MP 7,1 Mac Pro 8.1 - WHAT IF...

[prefuse07]

And that's why the Apologist comment, @iPadified, because you still continue to defend apple and their mistakes or poor decisions at all costs...

I mean, hell, you literally just compared a 13 year old computer with one released more recently. Of course the 5,1 is no longer high-end, but it certainly was when it was released, and the fact that it has been able to continue to live-on (thanks to enthusiasts) throughout these 13 years speaks more volume to it being high-end than anything else.

 

[exoticSpice]

And that's why the Apologist comment, @iPadified, because you still continue to defend apple and their mistakes or poor decisions at all costs...

I mean, hell, you literally just compared a 13 year old computer with one released more recently. Of course the 5,1 is no longer high-end, but it certainly was when it was released, and the fact that it has been able to continue to live-on (thanks to enthusiasts) throughout these 13 years speaks more volume to it being high-end than anything else.

Not really. My 2010 PC still works with upgrades as well. It cost me only $500.

Making a PC work after a decade is common in the PC world. The Xbox one/PS4 is still supported by MS and Sony and they came out in 2013. So PC and last gen Consoles are long lasting and are actually supported by OEMs even today. Case in point the PS4 can still play the new God of War.

Apple dropped support for the 2013 Mac Pro because Apple is more greeedy and vile than other companies.


If you are in the workstation market what you mainly want is a lot of I/O and you would regularly upgrade your CPU and GPU. There is nothing high end about a computer to last long.
But it's only high end if it has lots of PCIe slots, Nvidia GPUs and AMD CPUs now in the current era.

TLDR: long lasting should be standard and not only for high end computers. Most computers do last long but what makes them high end is at the time what hardware and features they offered that made cost so much.

 

[ZombiePhysicist]

7,1 was a reaction to tcMP but careful what you which for: entry prices got a bit steep for most. The 7,1 is much better than the 6,1 but also the 5,1. Large power supply and sufficient power to the GPU. After some digging, the three PCIe appeared to share 300W in the 5,1. Not even one 4090...Ergo, the 5,1 is not a Mac Pro by the 7,1 standard but at best a mid range tower which explains the pricing of them.

It does not change the fact that non-modular computers have taken over the world and PC towers, which 7,1 belongs to, are squeezed by sufficiently high performing non-modular computers and cloud computing. The world is not as it was during the 1995-2010 where towers had their peak.

Please stick to rational arguments. "Apple apologist" is not an argument.

Thanks for conceding and running away to your new goal post. Because so many people here were arguing that Mac Pro is not a niche. 🙄 Nearly the last bastion of a lost argument... oh but here is the last one, dictating to me, after shifting goal posts, about rational arguments. 🙄

You are are dismissed.

 

[innerproduct]

So this is the workstation chipset that we would have had if there was a new mac pro using intel. That coupled with 7900 duo cards for an up to date 2019 mp. I wonder what the diff will be in regards to apple silicon perf across the board. Single thread probably faster. But else? https://wccftech.com/supermicro-w79...sapphire-rapids-ws-cpu-support-16-ddr5-dimms/

 

[Mac3Duser]

if only the new mac pro could have sapphire rapids xeon...^^

 

[Kimmo]

So this is the workstation chipset that we would have had if there was a new mac pro using intel. That coupled with 7900 duo cards for an up to date 2019 mp. I wonder what the diff will be in regards to apple silicon perf across the board. Single thread probably faster. But else? https://wccftech.com/supermicro-w79...sapphire-rapids-ws-cpu-support-16-ddr5-dimms/

Interesting.

Here's a link to another article with additional specs:

Intel Sapphire Rapids-WS Xeon CPUs & W790 'Fishhawk Falls' Platform To Be Unveiled In February, Launch In April 2023

Intel is planning to launch its Xeon Workstation 'Sapphire Rapids-WS' CPUs & W790 Fishhawk Falls platform in February & launch in April 2023.

wccftech.com

 

[iPadified]

You're missing the point. My reference to PIXLAS was just to demonstrate that the AUX power 'limitation' was arbitrary, and related to GPU needs of the era. There was loads of spare capacity, and it would have been trivial for Apple to provide it had they revised the 5,1 - e.g. by using thicker traces on the motherboard.

Sure they could but there is misconception that 5,1 was good computer without flaws and these flaws was an Apple decision.

And that's why the Apologist comment, @iPadified, because you still continue to defend apple and their mistakes or poor decisions at all costs...

