AS transition, whilst mostly re-using existing components. It got a 30 second mention at WWDC 2023, before being quietly cancelled less than 3 years later.


About $250

That delivers what Frixo asked for, though 4x NVMe blades in RAID 0 would be severely bottlenecked by a 3GB/s connection. Still, it would aggregate their capacity, and would be fast enough for most needs in practice.
 
Last edited:
That is nothing new or out of character for Apple.

AirPlay and HandOff rely on WiFi, yet they discontinued their excellent AirPort router ~10 years ago. They once had the dedicated Time Capsule HDD for your Time Machine, but also discontinued after poor sales. They made Xserve, discontinued. I don’t think they ever made printers/scanners, at least not in the OSX/MacOS era, possible they did so in the 80’s or 90’s.

If the product doesn’t move, they chuck it.
I don't think it's strictly about numbers. Back in the day Airports, Time Capsules, laptops and towers regularly sold together because people wanted the 'just works' ecosystem. It wasn't perfect, but it was damn good.

The networking and backup products certainly sold. I don't have numbers, but I was traveling the country installing Xserves / RAIDs / Promise RAIDs for years. If they still made these products and kept the interfaces simple and beautiful, they would still be selling. All the glow Linux is currently enjoying with the Win11 departure... Apple would be eating their lunch.
 
I did love Airport and Airport Express, better than anything else at the time. Time Machine never made sense to me, so Time Capsule was not an option.

SJ famously said about discontinuation of XServe that people weren't buying them.
 
The M2 Ultra Mac Pro did not sell well, though. At least that is what we are told. On paper, it is a real powerhouse, but, we don't seem to hear that much about it. I assume that it would have sold better if it hadn't cost so much, but still, it should have sold better to "pro" users if that is what people really wanted.
The M2 Ultra wasn't what the target audience of a "true" Mac Pro wanted. That's just not compatible with the "consumer first" Apple Silicon designs. My best guess is that most real Mac Pro users jumped ship after 2013. By the time the M2 Ultra came around even the last few die-hard customers couldn't fake being interested in a big expensive box with too many limitations compared to what a big tower should be.
 
MPX was a reasonable solution and would likely have continued for a few generations. But from a standing start in April 2017, the 2019 MP only hit the market at the end of December 2019. Meanwhile, Apple A-series performance was a consistent upward line on a graph, and despite being recently dominant, Intel couldn't stop falling over their own feet. Basically, the 2019 MP was likely released later than hoped, and the Apple Silicon transition likely occurred earlier than predicted.

MPX was a horrible solution and a closed off system just like the 2013 debacle. They never once upgraded anything with the MPX modules and no other manufacturer would make MPX modules except a one time drive module from some 3rd party.

And they clearly didn't care about non MPX PCIe GPUs whatsoever and they stopped adding drivers beyond the 6900XT. They didn't even support the 6950XT. They could've easily added 7xxx support and even 9xxx support but they didn't. They obviously want us to get an M series mac.

Apple went too high end on the 7,1 instead of finding a middle ground. Some of the setups cost $30,000+. What made the 5,1 and below successful was that it was highly configurable and even the higher end dual Xeons weren't ridiculously priced.

I just think their strategy and thinking on the Mac Pro was wrong for a long time especially after Steve died. They tried to cater to the wrong crowd twice in a row after 5,1. 2013 Mac Pro saw no upgrades whatsoever, not even a price drop for 6 years. 7,1 went too high end and then the M series came out and that ended the Mac Pro. The M2 Mac Pro had no reason to exist at all. I actually feel worse for that crowd than the Intel 7,1. But I bet that the M2 Mac Pro didn't even sell well.

If you look at Apple's moves since 2022, it's pretty obvious that a Mac Mini, Mac Studio, MacBook Pro (which reaches desktop performance) + Apple Studio Display (or XDR) is the perfect solution for most people. There isn't even a iMac 27" (or iMac Pro) anymore so it seems like they are looking at what people want and going with that.

There is still a decently sized professional but very niche area of Mac users who require PCIe addon cards like music professionals, video professionals and people who want to add storage. But that's moved to Thunderbolt enclosures in racks and what not.
 
