Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

deconstruct60

macrumors G5
Mar 10, 2009
12,309
3,902
and if there is an 2022 mac pro X86_64 cpu / video card bump?
Will mac os end rosetta in MacOS 14? if they also end x86_64?

Will they keep rosetta and x86_64 around till they end rosetta

Apple has already shown that Rosetta2 isn't solely about MacOS.



Apple isn't going to make all x86_64 Linux apps going away in two MacOS iterations. They don't have primary control over that. Rosetta 2 in these VM imagines will indirectly encourage some subset of commercial Linux vendors to port, but Apple doesn't have a "hammer" here to threaten and cajole folks into moving faster.


The major difference between Rosetta and Rosetta 2 isn't the targets and sources. It is ownership. Apple didn't own Rosetta. It was owned by Transitive and later IBM. It cost money to license for a longer period of time and its ownership was changing ( Apple wasn't going to browbeat IBM into favorable licensing terms. )

Rosetta 2 is more so of a sunk cost that Apple has already invested. If Apple continues to ignore AVX and Virtualization calls of x86_64 along with any new extensions ( Intel AMX , etc. ) then the costs to keep Rosetta going aren't going to be large. The more M-series Apple sells the, then more systems can amortize moderate ongoing costs over. They aren't likely going to be in a hurry to kill it. ( took years of work to get done. Now is the "pay day" for that to doing charge backs into every new Mac sold. ) . Once it is profitable they are just getting a lower return on investment to stop early.

However, Rosetta is unlikely to be a "get out of jail free" card for macOS on Intel. Those systems will be rolling onto the Vintage/Obsolete list at a regular rate. That work is on going costs that deliver zero new revenue to cover expenses once Apple stops selling Intel systems. ( likely next year or so. Apple has goosed the 2018 Mini for the last almost two years paired with a M1 primary variant. Goosing the MP 2019 for another year or so wouldn't be odd. Especially if the M-series Mac Pro was half sized and missing some functionality. ). Apple charges up front future upgrades when you buy a new Mac but that becomes a fixed pot of money that only gets smaller with each new upgrade they issue. At some point the pot of money get used up. And Apple stops because they don't work for free and not a non-profit charity.

Rosetta2 would require a defined subset of the macOS libraries live on so that have x86_64 versions. But the bulk of the apps and complexity of macOS (the of the GUI layers that causual users attach to "mac OS" ) would be are not needed for the Rosetta infrastructure.
 
Last edited:

deconstruct60

macrumors G5
Mar 10, 2009
12,309
3,902
Not that I think Apple will abruptly cut any support. A possible scenario could be them extending security updates of the last x86-64 MacOS release beyond the usual two years. Such a move could mitigate anger among existing owners of relatively new Intel Mac. Unlike PowerPC to Intel transition, most Intel Mac's are still capable and useful machines.

And yet Apple got rid of the Mac Pro 2013 on this iteration. It is definitely an outlier. Apple left it on the "for sale" list for 6 years. Adding another 6 to that was likely going to get into trouble. The Mini 2018 being extended into 2022 is also likley to hit some later blow back. Decent chance they are 1.5 years on countdown clock even though kept selling them ( use M1 Mini intro as 'start' date for countdown.)

However, If look at rest of line up though pretty much anything pre-2017 has be axed. They are leaning more aggressively into no cutting any slack for something that is about to go onto the Vintage/Obsolete list. Looking more like if something could go onto Vintage during macOS 13 Ventura's first year of deployment, then it got axed/'Steved' . Maybe they ease up after gotten rid of everything not T2 based . By next iteration , anything not T2 is likely gone.
[ iMac Pro "end of sale" in 2021 and if shave another year off ; 2025. Just 3 more. ]


Sometimes Apple tries to kick the can down the road. Mac Pro 2010 hitting coverage retirement before the Mac Pro 2012 did. That was driven more by their inaction than on the 2012 hitting above some "don't retire me benchmark" performance number. Nor some modularity "get out jail free" card.

it is a definite prune deliberately approach.
 
  • Like
Reactions: JMacHack

kvic

macrumors 6502a
Sep 10, 2015
516
459
LPDDR5 and DDR5 do not approach ECC the same way. Nor are their standard bus widths the same (which partially drives the different choices ).