I mean, hell, you literally just compared a 13 year old computer with one released more recently. Of course the 5,1 is no longer high-end, but it certainly was when it was released, and the fact that it has been able to continue to live-on (thanks to enthusiasts) throughout these 13 years speaks more volume to it being high-end than anything else.

My point was that the 5,1 was not high end in the GPU department and lacked ability to support (future) high end GPU due to poor power delivery. In my opinion a computer you need to mod to do what is supported to do is built wrong from the factory. Is something wrong with that criticism?

Did I not say earlier the tcMP fault was the name and the lack of a co-existing modular Mac Pro such as 5,1? Something wrong with that criticism? Next fault was that Apple did not update the tcMP but I think the faults was partly the lack of Intel and AMD offering with the correct power brackets. They were awfully slow to move from 28nm. Something wrong with that critisism? No wonder Apple went ASi. When we are at it, where is the support for Nvidia in Mac Pro? The Mac Pro is nearly dead in the 3D space and likely scientific compute space without Nvidia support. Something wrong with that criticism?

I think Apple has made many bad decisions in the Mac Pro space but offering non-modular desktop computers was not one of them.

Apparently criticizing a Mac Pro is Apple apologizing. Strange selective idea. Please explain.

 

[mattspace]

Next fault was that Apple did not update the tcMP but I think the faults was partly the lack of Intel and AMD offering with the correct power brackets. They were awfully slow to move from 28nm. Something wrong with that critisism?

Yes, something is wrong with that criticism. It's effectively like claiming gravity is at fault for the consequences of a decision to jump off a cliff.

Improvements in graphical performance post 2013 required more power - that is the overriding reality of the world. Apple, as usual, built a computer for a fantasy scenario that no one else was going to subscribe to, in which everyone just cruised on process shrinks to get any (if any) performance improvements.

Apple's decision to build a computer that could be thermally constrained, showed a gross failure of both imagination, and put paid to the myth that the people who work there are in any reasonable definition of the term "smart".

 

[maikerukun]

So the best comment here is that nobody knows 😂. We can debate until we are blue in the face but this was a fantastical question meant to spark imagination and discuss what in a perfect world might be happening.

I still think they’re making a brand new top secret chip they’ve been working on specifically and only for the 8.1 and that it’s going to somehow integrate both technologies.

I also expect it to be 2x faster than my maxed out 2019 in 3D rendering of all forms.

What is YOUR ideal reveal for the 8.1? I’m super excited for this thing and hoping it’s amazing because if not, I won’t be leaving my 7.1. Back in my day I had to walk 5 miles with the 7.1 on my back just to get solid tender times. You youngsters with your do it all chips don’t know what it means to almost burn down your house with power surges from juicing up your system to kingdom come. You’re all about efficiency these days but I want a monster to destroy the city while I play with the buildings like legos 😂😂😂

 

[Kimmo]

What is YOUR ideal reveal for the 8.1?

Soon, very soon!

 

[iPadified]

Yes, something is wrong with that criticism. It's effectively like claiming gravity is at fault for the consequences of a decision to jump off a cliff.

Improvements in graphical performance post 2013 required more power - that is the overriding reality of the world. Apple, as usual, built a computer for a fantasy scenario that no one else was going to subscribe to, in which everyone just cruised on process shrinks to get any (if any) performance improvements.

Apple's decision to build a computer that could be thermally constrained, showed a gross failure of both imagination, and put paid to the myth that the people who work there are in any reasonable definition of the term "smart".

The ability to heat dissipate determines the performance of any computer. During the Intel /AMD era, things got worse for every generation because of lack of innovation. There was an increasing discrepancy between what many people wanted (smaller and silent computers) and the chips development did not support that. The result was ASi and we are back on track in the thread.

So the question still remains, will Apple use any performance/watt advantage to build a competitive workstation. I see lots of clouds and it is not the hardware but the lack of software support and a very tiny community.

PS. There are some of you that should get over the tc2013 as you seem to take that machine as a personal insult. It was not, it was just not a machine for you. DS.

 

[Mac3Duser]

I love the tcMP.
In my opinion, there are 2 ways for the 8,1:
- 1. the ASI mac Pro is like a super Mac Studio, so we have a better design for air flow, maybe one M2ultra or two M2ultra (as we had bi-xeon mac pro), and upgrades possible by the consumer : SSD M2, pci-e cards but not graphic cards.
- 2. totally new ASI chip with a lot of pci-e lanes, MPX modules with AMD graphic cards

 

[mattspace]

The ability to heat dissipate determines the performance of any computer. During the Intel /AMD era, things got worse for every generation because of lack of innovation.

