So fully 6 years between them by 2019, that's crazy to think of the gap left. That a very long time in silicon terms.
The "pro workflow team" is encouraging. I'm hoping for myself it allows a lot of min/maxing, not fixed to relatively high end hardware on all parts like the iMac Pro. I need a lot of CPU for data science, but a GPU goes entirely unused, so I wouldn't need Navi Pro with HBM2 adding to the cost.
Yes, it's an embarrassment that Apple has not been able to keep the entire Mac lineup fresh each year. Let's hope they make the corrective actions necessary and it's not "just talk".
To be truly environmentally friendly, Apple would turn the industry of computer design on its head and go truly "modular" where to upgrade means just adding another module to an existing grid. Need more power, add more modules. Not much money, just start with one module. We have the fast networking technology and universal cabling now. Let's see it, Apple. Just bring back Xgrid!
"Throughout, the idea of modularity was omnipresent. An iMac Pro with two iPad Pros hooked up to it allows for direct control, shortcuts and live access to the Logic manual all while you’re mixing a song on the main device. "
Uh wait, you can do that? If not...Why not?! I always wanted tighter integration to use iPads to make Macs better.
Likewise. Also, the bigger point is being missed by Apple here... people want touchscreen capability on their displays!
God, I hope Jony doesn't have any input into the design of this machine. Otherwise we'll be seeing something that looks pretty and is totally unfit for pro use.
You beat me to it. Looks like Ive has been sidelined on this project. Rightly in my opinion, since the MacPro should be all about engineering.
Shouldn't they discount the current model? 15-20% discount might keep a few more on the platform
Why do you say that? Jony Ive rarely does product marketing interviews. Also anyone who believes the current Mac Pro is a cylinder because Jony Ive demanded it is demented. Craig Federighi even said they didn’t go into the project with a specific shape in mind. The company decided what they thought the future of the pro market would be and designed a product around that.
So fully 6 years between them by 2019, that's crazy to think of the gap left. That a very long time in silicon terms.
The "pro workflow team" is encouraging. I'm hoping for myself it allows a lot of min/maxing, not fixed to relatively high end hardware on all parts like the iMac Pro. I need a lot of CPU for data science, but a GPU goes entirely unused, so I wouldn't need Navi Pro with HBM2 adding to the cost.
Yes, it's an embarrassment that Apple has not been able to keep the entire Mac lineup fresh each year. Let's hope they make the corrective actions necessary and it's not "just talk".
To be truly environmentally friendly, Apple would turn the industry of computer design on its head and go truly "modular" where to upgrade means just adding another module to an existing grid. Need more power, add more modules. Not much money, just start with one module. We have the fast networking technology and universal cabling now. Let's see it, Apple. Just bring back Xgrid!
Is there a source that shows making everything “modular” would make their products more environmentally friendly.
How would using slotted RAM and NAND in the iPhone, the device they sell he most of, even be feasible? All their Macs are such comparatively low volume that it’s a moot point if you’re saying that the MacBook should all that plus the CPU, WiFi cards, slotted BT modules, and on-and-on.
Everything I’ve read is that hardly anyone ever upgrade their devices so would that be a waste of resourcea and more chances for HW issues to make everything modular?
I bet Apple do a lot more to be environmentally friendly by simply not including a PSU with every iDevice purchase without charging you extra for it. It’s the thickest item in the box, they don’t tend to go bad, they contain a lot of harmful metals, and we all seem to have a drawer of them.
I didn’t like that narrative... “Not this year”, “understanding from soup to nuts” [still?], “internal workflows with real content” sound like bla bla bla to my ears.
Those quotes from Apple sound horrible. They should have just said the Mac Pro will be available in 2019, and left out these weird statements or rationale, which sound devoid of sense to me. Designing for workflows? That’s what drove them down the 2013 Mac Pro route.
The point of the top end machine is to allow for enough flexibility to account for unknown workflows in the future, instead of having to start over again like they are when the future workflow is mispredicted.
So fully 6 years between them by 2019, that's crazy to think of the gap left. That a very long time in silicon terms.
The "pro workflow team" is encouraging. I'm hoping for myself it allows a lot of min/maxing, not fixed to relatively high end hardware on all parts like the iMac Pro. I need a lot of CPU for data science, but a GPU goes entirely unused, so I wouldn't need Navi Pro with HBM2 adding to the cost.
We will see buut im still not convinced Apple gets it! As you point out "pro" needs vary widely and Apple just doesnt sem to understand this.