"... As DDR5 and LPDDR5 support much higher data-rates than their predecessors, they support additional ECC features for enhancing the robustness of the memory subsystem. On-die ECC in DDR5 and Link-ECC in LPDDR5 are two such RAS schemes to further bolster the memory subsystem RAS capabilities. .."
https://www.synopsys.com/designware-ip/technical-bulletin/error-correction-code-ddr.html


There is no default "built in" ECC for LPDDR5. The standard LPDDR5 is aimed protecting the amphoral transfers of data on the bus/channel from the Memory package to the SoC package. When the data is at rest it doesn't protect anything. ( that saves a substantial amount of capacity as a trade off. )

DDR5 is the one that has "at rest" requirement turned on all the time. The link protection is optional and delivered via a wider side channel.

LPDDR5 can do it without a side channel, but that will take a bandwidth hit as the ECC data has to trail behind the data all the time on a relatively narrow channel ( not travel in parallel. ) . Seems unlikely Apple is going to toss data bandwidth out the window.

Apple is pragmatically running a "poor man's HBM" memory solution using LPDDR building blocks. So they cluster 4 LPDDR buses together and aim them at the same RAM die stack . They could 'weave' the parallelism in that subsystem to avoid the inline bandwidth slow down depending upon how they are clustering together the multiple die read/writes ( or not ) . Apple is already running a high rate of die access concurrency into a single stack so won't raise power dissipation issues of the stack much with ECC date traffic weaved in.


Apple isn't using standard , off-the-shelf LPDDR. The basic materials are the same (so don't need different RAM dies) , but how they are stacked and connected isn't the "norm". Especially, those chasing higher capacity at lower costs (via more banks and a fair amount of non-concurrent die access. ) It is different than the norm but Apple asks just one or two makers to make very high volume of it ( basically the whole line up using the same basic building blocks with some minor differences on packaging for the Pro/Max to save more space. Two side-by-side die stacks in those packages so easier to pack them in around the sides than one-stack-per-package plain M. ) .




Actually not traditional ECC. DDR5 still can be run without the link protection to/from the package. There is explicitly optional ECC DIMMs to do both. Still such a think as "DDR5 " and "DDR5 ECC" distinction.

As for AS Mac Pro ... not really likely. DDR5 isn't oriented toward doing the "poor man's HBM" solution that Apple is hooked into. It is a wider bus so not going to as compactly fit a relatively high number of DDR5 channels into a small space. Not impossible to do, but harder to do than with the narrower LPDDR5.

There are really good reasons why don't see high end GPUs using DDR5. The number of cores making the number of pragmatically more random requests means more reasonable size paths/channels is better than fewer wider paths.




Seriously not the practice that Apple has engaged in so far. They use the same blocks of 4 channel LPPDR memory controller and just replicated it into bigger aggregations up the whole food chain. From A-seires up to Max.
The overhead of doing it is spread over all the implementations.

It would be surprising if it went onto the primary die though. Intel has done lots of work to weave Optane DIMMS into the same space are DDR DIMMs. Mismtached memory types would leak into the operating system kernel infrastructure also.

Apple doing something "Simple" and "cost effective bill of materials" would be doing homogenous memory.

Regarding LPDDR5/DDR5 ECC..

In my original post #108, substitue "built-in ECC" with "on-die ECC" and "traditional sense of ECC" with "side-band ECC". See if this help you understand my post better.

Definitions of "on-die ECC" and "side-band ECC" in [0].

[0] different kinds of ECC

***

Regarding AS Mac Pro with DDR5 DIMMs..

I have my thought how Apple could implement it. I also love to let people's imagination fly, and hear what they think (including nay sayers) first.

***

To steer the discussion of this thread a bit regarding the next Mac Pro aka AS Mac Pro...

Personally I would consider AS Mac Pro a flop without meeting the following criteria:
  1. PCIe slots, and more Cheese Grater like
  2. user expandable GPU capability after purchase
  3. huge memory capable, and user expandable capacity after purchase
  4. possibly user expandable CPU capability after purchase
Mac Studio proves itself a true successor to Trashcan. So people can settle with #1 being highly probable.

My thought on #2 is Apple dGPU. On #3 is DDR5 DIMMs. On #4 is socketed Apple silicon SoC though perhaps quite unlikely, hence lowest rank.
 

m1maverick

macrumors 65816
Nov 22, 2020
1,316
1,238
Um you CLEARLY do not understand how law works. Successful suit sets precedents. So when you lose for planned obsolesces once, you likely will lose when youve done it again. Also, the misrepresentation is that they promised, when they announced the M1, that they would keep intel around for many years to come. If they kill it 1 year after the last intel machine is sold, that will be a material misrepresentation.
So the basis of your lawsuit would be "planned obsolescence"?
 

mattspace

macrumors 68040
Jun 5, 2013
3,179
2,879
Australia
To steer the discussion of this thread a bit regarding the next Mac Pro aka AS Mac Pro...