Nonsense. Look at how radical performance improvements were in the GPU space. There was no "lack of innovation" - that is a myth invented to wallpaper over the truth, that Apple was off in the weeds trying to force a future of performance constraint to suit decorative enclosures that no one else wanted.

Everyone else innovated on their cooling game - and cooling is THE big innovation across a range of technologies. Look at the fastest, most extreme performance cars, what do they share in common - massive cooling capacity. Bugatti Chiron - 10 radiators, with 2 auxiliaries for the main one alone.

Apple kept producing thermally limited design after thermally limited design that gambled on unknown futures for component heat production, and then had their army of flying media monkeys whine in the press about component producers not "innovating".

It's a Downfall meme clip.

There was an increasing discrepancy between what many people wanted (smaller and silent computers) and the chips development did not support that. The result was ASi and we are back on track in the thread.

Literally the only area of computing that had any substantial growth in the last 5-10 years is high performance 3d visualisation / gaming desktops.

Funny that, it's exactly the same systems that use the big graphics cards which Apple consistently fails to build systems to host. It's almost as if Apple's finger is jammed up its nether regions, instead of being on the pulse of what everyone else is doing.

And no, Apple sales are not proof of it's visionary strategy, because without an alternative vendor of macOS hardware / paradigms, sales of Apple devices is not falsifiable evidence in support of people actively choosing Apple's choices.

So the question still remains, will Apple use any performance/watt advantage to build a competitive workstation. I see lots of clouds and it is not the hardware but the lack of software support and a very tiny community.

Firstly, Performance / watt advantages will disappear as performance increases. AS is good at low-hanging fruit - minimum power draw, but it offers practically no benefit when it's actually doing work & at maximum power draw.

Secondly, power draw is irrelevant in a desktop system. Professionals who can afford "professional" computers can afford "professional" power bills. Power is tax deductible, and just a (minor) cost of business, like the computer itself.

PS. There are some of you that should get over the tc2013 as you seem to take that machine as a personal insult. It was not, it was just not a machine for you. DS.

No, it was a bad design. A textbook bad design - poorly realised, and motivated by hubris and butthurt. It, and everyone involved in it, should be pilloried until their only remaining professional epitaph is "had a hand in the single worst computer ever sold". It was not "a good design for some consumers", because none of the (alleged) advantages it offered, were uniquely tied to the sacrifices it made.

 

[Yebubbleman]

As you all know, Apple will have to pry my Intel based Mac Pro 7.1 out of my cold dead fingers, because I won't let go. I won't leave behind it's 28 cores, it's 2 w6800x Duos, it's ridiculous GPU power while rendering in Octane and Unreal...and yeah, I'll pick up whatever is coming next with the AS Mac Pro 8.1, and it'll be something I use ALONGSIDE my 7.1

As someone that owns way too many computers, I won't debate the desire to own multiple generations of Mac Pro, especially when one is the pinnacle of the Intel era (which already entails a fair amount of computing workloads that are simply not available on an Apple Silicon Mac and likely never will be).

That all being said, you do know that this won't be "MacPro8,1", right? There literally won't be such a Mac now that Apple has moved towards a general "MacX,Y" naming scheme (starting with the Mac Studio, but continuing on with the M2 MacBook Air and the M2 13-inch MacBook Pro).

Unless...

And I never thought about this until now, because it seems crazy, and because nobody has ever mentioned it with all the fuss around AS, but...

What if...

Apple has been working on an entirely separate chip very specifically designed just for the current Mac Pro? What if they realized long ago that the M family wouldn't scale and be able to outperform the monster they designed with the current 7.1 Mac Pro?

Uh...I'm pretty sure this is the dominant theory around the campfire. Far from "nobody has ever mentioned it".

In fact, it would only be logical for the Apple Silicon Mac Pro to keep in with the Intel Mac Pro in that it does not use the same class of processor as the rest of the lineup.

Furthermore, it's very clear that Apple doesn't intend to keep its desktops updated anywhere near as often as its laptops. As it stands currently, the only Mac product line to even see annual changes is the MacBook Pro. There was no new MacBook Air in 2021 and, with the exception of 2019-2021, changes to the iMac line were only happening once every other year (though, you could argue that the only real exception to this was the 2020 27-inch iMac, given that the Mac Studio was its replacement whereas the 2021 M1 iMac was not). The Mac Pro won't be an exception at all. We're going to see 3-6 years in between Mac Pro refreshes AS THE NORM.