Apple whines about single digit sales for the Mac Pro but you have to ask what did they expect from the trash can. As you noted not everybody needs dual GPU cards, in fact integrated GPUs can be good enough. Likewise for some storage is the big issue. Which brings up the idea that the machine must be a faamiky of components one if which needs to be a storage array.
What is really interesting here is all the focus on performance testing. If this was an Intel platform performance would be limited by the Intel supplied chips. In otherwords not much they can control performance wise. This makes me wonder if the machine will be one if the rumored ARM based machines. It would throw the industry a curve to have the first ARM based Mac be based on an ARM processor. There are at least two commercial chips available today that could lead to a very high core count Mac Pro. So maybe a 64 core Mac Pro is on the way.
Crazy? Maybe but some of the statements just seem weird if this was an Intel based machine.
I'm curious if "2019" means that they will still officially announce it at WWDC and set a late winter/ spring release(early 2019), or are they really pushing it to late 2019 and possibly won't mention it again for more then 12 months from now.
Those quotes from Apple sound horrible. They should have just said the Mac Pro will be available in 2019, and left out these weird statements or rationale, which sound devoid of sense to me. Designing for workflows? That’s what drove them down the 2013 Mac Pro route.
The point of the top end machine is to allow for enough flexibility to account for unknown workflows in the future, instead of having to start over again like they are when the future workflow is mispredicted.
I found it weird too. And it came across like Apple finds this really hard work and wants everyone else to know how hard it is. I’m sure that’s not how they intended it but that’s how it comes across, One Mac developer on Twitter quipped, stick a Xeon in a box and call it a day. I think Apple’s messaging here is off. And after all these years of being in the Mac businsss they have to go out and hire professionals to see how they work and what their workflows are? Seriously?
So fully 6 years between them by 2019, that's crazy to think of the gap left. That a very long time in silicon terms.
The "pro workflow team" is encouraging. I'm hoping for myself it allows a lot of min/maxing, not fixed to relatively high end hardware on all parts like the iMac Pro. I need a lot of CPU for data science, but a GPU goes entirely unused, so I wouldn't need Navi Pro with HBM2 adding to the cost.
We will see buut im still not convinced Apple gets it! As you point out "pro" needs vary widely and Apple just doesnt sem to understand this.
Apple whines about single digit sales for the Mac Pro but you have to ask what did they expect from the trash can. As you noted not everybody needs dual GPU cards, in fact integrated GPUs can be good enough. Likewise for some storage is the big issue. Which brings up the idea that the machine must be a faamiky of components one if which needs to be a storage array.
What is really interesting here is all the focus on performance testing. If this was an Intel platform performance would be limited by the Intel supplied chips. In otherwords not much they can control performance wise. This makes me wonder if the machine will be one if the rumored ARM based machines. It would throw the industry a curve to have the first ARM based Mac be based on an ARM processor. There are at least two commercial chips available today that could lead to a very high core count Mac Pro. So maybe a 64 core Mac Pro is on the way.
Crazy? Maybe but some of the statements just seem weird if this was an Intel based machine.
"This makes me wonder if the machine will be one if the rumored ARM based machines."
But the very recent rumor with respect to ARM Macs mentioned that, that transition won't begin until around 2020.
"Tiny % of market"...could this be because Apple is CLUELESS and makes a foolish "pro" machine that is a joke. The Mac Pro was/is nothing, that is why sales are so minimal. NOW they have a "pro" dev team? This is an admission that they have cared nothing for pro users and pro systems, a bald admission really. Chumps can't make a great computer. Apple is the iphone company, like most corporate grifters they go only where the easy cash is. Apple, particularly Apple under Cook, has become almost nothing. Burned out, over-rated, drenched in sloth. This is a monkey butt situation.
You have no idea of what you're talking about! Why so much hate for an American success story? Are you anti-American? Not to mention a computer success story. MANY companies have stood on the back of Apple and copies everything they have done. This includes just about all Asian Tech over the last decade computer and phone wise. SO EASY for you to sit there and judge something you were not involved with? Why are you even here? You have nothing better to do a slam someone?
So fully 6 years between them by 2019, that's crazy to think of the gap left. That a very long time in silicon terms.
The "pro workflow team" is encouraging. I'm hoping for myself it allows a lot of min/maxing, not fixed to relatively high end hardware on all parts like the iMac Pro. I need a lot of CPU for data science, but a GPU goes entirely unused, so I wouldn't need Navi Pro with HBM2 adding to the cost.
We will see buut im still not convinced Apple gets it! As you point out "pro" needs vary widely and Apple just doesnt sem to understand this.