Personally I would consider AS Mac Pro a flop without meeting the following criteria:
  1. PCIe slots, and more Cheese Grater like
  2. user expandable GPU capability after purchase
2. User Upgradable, off-the-shelf, *display* GPU.
 

kvic

macrumors 6502a
Sep 10, 2015
516
459
2. User Upgradable, off-the-shelf, *display* GPU.

Allowing AMD GPUs will be an embarrassment for Tim's Apple.

According to him, Apple uses 3rd party parts if Apple doesn't produce or cannot do it better. Only if Apple can do it better, then that part will be taken in house. Apple has been designing and using its own GPU part for many years. Now also cover laptops, desktops. Soon will go into workstations.

Allowing AMD GPUs indicates Apple admits its own GPU cannot compete among the best in the world. I would think in the medium term, 3-5 years, Apple is going to compete head on against AMD/Nvidia.

People will have to wait for a couple of years if 3rd party GPUs will be allowed.
 

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
And yet Apple got rid of the Mac Pro 2013 on this iteration. It is definitely an outlier. Apple left it on the "for sale" list for 6 years. Adding another 6 to that was likely going to get into trouble. The Mini 2018 being extended into 2022 is also likley to hit some later blow back. Decent chance they are 1.5 years on countdown clock even though kept selling them ( use M1 Mini intro as 'start' date for countdown.)

However, If look at rest of line up though pretty much anything pre-2017 has be axed. They are leaning more aggressively into no cutting any slack for something that is about to go onto the Vintage/Obsolete list. Looking more like if something could go onto Vintage during macOS 13 Ventura's first year of deployment, then it got axed/'Steved' . Maybe they ease up after gotten rid of everything not T2 based . By next iteration , anything not T2 is likely gone.
[ iMac Pro "end of sale" in 2021 and if shave another year off ; 2025. Just 3 more. ]


Sometimes Apple tries to kick the can down the road. Mac Pro 2010 hitting coverage retirement before the Mac Pro 2012 did. That was driven more by their inaction than on the 2012 hitting above some "don't retire me benchmark" performance number. Nor some modularity "get out jail free" card.

it is a definite prune deliberately approach.

It does not in any way excuse getting rid of the 2016 MBP. IT had a T2. It's WAY more powerful than the 2018 MBAir or the 2017MacBook. It's pure evil gouging IMO.
 

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
So the basis of your lawsuit would be "planned obsolescence"?
I gave 3 basis. Planned obsolescence. Misrepresentation of products. Antitrust. The first 2, apple has lost lawsuits on similar bases. The third, they are currently being investigated and/or charged with by EU/Japan/US.
 

m1maverick

macrumors 65816
Nov 22, 2020
1,316
1,238
I gave 3 basis. Planned obsolescence. Misrepresentation of products. Antitrust. The first 2, apple has lost lawsuits on similar bases. The third, they are currently being investigated and/or charged with by EU/Japan/US.
One cannot file a lawsuit of "planned obsolescence" or "misrepresentation" in and of it self. If one wanted to file such a lawsuit one would have to show how they were injured be either (or both) of them. Merely stating "planned obsolescence" or misrepresentation is insufficient.
 
  • Like
Reactions: JMacHack and mode11

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
One cannot file a lawsuit of "planned obsolescence" or "misrepresentation" in and of it self. If one wanted to file such a lawsuit one would have to show how they were injured be either (or both) of them. Merely stating "planned obsolescence" or misrepresentation is insufficient.
Misrepresenting goods sold has numerous state and federal statutes that would apply. Part of the FTC's purpose is to prevent misrepresenting products. Thats why there were all those suits in california over diagonal monitor measurements and hard drive capacity size. Not to mention benefits of drugs etc. So, youre wrong.

https://www.shouselaw.com/ca/personal-injury/intentional-misrepresentation/


You cannot misrepresent products or their material features to the public. Planned obsolescence really goes back to misrepresentation and reasonable expectations of support.

Antitrust goes to market abuses. Not clear if those will pan out with the App Store, but basically, any time you have some monopoly and you engage in predatory practices and market abuses, which planned obsolescence can count towards and there is a potential for liability.
 

goMac

Contributor
Apr 15, 2004
7,662
1,694
2. User Upgradable, off-the-shelf, *display* GPU.
I think this WWDC is the WWDC the third party GPU dream died.

No GPU driver changes. No eGPU support. No return of the AMD drivers.

Apple could do first party GPUs but third party cards don't see to... be in the cards.
 

goMac

Contributor
Apr 15, 2004
7,662
1,694
It does not in any way excuse getting rid of the 2016 MBP. IT had a T2. It's WAY more powerful than the 2018 MBAir or the 2017MacBook. It's pure evil gouging IMO.
2016 had T1, not T2. Which was very different.