But, much in the same way that there are people still chugging away on 2010-12 Mac Pros some 10-12 years later, the first Apple Silicon Mac Pros will definitely be apt to remain supported for the same 9 years that the 2012 and 2013 Mac Pros were. Hell, you have A9 iPhones having been dropped from iOS 16 support, yet iPadOS 16 definitely still supports the first generation 12.9" iPad Pro's A9X. Why? Because relative to A9, that particular SoC was (and still sort of is) a beast.

What if, they keep everything about the current awesome design that is the 7,1 Mac Pro, and just cut out the intel?

If you're talking dreamland, sure, that would be cool.

Back in reality, that ain't gonna happen. Apple has been very clear, since day one of them announcing the switch to Apple Silicon to begin with, that the RAM and GPUs will be integrated with the SoC. So, while Apple could toss you a bone and give you a PCIe x16 slot with a ton of bandwidth, you will never be able to put in an AMD or NVIDIA graphics card into that slot. Similarly, they might keep the MPX module convention around (seeing as MPX isn't limited to graphics cards), but you won't see graphics cards occupying those slots.

So, looking at the current Mac Pro, you won't see those RAM slots, and you won't see GPUs occupying those slots.

I don't see why a socketed SoC isn't possible. But get ready to cry at the prices of those replacement chips.

64 cores, w7800x duos, still 100% customizable in every way, but running on a brand new custom designed, and for the Mac Pro only CPU and motherboard...

You might see 64 graphics cores on the SoC. You will not see a W7800X grace that box. Again, third party graphics are not a thing in Apple Silicon Mac land.

I'll admit, I'm 5 glasses of moscato into my home girls dinner party right now, but it came to mind and so I had to hop on here and just vomit my nonsense all over the message board and see how ya'll responded LOL.

But seriously though..what if...

Not a bad topic to engage with intoxicated. I do it with my buddy all the time!

If those PCIe slots support the upcoming AMD 7000 series then this could be something interesting

They won't. You can wake up from that particular dream right now.

Here is an important question, and one that goes to the OPs point of why not a completely new chip. Pro gear needs ECC. If you have 256GB youre going to have a bit flip more than once a day. That is *not* acceptable in a pro market device. None of the apple chips have ECC as far as I know. As such, might require a completely new chip that does have it.

We seemed to survive the PowerPC era of Power Macs just fine without ECC. Not saying that we won't want it for the Apple Silicon era of high-end Mac computing, but is it as much of a need now that we're not in Intel land and given how efficient Apple's unified memory architecture is? Not saying it wouldn't help, but is it as much of a must?

I can't believe we're nearly in 2023 and I'm still saying this, but it really remains to be seen what Apple will do with AS and 3rd-party GPU support.

We're nearly in 2023 and you still haven't watched the video from WWDC 2020 where Apple completely debunked this: https://developer.apple.com/videos/play/wwdc2020/10686/

Go to 1:00 in. There will be no discrete GPU in the way that we had on 15, 16, and 17 inch MacBook Pros (nor any of the PowerBook G4s that preceded them), Mac Pros, nor iMacs. The GPUs will always be on the SoC. End of story. In turn, while you can certainly make the Intel version of macOS run with third party GPUs (given that's all there were) and add/hack on drivers for ones that aren't already there, YOU CANNOT DO THIS WITH THE APPLE SILICON VERSION OF macOS!

A real dream would be to chuck an MPX module into my 7,1 that's whatever M2 pro thing they come out with, but then I guess that would negate the need for any of the RAM, certainly the Xeon and maybe the GPU's as well, at which point… what, my PCIe m.2 ssd's and a headphone port are what I'm adding? 😬

You know that PCIe slots are useful for more than graphics cards, right? In fact, upgradeable graphics, while a big sticking point for 2013 Mac Pro and iMac Pro (2017) complainers, is far from the only reason why those customers still wanted PCIe slots. There are plenty of high-end video and audio capture cards that you're not simply slapping into a Thunderbolt 3 expansion chassis and calling a day with. THAT'S who the Mac Pro is for.

It's normal. What is the "Mac pro" philosophy? It is very simple. It's having a pc tower and being able to change the components as you want, starting from a not too expensive base (more or less 2500 dollars).

There hasn't been a Mac Pro with a base cost of $2500 in at least 9 years.