Apple whines about single digit sales for the Mac Pro but you have to ask what did they expect from the trash can. As you noted not everybody needs dual GPU cards, in fact integrated GPUs can be good enough. Likewise for some storage is the big issue. Which brings up the idea that the machine must be a faamiky of components one if which needs to be a storage array.
What is really interesting here is all the focus on performance testing. If this was an Intel platform performance would be limited by the Intel supplied chips. In otherwords not much they can control performance wise. This makes me wonder if the machine will be one if the rumored ARM based machines. It would throw the industry a curve to have the first ARM based Mac be based on an ARM processor. There are at least two commercial chips available today that could lead to a very high core count Mac Pro. So maybe a 64 core Mac Pro is on the way.
Crazy? Maybe but some of the statements just seem weird if this was an Intel based machine.
But the very recent rumor with respect to ARM Macs mentioned that, that transition won't begin until around 2020.
I'd like to note that Jobs underpromised and overdelivered on the PPC to Intel transition. That said, I don't see Intel Macs going away even after ARM Macs are part of the lineup. This is a very different need compared to the faltering PPC chips. While Intel is faltering, it's faltering for all Windows PCs, too, so Apple wouldn't be playing catch up here, but instead jumping ahead.
As somebody who desperately needs a new headless desktop Mac, it's incredibly frustrating that Apple can't ship one. Apple's determination to create the 'perfect' Mac Pro instead of just shipping a damn box leaves us with no options. At this point I'd take a damn G5/DTK chassis
It has been a full year since Apple's Mac Pro apology and all they could bring to the table was 'we have a slightly better idea what to do now, because we asked some people'? 🤷♂️ The transparency is great, but c'mon
This is the problem with pre-announcing dates waaaaayyyyy in advance. Which old Apple knew, but new Apple is doing more and more these days. It’s honestly one of the weirdest changes.
I'm all for talking with users, but one thing I found weird was Apple hiring outside users to watch their workflows.
Doesn't Apple already employ some of the biggest groups of Mac pro users? E.g., developers, video makers. Seems like there's an internal communication problem.
Those quotes from Apple sound horrible. They should have just said the Mac Pro will be available in 2019, and left out these weird statements or rationale, which sound devoid of sense to me. Designing for workflows? That’s what drove them down the 2013 Mac Pro route.
The point of the top end machine is to allow for enough flexibility to account for unknown workflows in the future, instead of having to start over again like they are when the future workflow is mispredicted.
I found it weird too. And it came across like Apple finds this really hard work and wants everyone else to know how hard it is. I’m sure that’s not how they intended it but that’s how it comes across, One Mac developer on Twitter quipped, stick a Xeon in a box and call it a day. I think Apple’s messaging here is off. And after all these years of being in the Mac businsss they have to go out and hire professionals to see how they work and what their workflows are? Seriously?
The only reason I can think of for making those statements is that thought they would make 2018 last year. Now in 2018, it’s obvious they can’t make 2018, and have to tell their customer base that it will be 2019, and they need to add some color to the announcement or the press release.
They aren’t going the generic white box route (stick a Xeon, ATX board in a box) that almost anyone could build. That adds no value from a hardware perspective. It’s going to be a custom form factor, with their take on cooling design and footprint. I wouldn’t bet on it using standard PCIe cards either. It may likely have room for a couple of 11.5” PCIe cards, but I wouldn’t bet on it. I’m taking they want it to be more “modular” so that they aren’t caught being unable to upgrade after spending tens to hundreds of millions on the manufacturing process and supply chain.
They could still add a Broadwell Xeon, a couple of lower power Vega GPUs, add TB3 and Gbit Ethernet into the cylinder and call it a day too. But, all funding for the cylinder likely was zeroed out in 2015 or whenever it was, so that is not even an option.
Will it take them to 2019 to give the mini a decent bump too? I have 4 x 2011 i7 server models waiting for an upgrade. It gets a little bit tiresome to scour e-bay for used 2011/2012 server configs.
No not if they are using off the shelf Intel parts. This leaves several other possibilities:
1. Intel and Apple are working on a custom chip set. Right now there seems to be zero possibilty here due to the low volume. 2. Apple will be using AMD parts here. Custom is possible and maybe cost effective due to AMDs need for design wins. Even if they dont go custom AMDs chips now have respectable performsnce. 3. Apple goes ARM with an in house designed chip. Assume msny cores here, im think 32 would be the minimal number for 2019. This would flip Apples transition to ARM upside down, leading from the high performance end instead of a Mac Book. 4. They could go ARM with a third party server chip. This has some interesting possibilities especially if they worked a technology sharing deal with the chip marketer. In any event a 64 core machine would be easy.
