That $650.00 12 core 5,1 upgrade I put together a month ago looks like an even better investment now.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Apple's Revamped Mac Pro to Launch in 2019
- Thread starter MacRumors
- Start date
- Sort by reaction score
Omg epic lol? Maybe you should stop watching keynotes for a while.
LOL.
http://web.archive.org/web/20050205230843/http://g4noise.com:80/news.php?ID=72
https://www.macworld.co.uk/feature/apple/antennagate-11-biggest-apple-scandals-3620624/
You mean take extra time to tune the Reality Distortion Field emitter to a wavelength that convinces people that bad products are actually good products.
EDIT https://www.macrumors.com/2018/01/30/apple-focus-on-software-quality-extends-to-mac/
"We make objectively bad software and need to take some time to reconsider our life choices".
I guess because they wanted to start from scratch- what we normally see is a continuous product pipeline with them working on the new machine before the previous one ships. In this case they dropped developing the Mac Pro entirely and then started it back up again sometime in 2017...
Why doesn't the Apple pro-Mac team just read up on Hackintosh? Done.
Wow, so in FORTY-TWO Years of products, the BEST you can come up with is TWO failures?
I'd call that a pretty good track-record!
Now, let's compare that with Microsoft; the inventors of "Patch Tuesday"...
And if you want to restrict it to Hardware Products, we can play that game, too:
https://www.engadget.com/2017/11/21/surface-book-2-battery-charging-gaming-issues/
https://en.wikipedia.org/wiki/Xbox_360_technical_problems
https://www.theverge.com/2018/2/1/1...ce-pro-4-screen-flickering-issues-flickergate
https://support.microsoft.com/en-us/help/4023476/surface-surface-touchscreen-problems-with-touch
https://www.digitaltrends.com/computing/surface-pro-4-16-common-problems-solutions/
https://answers.microsoft.com/en-us...4-screen/40e0962f-683e-4e4f-8466-5b5a1be386ad
https://answers.microsoft.com/en-us...ce-pro-3/b7e32892-5fc3-4cd2-91c6-0a69adf1b5b6
https://www.theverge.com/2017/8/14/16142490/microsoft-surface-consumer-reports-memo-return-rates
https://www.theregister.co.uk/2016/05/27/sleep_of_death_windows_10/
https://www.reddit.com/r/Surface/comments/6iz677/problems_w_new_surface_pro_2017/
https://support.microsoft.com/en-us...bleshoot-display-problems-with-surface-studio
https://support.microsoft.com/en-us...oubleshoot-power-problems-with-surface-studio
https://www.windowscentral.com/how-force-surface-pc-shut-down-and-restart-when-its-frozen
Shall I go on?
Intel have x86 architecture chips. Everyone wondered how successful they would be at running low power mobile versions of them. Turned out the answer was not very. But even Intels chip experts didn’t know that until they tried and spent a very large amount of r&d money.
I think the same experiment would be interesting with ARM in the other direction. The reason Intel had problems running low power x86 chip is they are CISC and as such do waste power when running simple operations. ARM however is RISC and as such has excellent power efficiency but a reduced instruction set. However, the iPad Pro has shown an ARM chip can perform advanced photo and video processing and run every aspect of a desktop operating system. So the reduced reduction set does not appear to be the barrier it had been assumed to be. Given that is so, it would indeed be interesting to see an ARM chip produced in a large die high TDP form factor.
We won’t know how it performs until we see it. Also a lot will depend on the compiler and software integration, especially for RISC. However, that is apples specialty, tight hardware and software integration.
Haven't read the whole thread so maybe someone has already mentioned this, but it seems to me an odd timeline given the earlier report (a week or so ago) that Apple is switching to their own proprietary CPU in 2020.
So you buy into the brand-new hotness of the Mac Pro in 2019, and in 2020 Apple switches to their own CPU? Wouldn't that destroy the "modular" aspect to switch CPU architecture?
Just seems weird to me; and totally Apple.
So you buy into the brand-new hotness of the Mac Pro in 2019, and in 2020 Apple switches to their own CPU? Wouldn't that destroy the "modular" aspect to switch CPU architecture?
Just seems weird to me; and totally Apple.
Cause and effect.
It was precisely because of all the outcry (right after the 2016 MBP reveal) that led Mac users to complain that Apple was abandoning the Pro market. I think that at one point, it got so bad that Apple had to come out publicly and announce that they were going to work on a new Mac Pro.
