Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
With 1.5 TB of RAM, they can give it to me without SSD :D :D

Jokes aside, good to see something "moving". However, as much as it is a high-core count CPU, I would expect a higher SP performance for the mMP. Maybe Apple will go for custom versions like they did with the iMP's Xeon Ws or I hope the 8/10/12 core base configuration will run sensibly faster.

What is this, then?

https://browser.geekbench.com/v4/cpu/10247246
 
Last edited:
Have Geekbench results ever actually leaked a real product versus made-up SKUs?
The thing with this entry is that 8280M launched, but is not widely available, to anybody. This could have been only in engineering sample product.

And who uses Eng Samples to run hackintoshes, apart from Apple themsleves?

Also quite ironic. Apple engineers using Geekbench to evaluate performance of professional machine. The worst benchmark on the planet, to evaluate anything with it.
 
  • Like
Reactions: barmann
The thing with this entry is that 8280M launched, but is not widely available, to anybody. This could have been only in engineering sample product.

And who uses Eng Samples to run hackintoshes, apart from Apple themsleves?

Also quite ironic. Apple engineers using Geekbench to evaluate performance of professional machine. The worst benchmark on the planet, to evaluate anything with it.

Eh, I dunno. It serves its use as a synthetic benchmark for back-of-the-napkin "how much more power will I be getting"-type questions. Anyone solely relying on synethics versus actually comparing their workloads is foolish.
 
Eh, I dunno. It serves its use as a synthetic benchmark for back-of-the-napkin "how much more power will I be getting"-type questions. Anyone solely relying on synethics versus actually comparing their workloads is foolish.
Then why ARM CPUs are on the toes of x86-64 CPUs, in Geekbench, and when it comes to anything heavy lifting they fall behind badly, VERY badly?
 
Then why ARM CPUs are on the toes of x86-64 CPUs, in Geekbench, and when it comes to anything heavy lifting they fall behind badly, VERY badly?

Presumably because x86 CPUs have a bunch of optimized software, or because they're comparing differences in loads (sustained versus burst clocks.) I'm not sure what heavy lifting you're referring to so I can't really answer the question.
 
Then why ARM CPUs are on the toes of x86-64 CPUs, in Geekbench, and when it comes to anything heavy lifting they fall behind badly, VERY badly?

It's heat/improper cooling. Benchmarks don't heat the CPU for long enough, so the ARM CPUs look good. Do serious work for longer and the CPU will overheat and start down clocking.

If Apple stuck a cooler or a fan on an iPad, that might fix the problem. But now you have a bunch of idiots thinking Apple has achieved high end laptop CPU or desktop CPU performance with no cooler.

Apple ARM CPUs don't really have a traditional burst mode. But there is an extremely large difference between their bursty speeds and their sustainable speeds. Much more of a difference than you'd see on an Intel CPU.
 
  • Like
Reactions: Nugget
It's heat/improper cooling.
The cMP is the last Apple product to be considered to have "proper cooling".

Oh wait, never mind, this Apple product never throttled due to overheating.
apple-iPod-socks.jpg
 
  • Like
Reactions: apolloa and Biped
https://twitter.com/momomo_us/status/1087909294987653120

Im baffled nobody has seen this ;). Xeon Platinum 8280m on Apple motherboard, and ... Linux platform. Cascade Lake SP.


An "Open Compute" semi-custom server node for Apple's cloud services ..... wouldn't be surprising.
Linux ... yep. Apple making their own server nodes llke Facebook , Google, AWS , Microsoft, and sizable cloud service vendors ) ... yep. Putting those up for sale.... No ( just like Facebook , Google , AWS , Microsoft , etc. are not ).

Apple doing their own 'trusted' ( taper protected ) servers. Sure Google does it.

https://www.anandtech.com/show/13248/hot-chips-2018-google-titan-live-blog-6pm-pt-1am-utc

Pretty sure that Apple doesn't want this story to turn real.

https://www.macrumors.com/2018/10/19/tim-cook-calls-for-bloomberg-hack-story-retraction/

Especially after made a super big deal about how it wasn't true. The "requirement" to deploy servers to China so 'partners' can run cloud services there .... even more so.