MPX was a horrible solution and a closed off system just like the 2013 debacle.

The 2013 MP was completely closed - the cards were in a proprietary form factor so no standard ATX GPU would fit. MPX was at least fully backwards compatible - it didn't interfere with the regular PCIe slot and indeed, many people installed off the shelf PC graphics cards.

My criticism of MPX would be that it was over-engineered and unnecessary, and the sort of pointless reinventing the wheel that delayed the 7,1's release. Power could have been adequately provided by high power cable sockets next to each slot, and Thunderbolt video could just have been a mandated output on the back of MPX cards - no need to pipe it round the system.

They never once upgraded anything with the MPX modules and no other manufacturer would make MPX modules except a one time drive module from some 3rd party.

The AS transition meant Apple themselves lost interest, and the market was too small for anyone else to bother. Especially given the slots were backwards compatible anyway - people could just buy a PC model.

And they clearly didn't care about non MPX PCIe GPUs whatsoever and they stopped adding drivers beyond the 6900XT. They didn't even support the 6950XT. They could've easily added 7xxx support and even 9xxx support but they didn't. They obviously want us to get an M series mac.

Not sure how 'easy' it would be to add 7000 / 9000 series driver support. Someone would have needed to write the drivers. In the past, that came for 'free', since they were already written for MBPs / iMacs.

Of course, with Apple being among the richest companies on Earth, it wouldn't have been much to ask for. Apple could have easily paid AMD to develop Mac drivers for their latest cards. The problem was, Apple were already struggling to match the GPU horsepower of a 2019 MP with their Ultra SoCs. Allowing MP users to stuff their machines with 9000 series cards would just raise the bar ever higher. They wanted the Ultra to be able to eclipse the MP in all ways. Even now, it can't match the RAM capacity of the 7,1.

Apple went too high end on the 7,1 instead of finding a middle ground. Some of the setups cost $30,000+. What made the 5,1 and below successful was that it was highly configurable and even the higher end dual Xeons weren't ridiculously priced.

Yeah. As I said earlier, I expect they did the math and reckoned the sweet spot for profitability was a smaller number of more expensive machines. They'd exhausted all the other options.

I just think their strategy and thinking on the Mac Pro was wrong for a long time especially after Steve died. They tried to cater to the wrong crowd twice in a row after 5,1. 2013 Mac Pro saw no upgrades whatsoever, not even a price drop for 6 years. 7,1 went too high end and then the M series came out and that ended the Mac Pro. The M2 Mac Pro had no reason to exist at all. I actually feel worse for that crowd than the Intel 7,1. But I bet that the M2 Mac Pro didn't even sell well.

I think the 7,1 crowd had it worse. At the time of release, there was a load of pent-up demand, so likely an initial surge of orders. This was before the impending transition had been announced, and the 5,1 had enjoyed a famously long life in the field. (Non-corporate) buyers could justify the expense on the basis it would be a '10-year' machine. The AS transition, announced just 6 months after the 2019 finally started shipping, meant Apple would clearly focus all their energy on the new shiny. They had zero incentive to help you enhance / upgrade Intel machines - it would just delay you moving to an M series machine.

At least M2 MP buyers knew exactly what they were getting into. Apple's cards were finally on the table, confirming that add-in GPUs were never going to be a thing for AS. They even put the price up by another £1K IIRC.

If you look at Apple's moves since 2022, it's pretty obvious that a Mac Mini, Mac Studio, MacBook Pro (which reaches desktop performance) + Apple Studio Display (or XDR) is the perfect solution for most people. There isn't even a iMac 27" (or iMac Pro) anymore so it seems like they are looking at what people want and going with that.

There is still a decently sized professional but very niche area of Mac users who require PCIe addon cards like music professionals, video professionals and people who want to add storage. But that's moved to Thunderbolt enclosures in racks and what not.

Apple Silicon is great for everything except slotboxes, which Apple aren't too keen on anyway. It's easy to see why they're OK with losing those customers to Windows / Linux.
 
What made the 5,1 and below successful was that it was highly configurable and even the higher end dual Xeons weren't ridiculously priced.