If the ASI mac pro base model is too expensive and no one can add its ram , ssd and graphics card, it will fail because the mac studio will be fine for most pro use.

The Mac Pro will be for those that (a) need whatever faster SoC Apple has planned. If that SoC is socketed, all the better. I have no reason to believe it is, but also no reason to believe that it wouldn't be a good idea for Apple to do that (and sell aftermarket upgrades) as this is the only way users can upgrade RAM or graphics after the fact.

SSDs will probably be handled exactly the same way they are on the current Mac Pro; which is to say that the modules just contain the NAND while the SSD's controller comes from the SoC which will be cryptographically paired to the modules. Remove the modules and the data on them is useless. Upgrade the modules by doing a DFU restore of the Mac Pro (thereby rendering whatever data was on the old modules as good as gone). I'm sure Apple will throw in those same SATA ports and any other drive expansion that was present on the current Mac Pro. Nothing about Apple Silicon precludes those from existing.

Users will buy the Mac Pro over the Mac Studio for the same reason why they didn't buy the 27-inch iMac instead; they will want PCIe expansion and way higher-end options than otherwise exist. Toward that end, you're likely going to see SoCs on the high end of the Mac Pro that won't ever be in a Mac Studio.

the 7.1 base model has a great case but it was way too expensive (reminder: at first it was sold with 256GB and 32 GB of ram).

While I don't disagree that the current Mac Pro was/is way too expensive, you have to take into account that there are fewer people that NEED this kind of Mac than was the case during the 1,1-5,1 days, or even the Power Mac G5 and Power Mac G4 days. You only really need a Mac Pro if you need (a) internal expansion and/or (b) CPU/GPU horsepower needs that exceed what you'd otherwise have in a Mac Studio or 27-inch iMac.

A way of saying to 90% of 5.1 community: "this computer is no longer for you".

Again, unless you need internal expansion or seriously powerful CPU/GPU horsepower (and mind you, the M1 Max and M1 Ultra DO outperform most Mac Pro configurations in this arena), you do not need a Mac Pro. The message isn't that "This computer is no longer for you", but rather "you no longer need this class of machine because now the lesser models are adequately performant".

Apple should regain the 1.1-5.1 community with the mac pro ASI. It wouldn't be a useless step back.
users expect only one thing: to be able to put the cards they want in a large format case.

Most cards that you'd throw into a current Mac Pro are high-enough end to justify the cost of the machine. You're not as likely to be as casual about the PCIe cards you throw into a Mac Pro as you would be a PC you built yourself, any Power Mac, or any 2006-2012 Mac Pro tower because cards for Mac aren't designed to be casual. Most of them are high end. The days of casual Mac tower usage are over and done with thanks to the 2013-2019 gap wherein the best Xeon Macs we had were the trash can and the iMac Pro.

Yeah, but the iMac Pro, and the 7,1 were never intended to exist as complimentary products.

The iMac Pro was a stillborn remnant of a dead product strategy - replacing the Mac Pro as a product, with an iMac as a product. There was no 7,1 in the iMac Pro timeline future.

Uh...the iMac Pro was literally made as a stopgap for the 7,1; designed for folks that begrudgingly switched to standard 27-inch iMacs because the trash can aged without updates. The 7,1 was literally the entire point of the iMac Pro's timeline. If that's what you were trying to say, then you went about it sorta roundabout.

Base model 128-196GB of RAM.

Depends on whether or not the M1 Ultra or some other A15 or A16 based equivalent is the basis or not. If it is, then 64GB will probably be the base model specs. If not, then 128GB as a starting RAM amount makes sense.

As the 7,1 showed - there was never an "iMac Pro without the cost of a 5k retina display" option. The availability of slots was priced higher than the (as Apple described it) "free" display that was removed.

"I want more ram than the maxed studio"
"buy a Mac Pro"
"But I don't want to pay for PCI slots"
"too bad"

The maxed studio has 128GB of RAM. You're paying for a hell of a lot more than just PCIe slots by buying ANY Mac with more than 128GB of RAM in tow. Hell, consider all of the Macs that Apple has produced that have 128GB of RAM as an option. None of them were/are even remotely low-end and that has nothing to do with PCIe slots.

 

[mattspace]

Uh...the iMac Pro was literally made as a stopgap for the 7,1; designed for folks that begrudgingly switched to standard 27-inch iMacs because the trash can aged without updates. The 7,1 was literally the entire point of the iMac Pro's timeline. If that's what you were trying to say, then you went about it sorta roundabout.