The super, ultra deluxe Xeon Platinum chips in a single user workstation isn't particularly likely. intel's price point on those are stratospherically high to the point that they primarily only make sense when can load them up to tons of users (consolidate workload) or have extremely high budge/value data to do.

Single core mark this 45-46K score is worse than most iMac Pro marks in Geekbench.

Linux 14.13 was EOL back in 2017. So someone "hot rodding" in 2019 is on the non bleeding edge uptick path.

Hackintosh is another creditable path.


It is not impossible some dumped macOS version because maybe it has a info associated with it. Or that there isn't even a MacOS for the board ready. I wouldn't put that option more likely than the other two.
 
Apple are the bunch of idiots that are thinking that....??

gomac can add what his observation, but I think that is far more so a reference to the folks who are proposing to throw out the x86 entirely out from the whole Mac line up and go 100% ARM across the board (including Mac Pro). It doesn't make much rational sense at all when look at the real numbers.

I think Apple clearly knows they do not have a does everything chip. However, they'll sit back and soak up the "faster than a PC" hype to sell more iOS devices (where really can't run macOS or Windows in dual boot mode so could put higher end apps and compare sustained workloads.) But they aren't deeply drinking the Kool-aid and lining up to throw a huge chunk of the Mac line up under by the hype bus.

On the other hand will Apple do a laptop with no cooler. Probably. The MacBook points in that direction. Will it still be a "Mac" book once they prune the passive cooler to a small as possible. Personally, I don't think so. Some folks do. If it does come it will probably be primarily an "even thinner and even lighter" Mac far more so than caring anything about performance. If Apple continues down that path to the absurd levels, then the "highly lacking in wisdom" label will be applicable to them too. If that turns out to be an overwhelming obsession for them then throwing an iPhone SoC into a uber , ultra, super thin "Mac" would be a way to get there.
 
gomac can add what his observation, but I think that is far more so a reference to the folks who are proposing to throw out the x86 entirely out from the whole Mac line up and go 100% ARM across the board (including Mac Pro). It doesn't make much rational sense at all when look at the real numbers.

I think Apple clearly knows they do not have a does everything chip. However, they'll sit back and soak up the "faster than a PC" hype to sell more iOS devices (where really can't run macOS or Windows in dual boot mode so could put higher end apps and compare sustained workloads.) But they aren't deeply drinking the Kool-aid and lining up to throw a huge chunk of the Mac line up under by the hype bus.

On the other hand will Apple do a laptop with no cooler. Probably. The MacBook points in that direction. Will it still be a "Mac" book once they prune the passive cooler to a small as possible. Personally, I don't think so. Some folks do. If it does come it will probably be primarily an "even thinner and even lighter" Mac far more so than caring anything about performance. If Apple continues down that path to the absurd levels, then the "highly lacking in wisdom" label will be applicable to them too. If that turns out to be an overwhelming obsession for them then throwing an iPhone SoC into a uber , ultra, super thin "Mac" would be a way to get there.
Pushing that line could actually be a marketing master stroke for them then.
 
gomac can add what his observation, but I think that is far more so a reference to the folks who are proposing to throw out the x86 entirely out from the whole Mac line up and go 100% ARM across the board (including Mac Pro). It doesn't make much rational sense at all when look at the real numbers.

Yeah, it's a bit of that. Apple A series work is quite good. They're beating Intel to a smaller process. But the A series also isn't magic. I'd expect an Intel CPU on the same process would still probably perform the same or better.

I think a switch to ARM or away from Intel will probably be for business and control reasons. But people assume that out of nowhere Apple is out-designing Intel, and I don't think that's true. It still looks like a lot of smoke and mirrors to me.
[doublepost=1548391013][/doublepost]
Pushing that line could actually be a marketing master stroke for them then.

I'm worried it would be like the PowerPC days all over again. The marketing line will only work if it's true. People aren't going to keep buying Mac Pros on the word of Apple's marketing department.

But it's too soon to know for sure. Intel could decline further. Apple could really put a lot more resources into their own CPUs. And AMD is hanging out off to the side.
 
  • Like
Reactions: ssgbryan
Pushing that line could actually be a marketing master stroke for them then.