I think the whole issue goes back to Apple having a differing vision of what "Pro" entails from its user base. I believe that the iMac Pro had always been Apple's answer to the 2013 Mac Pro, but users simply wanted a tower PC which they could upgrade themselves.
I am of the opinion that any Mac Pro device would not have sufficient sales to even cover costs of development. However, there are advantages to having a halo Mac product which goes beyond paper profits alone. You need a Mac to develop iOS apps, so it could be a small price to pay to retain your developer base and ensure a healthy app market for your iOS devices. At the end of the day, Apple probably weighed the pros against the cons, and decided that they simply could not afford to alienate their Pro user market. Yet.
That's the issue. I don't think the new Mac Pro 2019 is going to be modular in the conventional sense. It's likely not going to have parts that are easily accessible or replaceable. Apple's idea of "modular" possibly means you can tack on additional accessories for added functionality, which again means spending more.
The iMac Pro has replaceable ram but it's very difficult to access and swap on your own. The Mac Pro may well have upgradable components; they just won't be very high on Apple's list of priorities.
I think the main problem with trying to predict and optimise for user workflows is that sometimes you just end up betting on the wrong horses. Workflows can and will change over time. That was the whole problem behind the 2013 Mac Pro. Apple bet on dual-graphics card workflows, while the industry would ultimately consolidate around single, more powerful cards.
I sure hope Apple knows what they are doing, because it kinda sounds like they have learnt nothing from the 2013 Mac Pro debacle.
I just read the tech crunch article. Very insightful. I'm glad to see Apple is spending time working through the issues to make the best possible solution. Unfortunately for me, timing is a bit off. Earlier this year marked my companies 4th year with the 2013 Mac Pro in our main edit bay. It's support dried up a year ago and it depreciated to the point that it can be mothballed to an interns desk to edit facebook videos. Sadly, the 2013 Mac Pro didn't make it for the count. Like clockwork, she started experiencing bizarre graphic issues. After 4 repair attempts, Apple gave up and traded me for an equivalent spec'd iMac Pro. Great for me and the company, but I still have 9k budgeted to replace equipment this year that I must spend. Hmm, the 8-year old 2010 Mac Pro at the interns desk is still chugging along fantastically. Glad we had it on hand to swap into the main bay while we were without hardware briefly. I was really hoping to spend my budget on a new Mac Pro, but will be forced now to grab an iMac Pro. I have to spend this year. Unfortunately, that means I won't be able to touch the "2019 Mac Pro" or it's design until about 2022. 
@sigmadog – weird indeed. It reminds me of all the effort Apple put into the Quad Core G5 (even reengineered with PCIe, SATA) in late 2005 only to announce it's death not very long after.
@sigmadog – weird indeed. It reminds me of all the effort Apple put into the Quad Core G5 (even reengineered with PCIe, SATA) in late 2005 only to announce it's death not very long after.
Assuming for the moment there’s at least some validity to the “dropping Intel” rumors, Apple would start the transition with the low power Macs first: MacBook and MacBook Air. Certainly there’s nothing to indicate they’ll have CPUs ready anytime soon in the 100/150 Watt range that would allow them to replace higher-end iMacs or the Xeons used in Mac/iMac Pro.
But I’m not at all convinced of the veracity of the rumor. It could be that Apple’s planning nothing more than a “MacBook ARM” that will mainly be targeted at schools and/or lower-income countries where even a $500-800 Intel-based Mac is not affordable. I’m still not convinced of any wholesale plan to ditch Intel (or AMD).
Last edited:
While I'd find the move to ARM exciting and even within realms of possibilities that to be powerful enough and cost efficient enough to challenge Intel, it's the software side that would screw people over. Unless transitioned really, really well. The stuff you want to run fast on Mac Pro is a legacy nightmare to move to something like iOS and if you signal that that's where things are going... you need to be rather desperate for the Mac Pro. Which is rather unlikely in 2019, since how the heck did you survive so far. Might just as well take an iMac at that point or run that cMP a few more years until the software catches up.
One of the fundamental lessons should be that when people need more serious computers they start to specialize and want/need/wish for something that benefits them. And at the same time, these folks already know how to do their stuff, and what do you think you're going to change as Apple that's not yet being done on PC side? Realize that many cameras prefer XQD / CFast slots instead of SD cards?