You are incorrect in your interpretation of the timeline. The iMac Pro was never designed to be a stopgap, or a fill-in product between anything. It was a completed product, ready for manufacture before the 7,1 Mac Pro was even an idea.

The iMac Pro was created to be the sole high end professional Mac, as a replacement for the Mac Pro as a product. Had Apple not course corrected, the Mac Lineup going forward would have been Macbook/Air, Macbook Pro, Mac Mini, iMac, iMac Pro, and nothing else.

The 7,1 was an emergency panic product (that was barely a napkin sketch when announced in the apology tour) in response to the dawning realisation that the iMac Pro was doubling down on the aspects of the 2013 Mac Pro that the market had rejected - weak fixed graphics, and Thunderbolt for all expansion. That's why it was stillborn, and never received an update (except putting an "X" on the GPU's firmware descriptor).

 

[exoticSpice]

You are incorrect in your interpretation of the timeline. The iMac Pro was never designed to be a stopgap, or a fill-in product between anything. It was a completed product, ready for manufacture before the 7,1 Mac Pro was even an idea.

The iMac Pro was created to be the sole high end professional Mac, as a replacement for the Mac Pro as a product. Had Apple not course corrected, the Mac Lineup going forward would have been Macbook/Air, Macbook Pro, Mac Mini, iMac, iMac Pro, and nothing else.

The 7,1 was an emergency panic product (that was barely a napkin sketch when announced in the apology tour) in response to the dawning realisation that the iMac Pro was doubling down on the aspects of the 2013 Mac Pro that the market had rejected - weak fixed graphics, and Thunderbolt for all expansion. That's why it was stillborn, and never received an update (except putting an "X" on the GPU's firmware descriptor).

Apple's Mac team was brain dead from 2012 - 2019. Every Mac sucked.

Apple's top engineers were made to work on the ARM transition. That's why the Intel Macs in that era sucked so much. The Mac department was running on third rate engineers and some which got into trouble into law later on.

What saved the Mac was Pros Workflows team which Apple created after numerous failures that using silly engineers for the Mac would kill the whole line up.


Really tell me which was Mac good in this 2012-2019 era. The MacBooks had stupid Keyboards and failing screens and ports that did not make sense. These MacBooks were made thin as a cracker and cooling suffered

The iMac was ugly thick bezels when other AIOs were stunning.

The Mac Pro was a stupid trash can.

The only good Mac was the Mac mini but even that suffered for 4 years.

It was not until the 2019 Mac Pro and 2019 16" MBP that Apple got it together. The Pro workflows team designed every Mac from mid 2019 onwards. The Macs are now designed as tools and not as showpieces.

Ok history time over.

Today the M2 Max chips leaked and it's max memory capacity is 96GB which is good for a workstation laptop.

This also means that M2 Ultra will have 192GB RAM and the M2 Extreme will have 384GB of RAM.

 

[iPadified]

Nonsense. Look at how radical performance improvements were in the GPU space. There was no "lack of innovation" - that is a myth invented to wallpaper over the truth, that Apple was off in the weeds trying to force a future of performance constraint to suit decorative enclosures that no one else wanted.

Everyone else innovated on their cooling game - and cooling is THE big innovation across a range of technologies. Look at the fastest, most extreme performance cars, what do they share in common - massive cooling capacity. Bugatti Chiron - 10 radiators, with 2 auxiliaries for the main one alone.

Apple kept producing thermally limited design after thermally limited design that gambled on unknown futures for component heat production, and then had their army of flying media monkeys whine in the press about component producers not "innovating".

It's a Downfall meme clip.




Literally the only area of computing that had any substantial growth in the last 5-10 years is high performance 3d visualisation / gaming desktops.

Funny that, it's exactly the same systems that use the big graphics cards which Apple consistently fails to build systems to host. It's almost as if Apple's finger is jammed up its nether regions, instead of being on the pulse of what everyone else is doing.

And no, Apple sales are not proof of it's visionary strategy, because without an alternative vendor of macOS hardware / paradigms, sales of Apple devices is not falsifiable evidence in support of people actively choosing Apple's choices.



Firstly, Performance / watt advantages will disappear as performance increases. AS is good at low-hanging fruit - minimum power draw, but it offers practically no benefit when it's actually doing work & at maximum power draw.

Secondly, power draw is irrelevant in a desktop system. Professionals who can afford "professional" computers can afford "professional" power bills. Power is tax deductible, and just a (minor) cost of business, like the computer itself.