Not sure how much it matters though. T2 never came to Intel iMacs, and some of those are still supported.
 

m1maverick

macrumors 65816
Nov 22, 2020
1,316
1,238
Misrepresenting goods sold has numerous state and federal statutes that would apply. Part of the FTC's purpose is to prevent misrepresenting products. Thats why there were all those suits in california over diagonal monitor measurements and hard drive capacity size. Not to mention benefits of drugs etc. So, youre wrong.

https://www.shouselaw.com/ca/personal-injury/intentional-misrepresentation/


You cannot misrepresent products or their material features to the public. Planned obsolescence really goes back to misrepresentation and reasonable expectations of support.

Antitrust goes to market abuses. Not clear if those will pan out with the App Store, but basically, any time you have some monopoly and you engage in predatory practices and market abuses, which planned obsolescence can count towards and there is a potential for liability.
What you have failed to provide, despite repeated requests to do so, is what misrepresentation Apple made that would form the basis of the lawsuit you stated you'd join.
 

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
2016 had T1, not T2. Which was very different.

Not sure how much it matters though. T2 never came to Intel iMacs, and some of those are still supported.
Thanks for the correction, but yea, it seems more punitive to me, than based on substance.
 

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
What you have failed to provide, despite repeated requests to do so, is what misrepresentation Apple made that would form the basis of the lawsuit you stated you'd join.
No, what you have failed to comprehend is during the announcement of the M1 machines they said they would support intel for many years to come. Further, there is a reasonable expectation of support baked into the consumer protection laws.
 

m1maverick

macrumors 65816
Nov 22, 2020
1,316
1,238
No, what you have failed to comprehend is during the announcement of the M1 machines they said they would support intel for many years to come. Further, there is a reasonable expectation of support baked into the consumer protection laws.
Ah, now you're finally starting to leak the requested information. Can you provide the statement you're referring to? The one upon which would form the basis of the lawsuit?
 

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
Ah, now you're finally starting to leak the requested information. Can you provide the statement you're referring to? The one upon which would form the basis of the lawsuit?
Yea, go to the M1 release keynote, watch through it, they say they will support the intel for years. Also, it's a reasonable expectation, so even if they didnt say what they did, it would still be a cause of action under the law I cited.
 

deconstruct60

macrumors G5
Mar 10, 2009
12,309
3,902
2016 had T1, not T2. Which was very different.

Not sure how much it matters though. T2 never came to Intel iMacs, and some of those are still supported.

The iMac 5k did get T2 in 2020 . But , yes there is a somewhat quirky 2019 model that doesn’t that I had forgot about .



The 2019 iMac would be the only corner case though if it makes it past next year. It would go around same time the iMac Pro would go ( if not quicker. ). The 2017 iMac is likely dead next iteration though ( 9+ 4—> 2023 anyway . )

if Apple blows away kext ( kernel extensions ) next year , then it would not be surprising if that non T2 2018 runs into problems and falls early. (i.e. the boot model gets Apple silicon centric and need T2 to get going. ) .
 

kvic

macrumors 6502a
Sep 10, 2015
516
459
I think this WWDC is the WWDC the third party GPU dream died.

No GPU driver changes. No eGPU support. No return of the AMD drivers.

Apple could do first party GPUs but third party cards don't see to... be in the cards.

AMD GPUs will have to wait for a couple of years before given the chance (if any) on Apple silicon Mac. Apple seems confident enough they can compete against Nvidia/AMD in the coming few years for both 3D and ML performance.

Mac Pro 2019 is 6 to 9 months before discontinue (if it does happen). In 3 months time, AMD will release its RX7000 series. AMD has some very performant high-end parts in RX7000. So it's still possible Mac Pro 2019 will receive high-end RX7000 GPUs if Apple's most valued customers ask for it. Also, even if MacOS 13 or 14 are the last OS, the machines are still supported up to around 2027.
 

m1maverick

macrumors 65816
Nov 22, 2020
1,316
1,238
Yea, go to the M1 release keynote, watch through it, they say they will support the intel for years. Also, it's a reasonable expectation, so even if they didnt say what they did, it would still be a cause of action under the law I cited.
As this would be your lawsuit you would need to provide the court the exact statement(s) you would use as the basis of your lawsuit. The court needs to know specifically what misrepresentation Apple engaged in. Think of me as the court and you're filing the lawsuit. I, as the court, need you to tell me.
 

goMac

Contributor
Apr 15, 2004
7,662
1,694
Yea, go to the M1 release keynote, watch through it, they say they will support the intel for years. Also, it's a reasonable expectation, so even if they didnt say what they did, it would still be a cause of action under the law I cited.