Yeah. And it's quite hard work too since no one else does workstations either.
Haha! But it needs to be a pretty box. And there's just so few boxes by Dieter Rams to get inspired from.
You might wanna brush up on inner workings of processors - Intel ain't really CISC anymore but the conversions are pain. And there were RISC workstations long ago that were fully functioning computers so iOS on ARM has shown nothing we'd not know from the past when it comes to reduced instructions set. Turing completeness and all that.
https://stackoverflow.com/questions...l-hide-internal-risc-core-in-their-processors
That said, I'd be keen to see what they could do in ramping ARMs up.
You do realize that the head of design is Ive, who has been by far the most powerful executive under both Jobs and Cook, right? If anything, he has gotten more control under Cook (adding things like software experience and building design), and started delegating that control as a result.
[doublepost=1523077391][/doublepost]
It is important in that it generates cash for Apple. The problem is that it is advertised as a switcher machine for people who are just going to bring their existing (aka old) monitor, keyboard, and mouse. This leaves it stuck in the past a bit - I'd actually be a bit surprised to see it get tb3/usb-c when it is finally revved - certainly not tb/usb-c exclusively, unless it is being revved in 2021...
[doublepost=1523078649][/doublepost]
Sure, they told you what they were thinking last year. But to elaborate:
They wanted to build a system with two GPUs and a crapload of cores because everything was pointing toward pro software needing to become geared toward massively parallelized systems. They wanted to make a small, sexy machine that had more cylinders than any PC could have in a case 8x the size. So they designed a system around a very specific thermal profile. They put in two GPUs- one which couldn't even do video out, because it was dedicated to data processing.
Then the industry went on to continue to push for a few threads with lots of CPU/GPU horsepower behind them. Since CPUs and GPUs were growing in transistor count (and thermal profile) at different rates, upgrading to a newer GPU became really challenging - they would basically have to under clock a new GPU design to match the heat output of the intel CPU side of the triangle.
After likely a few failed attempts (possibly including moving to Vega 64), they decide they need a new design. It will take long enough they decide to just be upfront, rather than having the professional industry all continue to think Apple forgot about them.
I honestly don't know if "modular" in apple parlance means user-upgradeable. I suspect it means more that they don't have to redesign the system to deal with the heat/power consumption needs when new components from Intel, AMD or Nvidia come out.
Where Apple controls the whole workflow and all the underlying software and hardware, they have way more power to make meaningful changes (and make them faster). IMHO the attraction of pro users to the Mac platform has always been more efficient workflow, increasing productivity making the cost worthwhile.
I couldn't imagine the pain of waiting on Adobe to make some photoshop feature stop hanging the system, or convincing them to modify some dialog to try to get things from four clicks down to two. Not to say there aren't possibilities there (and I doubt people are on a pure apple software workflow doing pro work), but Apple is going to have a way easier time evolving their own tools.
[doublepost=1523084295][/doublepost]
A temporary thermal throttle during a synthetic benchmark is not unusual. The system goes from 0 to 100%, blasting through the low/no fan cooling zone and into a danger zone before the fans can even spin up.
I counted 13 standard ports.
Citation needed? Unless you are equating battery life with performance (they definitely *could* manage a slightly larger Wh battery in the tMBP - my understanding was there was a battery issue late in the design that caused the reduced size), or keyboard travel distance with performance.
Apple's primary design limitation was the 80W max TDP and > 1 week suspend time (with power nap engaged). This precluded desktop ram (to go over 16GB with Intel's current chip designs), all modern Nvidia parts, etc.
I can understand people wish apple made a portable workstation lab-top with a 225W TDP and a GTX 1080, that was usually too hot to actually have on your lap and could only last 45 minutes on battery power. But the reason apple didn't make that computer wasn't "thinness".
Apple 3
Lisa
Copeland (and the state that "classic" Mac OS was in before they basically switched to NextStep)
Pippin
Newton
eMate (bet you've all forgotten that one!)
Quicktake (early lead in affordable digital photography went nowhere)
The mess the Mac range got into in the "Performa" era.
Mac G4 Cube
~2011 MacBook GPU failures
2016 MBP keyboard failures
...and that's just off the top of my head.