The cMP is an anomaly in that for a short while (2006-2012) they were actually competitively priced, compared to similarly specced PC workstations. At launch, they would often be a few hundred bucks cheaper, but Apple never dropped prices so after six months or a year the PC’s would be cheaper.

But if you look at Apple’s entire 50-year history, the early Intel years were an exception, not the rule. It was a unique confluence of circumstances that forced Apple’s hand. The iPhone not yet dominated their sales, and they had to compete with Wintel on merit in an era were spec-wankers were still a very large part of their customer base.

I don’t think Steve Jobs loved that, frankly. If you look at the products Apple made before he was ousted, he was always an all-in-one guy.
Being the aesthetician that he was, I think those big rectangular boxes from the 1990’s-2000’s offended his sensibilities. And if the G4 Cube hadn’t failed so miserably, I think he would have gotten rid of the towers a lot sooner.

In that sense I think the 2013 Mac Pro was really his legacy.
 
Last edited:
  • Like
Reactions: BSDnostalgia
The cMP is an anomaly in that for a short while (2006-2012) they were actually competitively priced, compared to similarly specced PC workstations. At launch, they would often be a few hundred bucks cheaper, but Apple never dropped prices so after six months or a year the PC’s would be cheaper.

But if you look at Apple’s entire 50-year history, the early Intel years were an exception, not the rule. It was a unique confluence of circumstances that forced Apple’s hand. The iPhone not yet dominated their sales, and they had to compete with Wintel on merit in an era were spec-wankers were still a very large part of their customer base.

I don’t think Steve Jobs loved that, frankly. If you look at the products Apple made before he was ousted, he was always an all-in-one guy.
Being the aesthetician that he was, I think those big rectangular boxes from the 1990’s-2000’s offended his sensibilities. And if the G4 Cube hadn’t failed so miserably, I think he would have gotten rid of the towers a lot sooner.

In that sense I think the 2013 Mac Pro was really his legacy.

Agree with much of this, though SJ seemed pretty chuffed with the PM G3 case design when he unveiled it, lowering the side / logic board to oohs and aahs from the audience. His NeXTcube was also a paragon of expandability.

Specs tend to be more important to people when a technology is still maturing. In the 90's, computers weren't fast enough, but were developing at a crazy pace. It was an exciting time and regular upgrades made sense. Nowadays, multicore CPUs and SSDs are ubiquitous, RAM is plentiful, and computers are all pretty good.

Apple made good PPC desktops, they just struggled to compete with the performance of x86 desktops. Resorting to multiple CPUs tended to increase prices. When Apple transitioned to Intel, their tower immediately became faster, better value, and able use its cavernous interior for something other than CPU heatsinks. You may be right, though, that once the iPhone represented the majority of their income, Apple cared less about pleasing the 'spec-wankers' (pro users). Eliminating internal expansion certainly simplifies driver support, configuration options, and manufacturing, shipping and storage costs.
 
  • Like
Reactions: BSDnostalgia
Agree with much of this, though SJ seemed pretty chuffed with the PM G3 case design when he unveiled it, lowering the side / logic board to oohs and aahs from the audience. His NeXTcube was also a paragon of expandability.

Sure, when Apple did towers, their designs were better than anyone else’s. G3/G4 and G5/MacPro were absolutely a class above the Win PC’s of the day.

My MP4,1 has the CPU tray that slides out, brilliant stuff.
 
Last edited:
  • Like
Reactions: BSDnostalgia
The 2013 MP was completely closed - the cards were in a proprietary form factor so no standard ATX GPU would fit. MPX was at least fully backwards compatible - it didn't interfere with the regular PCIe slot and indeed, many people installed off the shelf PC graphics cards.

How were MPX modules backwards compatible when GPUs and drive carriages for example had their own power supply lane? (non standard). You can't take an MPX GPU and put it in another PC. You could plug any plain old PCIe card sure it would work, but MPX modules on their own were dead end products. There's no way Apple would've kept continuing to make new GPUs and what not and 3rd parties would never continue to build upon it. It was just too propietary. It just should've never been a thing. Just do plain PCIe cards with no custom onboard PCIe power lane.