I'm pretty sure you are conflating P.T. Barnum sales pitch sideshow barking with actually actual master class marketing. Marketing is actually figuring out what the market is ( the problems that need solutions/products and the size of the scope of the folks with those issues) and matching that to the viable solution your organization has ( viable solutions at win/win exchange points for supply/demand. )

The market needs even more fined grained 'princess and the pea' laptops. Probably. if give most users a choice between 15% lighter with $600 repair after 2-3 years of use (https://www.macrumors.com/2019/01/22/macbook-pro-flexgate/ ) and +1% over current weights and no $600 repair after 2-3 years of use, that vast majority are going to pick the latter.

Going into Rip van Winkle mode for 6 years is not a masterful marketing move either.

We'll solve the "fat" binaries soaking up extra space on relatively limited capacity SSD drives by forces all apps through the Mac App store .... not really masterful marketing either. ( that is really pushing users into a solution to a self created problem. )


Pushing the whole Mac line up onto ARM isn't a particularly viable solution. Apple can't do that with 1-2 A series models. Apple hasn't demonstrated the ability at all to run 5-6 different SoC line ups and keep them all at the bleeding edge. Control for control sake isn't marketing. "Apple has buckets of money in their money pit" ... simply spending giant wads of money isn't marketing either.

There is an article that Ars did with Apple after the A12X that is yet another example of where Apple talks about what they focus on in making the A series solutions.

"... Schiller added that the process of developing these chips starts years before they are released. That begins with teams meeting and talking about how to solve specific user problems on specific devices. ..."
https://arstechnica.com/gadgets/2018/11/apple-walks-ars-through-the-ipad-pros-a12x-system-on-a-chip/

There is a bigger quote above that in the article. Namely apple designs these A series SoC specifically for individual devices. The Mac line up isn't a device. Or 2-3 devices. The real master stroke marketing would be recognizing the Intel/AMD are doing the "device" optimizing across far more device classes than Apple is. Futhermore Apple can't even get a comprehensive Mac line up now out now at a steady rate. If Mac products are tied into more far flung teams at Apple is the rate really going to improve? A custom Mac chips going to get same resources as iPhone chips to "chase' their device teams. ( not very likely at all given the resource starve Mac Pro track record over last 10 years. ). Again part of marketing is lining up with what your org is pragmatically capable of doing with the allocation resources can plausibly get.

Finally to be blunt Apple hasn't particularly executed handing off a subset of duties to the T2 extremely well. Highly inconclusive how moving the whole stack would be even better or even an improvement from the security aspect. ( IHMO, it shows signs of being under resourced and only 5/6th baked. ). Putting your organization on a hype train product rollout isn't master stroke marketing either.
[doublepost=1548438961][/doublepost]
....

I'm worried it would be like the PowerPC days all over again. The marketing line will only work if it's true. People aren't going to keep buying Mac Pros on the word of Apple's marketing department.

IMHO a major part of why Apple "designed themselves into a corner" was a priority to get up on stage and do some 'chest beating' smack talking ( i.e., "can't innovate anymore, my ass! " ). Well it wasn't. A Rip van Winkle product is non innovation. They could have done a literal desktop that was smaller than previous Mac Pros , but they put a self imposed requirement on it to more than half the power, smaller footprint than a Mac Mini , more Thunderbolt ports than has every put on a system (stiill has the 'record' ), etc. etc. Essentially more stuff thrown on top so can strut around the stage to loud bass line and posturing.

The whole Wizard of Oz "I'm the great all power Oz" thing is going to work even less well the second time.


But it's too soon to know for sure. Intel could decline further. Apple could really put a lot more resources into their own CPUs. And AMD is hanging out off to the side.

If AMD keeps marking solid incremental progress on the CPU solution space, holding them off to the sidelines is yet another marketing failure. Apple looks for two (or more) vendors because they can "change horses" if one doesn't cut it. If Apple starts to expand the whole "one vendor or else we'll go inside" they are going to loose that dynamic over time.

Apple put resources into looking into building a whole car and that didn't turn out to be a productive investment either. If both AMD and Intel screw up for 2+ years in row .... sure Apple could possibly move to their own stuff. So far though that hasn't played out.

Apple has their own problems too. Apple fell short on iPhones and Intel took a 'hit' on Cell modem sales (certainly sold more than they did last year but not as much as they hoped. And Apple may dump them for someone else. ) The "infinite supply" money pit may not be there forever.
 