By the time Ventura ships, it will have been two years since the first Intel Macs launched. Or "a few years." And Ventura will still supports some Intel Macs.

There isn't really a case here. Apple has supported Intel Macs for a "few" years. They never were specific about which ones. And they haven't even stopped supporting Intel Macs yet.

Heck, Apple could say that "support" refers to continuing security updates.
 
  • Like
Reactions: m1maverick

ZombiePhysicist

macrumors 68030
May 22, 2014
2,795
2,700
By the time Ventura ships, it will have been two years since the first Intel Macs launched. Or "a few years." And Ventura will still supports some Intel Macs.

There isn't really a case here. Apple has supported Intel Macs for a "few" years. They never were specific about which ones. And they haven't even stopped supporting Intel Macs yet.

Heck, Apple could say that "support" refers to continuing security updates.


Yea apple has been for less and lost. By your way of counting, a few years could have started a decade ago when intel chips first gained support. A reasonable count is how many years of support after the last intel machines are still sold. They still sell intel machines in 2022 and by your count they could stop selling intel machines in 2023, but stop support for them in 2022. That won’t work out well for them in court.
 

Boil

macrumors 68040
Oct 23, 2018
3,283
2,899
Stargate Command
I think this WWDC is the WWDC the third party GPU dream died.

No GPU driver changes. No eGPU support. No return of the AMD drivers.

Apple could do first party GPUs but third party cards don't see to... be in the cards.

6jh53y.jpg
 
  • Haha
Reactions: ZombiePhysicist

Joe The Dragon

macrumors 65816
Jul 26, 2006
1,025
474
I think this WWDC is the WWDC the third party GPU dream died.

No GPU driver changes. No eGPU support. No return of the AMD drivers.

Apple could do first party GPUs but third party cards don't see to... be in the cards.
and what is the plan for the high end workstation that needs Multi screen video? useing the TB bus can kill IO for non video needs.
People who may need multi screen more basic video out and don't want USB based video (did apple lock that out yet?)

high end workstation with big GPU compute needs?

Right now apple sells the mac pro with an max gpu config of twin 64GB Video cards what is the plan to deal with work loads that really need an big ram pool just for video + an big system ram pool?
 

mattspace

macrumors 68040
Jun 5, 2013
3,179
2,879
Australia
and what is the plan for the high end workstation that needs Multi screen video? useing the TB bus can kill IO for non video needs.
People who may need multi screen more basic video out and don't want USB based video (did apple lock that out yet?)

high end workstation with big GPU compute needs?

Right now apple sells the mac pro with an max gpu config of twin 64GB Video cards what is the plan to deal with work loads that really need an big ram pool just for video + an big system ram pool?

Yup, or if you wanted to put several of the AMD pro GPUs with 6x mDP ports, and drive a video wall etc. Thunderbolt, and iGPU isn't up to that task.

My suspicion is we're not seeing AMD GPU drivers, because there's a timing tension between Apple and AMD over when AMD makes the reference driver for ARM. This isn't "Apple refusing to support it", it's "Apple can't get a supplier to commit to them".

AMD was selling every single GPU they could manufacture. What is their motivation to spend time and money on an ARM reference driver, when they're an x86-64 CPU manufacturer, there's no significant non-Apple, non-server ARM computing ecosystem, fabrication world wide is already constrained, and they're selling everything they already make?

Just like Apple had to pay the Blender foundation to support Metal, I suspect there's a significant negotiation over how much money Apple is going to give to AMD, for AMD to write a driver. AMD knows they have Apple over a barrel, because it doesn't matter what Apple does with integrated graphics, they can't launch a Mac Pro without a user-replaceable display GPU that is significantly more performant for the price than AMD's.

We've done "Pro workstation without user-replaceable display graphics" - it isn't a thing. It will never be a thing. Stop trying to make Fetch happen.

WHEN Apple has an ARM machine with slots available to buy, we'll see an ARM AMD driver story, and then all the machines released AFTER that point will support eGPU and there's your upgrade driver. The folks who bought M(x) machines before then, bought version 1 Apple products.

As for why there was nothing about it at WWDC - the most likely answer is it will be implemented transparently, and 3rd party developers won't have direct access to the hardware from their apps (which was the fundamental conflict that saw Nvidia abandon the Mac). They'll only be able to call to the higher level graphics infrastructure, which will delegate iGPU or dGPU - so there's no developer story to talk about.
 
Last edited:
  • Like
Reactions: ZombiePhysicist
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.