Then there are things like the original iPhone and the original MacBook Air which showed huge innovation and went on to great success but weren't actually much good until the first major update - Apple could afford a second go at getting these right because they were new product lines, and they had perfectly good iPods and MacBook Pros to offer their less adventurous customers in the meantime. As I've already mentioned, even the release of the retina MacBook Pros was accompanied by a substantial update of the classic MBPs to ensure a smooth transition period.
The recent problem is that - possibly starting with the Mac Pro - they've been letting established products go to seed and become hopelessly outdated before suddenly replacing them with "courageous" radical re-designs that require major re-thinks by customers.
Apple's real problem is this: the "cheesegrater" Mac Pro was left for several years without updates (in Europe it was actually discontinued a year before the nMP was even announced because Apple couldn't be bothered to add a fan guard to meet new EU safety regs) - and it was then replaced with a radically different type of machine that - even Apple now admits -wasn't suitable for many customer's needs. Now, that has been left untouched since 2013 and all there is is the promise of some totally undefined new Mac Pro sometime in 2019. All we know from this latest release is that Apple are intent on trying to do something "clever".
So, even if the new Mac Pro does appear next year, who has any confidence that it isn't going to be another one-shot wonder that never gets updated? What's the chances that, come 2024, when your lease is up and you're ready to update, Apple will still be selling the same old 3-year-old mMP (complete with Spectre and Meltdown) at its original price? Or maybe they'll have dumped it in favour of a machine with 32 ARM cores and 128 vector processors which would be truly amazing if it wasn't for the minor problem that you'd only be using it to run x86 emulation for compatibility with the software on which you depend...
[doublepost=1523104220][/doublepost]
...so they took a punt on a horse and lost. That's not the problem - if you don't take risks you don't win. The problem was - figuratively - that they'd already given up the day job (the traditional Mac Pro tower - which had gone for a couple of years without updates and had already been discontinued in Europe) on anticipation of that horse coming in. That's careless.
If Apple want to take a punt on some exciting new modular workstation concept in 2019 then that's great - the problem is that, meanwhile, they've let the nMP and the Mac Mini go obsolete so all they have to offer the pro/enthusiast/power user who needs something today is the iMac Pro - and not everybody wants an all-in-one.
Thing is, Microsoft can afford to play around with expensive concepts like the Surface Studio safe in the knowledge that if Windows users don't like it they can hop on the web and find a dealer that will build a mini-tower to their exact specification - and they'll still be a windows user. You can't do that with a Mac (newsflash: Apple could pull the plug on the hackintosh community overnight if they decided to enforce their software licensing, or do it inadvertently though an OS update - which means they're only suitable for hobbyists). Apple need a safe, solid, powerful headless desktop with PCIe slots in their range - not just to tick the "moar power!!!" box, but to tick the "other" box. E.g. I don't need an 18 core Xeon and 256GB of ECC RAM but I could really do with another half-dozen USB slots that weren't connected via a hub or dock (which can sometimes cause problems) and, while the 5k iMac display is nice, I'd really rather have a pair of slightly bigger 4ks.
The development for the iMac Pro started in 2016 (or even a bit earlier). In 2016, it was the "new Mac Pro" in Apple's mind.
Only because Apple seem to think that any new Mac Pro has to be some exotic new form-factor that pushes the boundary of industrial design. Designing a PCIe mini-tower is a solved problem. Foxconn could probably have a container ship load of them in nice stylish aluminium boxes sailing into SF Bay in time for WWDC. OK, realistically, nothing is ever quite that quick and simple but, had they decided to go that way back in early 2017 when they admitted the problem, a company with Apple's resources could have had product on the shelves by now.
Its quite true that the market for that is shrinking - on the one end, laptops and all-in-ones are getting more powerful, and on the other end seriously high number crunching requirements can be met on-demand by cloud computing or a rack of cheap gaming GPUs. All the more reason to not pour resources into something exotic to fill the shrinking gap when something simple and mature would fill the need.
I have the 2013 Trashcan Mac, along with an external DVD, 3 external enclosures which hold 12 SSD, and a 14-port USB-3 hub to handle my scanner, iDevices, and laser and photo printers.
I would love to have a tower which can support everything.
I would love to have a tower which can support everything.
I wouldn't mind this so much if I could buy the trashcan Mac Pro for a price that's reflective of where it's at in the life cycle of those parts.
Ha. I like your idea. On that basis they should pay US to take them.