No, it was a bad design. A textbook bad design - poorly realised, and motivated by hubris and butthurt. It, and everyone involved in it, should be pilloried until their only remaining professional epitaph is "had a hand in the single worst computer ever sold". It was not "a good design for some consumers", because none of the (alleged) advantages it offered, were uniquely tied to the sacrifices it made.

Cooling is one way to improve heat dissipation so it seem we agree on that one. I also think Apple dropped the ball on 3D but that was long before the tcMP.

Why would the performance/watt disappear and what is maximum power draw? What is maximum power draw? Do you have a certain power usage threshold for when a computer is doing work? What would you prefer: a system that draw much power or low power?

Did you have a tcMP on your desk? It was actually very nice. Small, silent at more powerful than a MacBook Pro. Was a modular Mac to compete with HP Z series - (well the high end) no. Did it do the job for what it was intended for - yes. Wrong for you but it did the job for others (just like the iMac Pro and Mac Studio). Why is that difficult to accept?

 

[mattspace]

Why would the performance/watt disappear and what is maximum power draw? What is maximum power draw? Do you have a certain power usage threshold for when a computer is doing work? What would you prefer: a system that draw much power or low power?

Because the "advantage" on power draw of AS is in idle and low-draw situations. In high-workload situations it's not any more power efficient than anything else.

So in other words, an AS laptop can sit at idle for longer, but will run flat just as fast going full strap.

Did you have a tcMP on your desk? It was actually very nice. Small, silent at more powerful than a MacBook Pro. Was a modular Mac to compete with HP Z series - (well the high end) no. Did it do the job for what it was intended for - yes.

It was intended to do the job of a high end Z series. Apple launched it as a "Pro" desktop, with "dual pro video cards". This idea that it was a niche product, really only intended for quiet low demand use, for a subset of potential mac pro customers is a revisionist myth, post-hoc justification for a bad product failing to change a market it was supposed to revolutionise.

Wrong for you but it did the job for others (just like the iMac Pro and Mac Studio). Why is that difficult to accept?

It did no job for anyone that a conventional machine could not have done. To claim it was anything other than a failure is like burning down 70% of your house, and talking about what a victory it is for fewer rooms to vacuum.

 

[exoticSpice]

AMDs 4nm Zen 4 APUs will be more efficient than M1 and M2. Watch AMDs space in laptops.

I don't find Apple Workstations interesting anymore. What can the Mac Pro do that other PCs workstations can't do!


Apple is making macOS worse every year.

 

[innerproduct]

Apple's Mac team was brain dead from 2012 - 2019. Every Mac sucked.

Apple's top engineers were made to work on the ARM transition. That's why the Intel Macs in that era sucked so much. The Mac department was running on third rate engineers and some which got into trouble into law later on.

What saved the Mac was Pros Workflows team which Apple created after numerous failures that using silly engineers for the Mac would kill the whole line up.


Really tell me which was Mac good in this 2012-2019 era. The MacBooks had stupid Keyboards and failing screens and ports that did not make sense. These MacBooks were made thin as a cracker and cooling suffered

The iMac was ugly thick bezels when other AIOs were stunning.

The Mac Pro was a stupid trash can.

The only good Mac was the Mac mini but even that suffered for 4 years.

It was not until the 2019 Mac Pro and 2019 16" MBP that Apple got it together. The Pro workflows team designed every Mac from mid 2019 onwards. The Macs are now designed as tools and not as showpieces.

Ok history time over.

Today the M2 Max chips leaked and it's max memory capacity is 96GB which is good for a workstation laptop.

This also means that M2 Ultra will have 192GB RAM and the M2 Extreme will have 384GB of RAM.

So what if the new Mac Pro simply is what rumours have been suggesting since the beginning: a dual or quad chip M* with some extra PCI lanes for things outside of graphic cards. Like for connectivity, pro tools/audio stuff etc. And that's it?
Then we would be looking at basically Mac Studio Ultra perf *1.2 for the low end and at best MacStudio Ultra*2.4 when it comes to CPU. That would be a great CPU, no doubt, and be good enough to compete with current gen intel and AMD CPUS.
The GPU on the other hand would be kind of terrible for the price point. As we have seen in other threads (and from personal experience) In Redshift (the only currently stable and useable GPU renderer on the Mac outside of Blender/cycles) the Mac studio Ultra renders the very representative benchmark in 6min 30s. IIRC, a single 6900XT managed the same render in less than 5 min.
So, if the new MP using M2 tech would scale better than the M1 Ultra, we might at the very very best see total linear scaling and end up with about sub 4 min for the new M2 Ultra and sub 2 min for the maxed out M2 Extreme.
If the price is right that is definitely useful performance, but about the same as a single, last gen nvidia card. There is oc a selling point that the RAM is unified but well see how much it matters. Seems to me the CPU will be so good that one would seriously consider going for a CPU renderer instead. (since those are still more feature complete and more responsive in general)