My criticism of MPX would be that it was over-engineered and unnecessary, and the sort of pointless reinventing the wheel that delayed the 7,1's release. Power could have been adequately provided by high power cable sockets next to each slot, and Thunderbolt video could just have been a mandated output on the back of MPX cards - no need to pipe it round the system.

As I said above, correct. They got rid of MPX in the M2 Mac Pro for probably this reason.

The AS transition meant Apple themselves lost interest, and the market was too small for anyone else to bother. Especially given the slots were backwards compatible anyway - people could just buy a PC model.

It's not interest, just over-engineered for no reason whatsoever. They were cool for marketing purposes. Other than that, useless. GPUs get dated very quickly. This is the problem with their thinking for the Mac Pro.

Not sure how 'easy' it would be to add 7000 / 9000 series driver support. Someone would have needed to write the drivers. In the past, that came for 'free', since they were already written for MBPs / iMacs.

Of course, with Apple being among the richest companies on Earth, it wouldn't have been much to ask for. Apple could have easily paid AMD to develop Mac drivers for their latest cards. The problem was, Apple were already struggling to match the GPU horsepower of a 2019 MP with their Ultra SoCs. Allowing MP users to stuff their machines with 9000 series cards would just raise the bar ever higher. They wanted the Ultra to be able to eclipse the MP in all ways. Even now, it can't match the RAM capacity of the 7,1.

I know this directly from someone on the inside, it's fairly easy to make AMD drivers because they used to have an AMD drivers team in-house (and before that NVIDIA). They probably fired them and/or didn't want to continue so they "forced-obsolete" it so people move to Apple silicon.

Apple has to notarize drivers, it's not that difficult especially since RNDA 2 > RDNA 4 wasn't a huge architecture jump anyway.

They could've at least gave us a 9xxx series drivers as a last hurrah for a dead product years ago but they didn't.

Apple Silicon is great for everything except slotboxes, which Apple aren't too keen on anyway. It's easy to see why they're OK with losing those customers to Windows / Linux.

Either way they lost a lot of customers in this niche product space after the 2013 debacle, and then by then the 2019 wasn't a feasible option. Most moved to Windows if they were doing anything 3D related for example. They just went "too pro" on the Mac Pro. Should've just made the case and made the whole thing as affordable as possible.

The big issue on Apple silicon is the on-die RAM/GPU, they probably saw no need to make a new SKU to support RAM slots and GPUs, they would lose performance anyway losing data on the logic board's tracers.


I don’t think Steve Jobs loved that, frankly. If you look at the products Apple made before he was ousted, he was always an all-in-one guy.
Being the aesthetician that he was, I think those big rectangular boxes from the 1990’s-2000’s offended his sensibilities. And if the G4 Cube hadn’t failed so miserably, I think he would have gotten rid of the towers a lot sooner.

In that sense I think the 2013 Mac Pro was really his legacy.

All I know is Steve wouldn't allow Apple to butcher the Mac Pro or let it die. I remember all his keynotes when he would get giddy when speaking about Mac Pro, you can tell it was a turning point for the company to cater to Pro customers something super speedy. He died in 2011, and the 2013 Mac Pro and forth is a Tim Cook debacle, he just doesn't get it. By killing off the Mac Pro they pushed many nich professionals into the hands of Windows (sometimes Linux). Steve always respected the pros.

The G4 Cube was also Steve's but it was never supposed to replace the Mac pro desktop variant. It was underpowered and just mostly aesthetics. The G4 Cube is somewhere between a Mac Pro and a Mac Studio. 2013 is not his baby. All Tim Cook did was streamline the Mac product line more and gave birth to the death of the Mac Pro.

It's sad really. There are still a lot of companies/individuals which need PCIe slots and internal drives. They either moved to Windows/PC or will do external Thunderbolt if they truly love the Mac.
 
All I know is Steve wouldn't allow Apple to butcher the Mac Pro or let it die.
Maybe not ~2013... when all Apple really had to do was make an official Hackintosh.

...but now, the choice is between focussing on Apple Silicon SoCs optimised for mobile and SFF, versus a PCIe-based Mac Pro workstation aimed at a shrinking market niche which meant throwing out most of the USPs of Apple Silicon. It's not 2013 any more.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.