There’s is so much Apple apologist content here. For the sake of everyone else, let’s just agree that I am too critical and you are not. For example, for a “Pro” product and priced accordingly, the MacBook Pro has a pathetic video card. Apple cannot add more powerful components to MacBook or MacBook Pro because “thin”. Glad you mentioned battery too, yeah, there is a gaping space in the case... why not have put a bigger battery? Yes, the Mac Pro does have some standard ports but few and no standard slots... in a workstation you need slots to accommodate industrial standards. Instead Mac Pro users have few or no upgrade paths and huge knots of cables hanging about. The iMac “Pro” is a nice machine but terrible form factor for a workstation. The thermal throttling is noticeable and affects real world performance. It literally always thermally throttles.
I am almost completely fed up with Apple at this point. As others pointed out, I simply don't believe anymore that we get that seemingly everyone wants, i.e. an upgraded "Cheese Grater 2.0" tower-style machine with PCI slots and all that. Yes, designing a computer takes it time, but sooo long, if all that they need is to make another tower? I do not buy that, and I more and more think that "modular" means something like a "lego computer" at best, with expansion modules available from Apple only of course. Starting at only $14.999 and up to $29.999 ...
It seems that Apple has completely lost track what a sizable part of their user base - and the most loyal one as well - actually wants. And I say "wants", not "needs" - and thats imho already part of the misconception, i.e. Apple does mix up "want" and "need", and on top of that they assume they know what we "need".
I also think that "pro whatever team" does not reflect their actual target audience, and they also do not get that quite a lot of users which would buy an updated tower are not fitting in the typical scheme of what they deem to be a "pro user".
Plus, they do not seem to get that the actual reason many (if not most) people run such systems is not a shiny, glitzy "beautifully designed" bling bling case available in rose gold. It is the function that counts, not the form! And more importantly, it is the (mac)OS that makes the actual difference, stupid!
So, I think I am simply forced to build another hackintosh system, as I already did almost immediatly after the 2013 announcement, since it was quite clear the 6,1 does not do what I want (and need), and it was also quite apparent at lewast to me that this would fail anyways.
But the whole hackintosh experience wasn't too much fun at least for me, and all the money I saved was burnt by this endless tinkering to get a decent running system. So I was waiting all that years in the hope that this would only be a stop gap measure ... And as macOS is still a huge argument for me (Windows and Linux are simply no alternatives for many different reasons) I think I will look for a decent X299 based system to replace my aging hack. And I think I have to do it now, because god knows what happens else in the future. I mean for example a "pro" machine with ARM chips would simply kill my workflow (as I rely on virtualization of all kinds of OS), and at that point I would even forced to move to windows. Please not!
So at least I hope I can milk my new hack until it is unable to support macOS version whatever in the maybe not so near future.
It is really ridiculous. Unbelievable.
It seems that Apple has completely lost track what a sizable part of their user base - and the most loyal one as well - actually wants. And I say "wants", not "needs" - and thats imho already part of the misconception, i.e. Apple does mix up "want" and "need", and on top of that they assume they know what we "need".
I also think that "pro whatever team" does not reflect their actual target audience, and they also do not get that quite a lot of users which would buy an updated tower are not fitting in the typical scheme of what they deem to be a "pro user".
Plus, they do not seem to get that the actual reason many (if not most) people run such systems is not a shiny, glitzy "beautifully designed" bling bling case available in rose gold. It is the function that counts, not the form! And more importantly, it is the (mac)OS that makes the actual difference, stupid!
So, I think I am simply forced to build another hackintosh system, as I already did almost immediatly after the 2013 announcement, since it was quite clear the 6,1 does not do what I want (and need), and it was also quite apparent at lewast to me that this would fail anyways.
But the whole hackintosh experience wasn't too much fun at least for me, and all the money I saved was burnt by this endless tinkering to get a decent running system. So I was waiting all that years in the hope that this would only be a stop gap measure ... And as macOS is still a huge argument for me (Windows and Linux are simply no alternatives for many different reasons) I think I will look for a decent X299 based system to replace my aging hack. And I think I have to do it now, because god knows what happens else in the future. I mean for example a "pro" machine with ARM chips would simply kill my workflow (as I rely on virtualization of all kinds of OS), and at that point I would even forced to move to windows. Please not!
So at least I hope I can milk my new hack until it is unable to support macOS version whatever in the maybe not so near future.
It is really ridiculous. Unbelievable.