Personally, the "extreme" risk to be to costly for what it offers and since I personally only used PCI slots for graphics and some networking before 10Gbe was standard , I will be better served by an updated Mac Studio I think.

My dream Mac Pro would be a system where I could get GPU performance that is in similar range as the PC competition and then be able to upgrade just the GPU part in 3 years to make the system viable at least 6 years.
The more I think of it the more reasonable the Mac Studio Ultra M2 edition with current M1 issues ironed out will be what I am looking for. That solution is still 3X as expensive as the MacPros I used to buy back in the day so even that is stretching the budget and ROI calculations.

 

[eatrains]

Apple's Mac team was brain dead from 2012 - 2019. Every Mac sucked.

Apple's top engineers were made to work on the ARM transition. That's why the Intel Macs in that era sucked so much. The Mac department was running on third rate engineers and some which got into trouble into law later on.

What saved the Mac was Pros Workflows team which Apple created after numerous failures that using silly engineers for the Mac would kill the whole line up.


Really tell me which was Mac good in this 2012-2019 era. The MacBooks had stupid Keyboards and failing screens and ports that did not make sense. These MacBooks were made thin as a cracker and cooling suffered

The iMac was ugly thick bezels when other AIOs were stunning.

The Mac Pro was a stupid trash can.

The only good Mac was the Mac mini but even that suffered for 4 years.

It was not until the 2019 Mac Pro and 2019 16" MBP that Apple got it together. The Pro workflows team designed every Mac from mid 2019 onwards. The Macs are now designed as tools and not as showpieces.

Ok history time over.

Today the M2 Max chips leaked and it's max memory capacity is 96GB which is good for a workstation laptop.

This also means that M2 Ultra will have 192GB RAM and the M2 Extreme will have 384GB of RAM.

The 2012 MacBook Pro set the standard for laptops. Retina screens, SSD, what was bad about it?

 

[enc0re]

To me, Mac Pro means expendable with a rack-mountable option. The sealed, desktop-only market is served perfectly well by the Studio.

Amazon now makes its own processors for its data centers. With all the digital services Apple is offering, maybe they are contemplating the same switch. If Apple became its own Mac Pro customer, the product would instantly go from niche to high volume.

 

[exoticSpice]

The 2012 MacBook Pro set the standard for laptops. Retina screens, SSD, what was bad about it?

The 2012 was an a okay gen.

 

[fuchsdh]

Yes, something is wrong with that criticism. It's effectively like claiming gravity is at fault for the consequences of a decision to jump off a cliff.

Improvements in graphical performance post 2013 required more power - that is the overriding reality of the world. Apple, as usual, built a computer for a fantasy scenario that no one else was going to subscribe to, in which everyone just cruised on process shrinks to get any (if any) performance improvements.

Apple's decision to build a computer that could be thermally constrained, showed a gross failure of both imagination, and put paid to the myth that the people who work there are in any reasonable definition of the term "smart".

Eh, I don't think flexibility for all scenarios is always the overriding goal, nor should it be. Apple's guess in 2013 was a reasonable one; that it was wrong is partly down to their own failures to support it (OpenCL) as well as the outside environment. But really the big issue was their refusal to either admit they were wrong (like they did with the G4 Cube) and pivot immediately, or make meaningful iterative improvements (like the iMac or the iPhone, which rapidly revised the original's weaknesses.) Nothing was stopping Apple from using the same principles of a thermal core and compact enclosure and adjusting it to use a single more powerful GPU. And, as I've said for years, whether or not the tube Mac Pro could have succeeded in an alternate universe where they actually supported the thing is an unknown. I was disappointed with aspects of the 6,1, but I would have been happy to buy a Broadwell model. Instead... they did nothing. Why would anyone buy into a product line they weren't supporting themselves with upgrades or new revisions?

 

[enc0re]

I would go so far as to say that the Studio effectively is a continuation of the tube Mac Pro. They just had to distance it a bit because of the negative reputation.