Well that is 4 possibilities too chew on. You need to ask yourself why they need so much time for the software issues they referenced. The only explanation i can come up with is that the machne will have a lot of non Intel hardware in it.
To put it another way, yeah something is fishy at Apple.
Comments
Yes, it's an embarrassment that Apple has not been able to keep the entire Mac lineup fresh each year. Let's hope they make the corrective actions necessary and it's not "just talk".
To be truly environmentally friendly, Apple would turn the industry of computer design on its head and go truly "modular" where to upgrade means just adding another module to an existing grid. Need more power, add more modules. Not much money, just start with one module. We have the fast networking technology and universal cabling now. Let's see it, Apple. Just bring back Xgrid!
How would using slotted RAM and NAND in the iPhone, the device they sell he most of, even be feasible? All their Macs are such comparatively low volume that it’s a moot point if you’re saying that the MacBook should all that plus the CPU, WiFi cards, slotted BT modules, and on-and-on.
Everything I’ve read is that hardly anyone ever upgrade their devices so would that be a waste of resourcea and more chances for HW issues to make everything modular?
I bet Apple do a lot more to be environmentally friendly by simply not including a PSU with every iDevice purchase without charging you extra for it. It’s the thickest item in the box, they don’t tend to go bad, they contain a lot of harmful metals, and we all seem to have a drawer of them.
Long live iMac Pro.
I know a lot of folks want the cheese grater back, but there's no reason for a computer to be that goddamned heavy.
2) What about a compromise with a tower design that is much smaller?
The point of the top end machine is to allow for enough flexibility to account for unknown workflows in the future, instead of having to start over again like they are when the future workflow is mispredicted.
Apple whines about single digit sales for the Mac Pro but you have to ask what did they expect from the trash can. As you noted not everybody needs dual GPU cards, in fact integrated GPUs can be good enough. Likewise for some storage is the big issue. Which brings up the idea that the machine must be a faamiky of components one if which needs to be a storage array.
What is really interesting here is all the focus on performance testing. If this was an Intel platform performance would be limited by the Intel supplied chips. In otherwords not much they can control performance wise. This makes me wonder if the machine will be one if the rumored ARM based machines. It would throw the industry a curve to have the first ARM based Mac be based on an ARM processor. There are at least two commercial chips available today that could lead to a very high core count Mac Pro. So maybe a 64 core Mac Pro is on the way.
Crazy? Maybe but some of the statements just seem weird if this was an Intel based machine.
But the very recent rumor with respect to ARM Macs mentioned that, that transition won't begin until around 2020.
Doesn't Apple already employ some of the biggest groups of Mac pro users? E.g., developers, video makers. Seems like there's an internal communication problem.
They aren’t going the generic white box route (stick a Xeon, ATX board in a box) that almost anyone could build. That adds no value from a hardware perspective. It’s going to be a custom form factor, with their take on cooling design and footprint. I wouldn’t bet on it using standard PCIe cards either. It may likely have room for a couple of 11.5” PCIe cards, but I wouldn’t bet on it. I’m taking they want it to be more “modular” so that they aren’t caught being unable to upgrade after spending tens to hundreds of millions on the manufacturing process and supply chain.
They could still add a Broadwell Xeon, a couple of lower power Vega GPUs, add TB3 and Gbit Ethernet into the cylinder and call it a day too. But, all funding for the cylinder likely was zeroed out in 2015 or whenever it was, so that is not even an option.
No not if they are using off the shelf Intel parts. This leaves several other possibilities:
1. Intel and Apple are working on a custom chip set. Right now there seems to be zero possibilty here due to the low volume.
2. Apple will be using AMD parts here. Custom is possible and maybe cost effective due to AMDs need for design wins. Even if they dont go custom AMDs chips now have respectable performsnce.
3. Apple goes ARM with an in house designed chip. Assume msny cores here, im think 32 would be the minimal number for 2019. This would flip Apples transition to ARM upside down, leading from the high performance end instead of a Mac Book.
4. They could go ARM with a third party server chip. This has some interesting possibilities especially if they worked a technology sharing deal with the chip marketer. In any event a 64 core machine would be easy.
Well that is 4 possibilities too chew on. You need to ask yourself why they need so much time for the software issues they referenced. The only explanation i can come up with is that the machne will have a lot of non Intel hardware in it.
To put it another way, yeah something is fishy at Apple.