We'll solve the "fat" binaries soaking up extra space on relatively limited capacity SSD drives by forces all apps through the Mac App store .... not really masterful marketing either. ( that is really pushing users into a solution to a self created problem .)

Fat binaries don’t worry me too much. In the classic days we had 68k/PowerPC. Then we had PPC/Intel. Then we had Intel 32/Intel 64. Now Intel 32 is dropping out and ARM will take it’s place.

And regardless of what happens on the pro line, I think ARM is most certainly happening at the entry level/low margin products. Technical considerations aside, it makes sense to not want to pay an Intel tax on those products. ARM is coming to macOS regardless of what happens to the Mac Pro.
 
Fat binaries don’t worry me too much. In the classic days we had 68k/PowerPC.

The point was did you have 128GB drives to go with Fat binaries and double digit megapixel cameras ?

Original Mac Mini (2005) pre "fat" binaries.

40GB baseline drive https://support.apple.com/kb/SP65?viewlocale=en_US&locale=en_US

First Intel Mini
60GB baseline drive https://support.apple.com/kb/SP34?viewlocale=en_US&locale=en_US

2009 Mini

120GB baseline drive https://support.apple.com/kb/SP505?viewlocale=en_US&locale=en_US

At this point the mini has backslide all the way back to 2009 point on storage capacity. (and never mind the price).

Apple's inflationary pricing policy on SSD drive capacity cost puts it on the table. Especially at the lower end of the product spectrum. Apple's laser fcous on dynamic, just-in-time, deployment from their "App stores" suggest they know "app package" bloat is an issue with their SSD pricing policy. I don't expect them to bring fat binaries back. Far more likley if ARM macs do come it will be "get you apps from the App store" solution. ( another reason to make a big deal about Microsoft Office getting there. If they can push Adobe and a few other large resistant players in for mainstream apps. That seems likely they 'cross platform' tactic. ). The folks with mainly high end apps just won't follow is all Apple prunes off is some super lightweight to point of crippling Mac laptop off the bottom. ( essentially same thing that is currently going on in Windows space in relation to ARM. )


Then we had PPC/Intel. Then we had Intel 32/Intel 64. Now Intel 32 is dropping out and ARM will take it’s place.

Apple chucking 32 from iOS and macOS space was a in part a space saving move. ( at least on iOS side).

The other extremely salient point was those were all transitions. 4-5 years later Apple was typically done with the other one. Unless ARM can expand to the "done" state in 4-5 year path. To almost permanently ( 8-10+ years ) path of the Mac line onto a two CPU path.... what is that added long term complexity actually buying? When has Apple wanted to take a mature product line to a more complex long term production plan?



And regardless of what happens on the pro line, I think ARM is most certainly happening at the entry level/low margin products. Technical considerations aside, it makes sense to not want to pay an Intel tax on those products. ARM is coming to macOS regardless of what happens to the Mac Pro.

Apple cranking their average selling prices ever higher but the Intel tax is 'evil". That's a chuckle. Unless Apple substantially changes their tune about cranking the Mac average selling prices higher that whole "money saving thing" is whole lot of horse hooey. If AMD gets consistently competitive then the Intel tax will go down a bit. Apple is upset because they don't have a $400-500 Chromebook competitor product in the Mac's line up ? There is highly little in recent (last 2-3 years ) Apple's behavior that indicates they are loosing sleep and are deeply saddened by not being there ( and mad at Intel because it is all their fault.) they killed off the MBA 11" pice point. They left the MBA 13" almost comatose for years and didn't touch the price much. They refreshed the MBA and pushed even farther away from the MBA 11" price point and left an even oder MBA 13" at exact the same price.
Mean old Intel makes Apple price their SSDs about twice as high as every else. *cough*. Intel is screwing up Apple system price points ... ROTFLMAO.

The notion that we are going to get more affordable Macs if Apple goes to ARM SoC is mostly delusional. Similar the notion that Apple needs to wrestle away INtel's profits so they can just add a deeper layer to their own money Scrooge McDuck money pit. That too isn't a value add for customers and would like bite Apple over the longer term.


Second, unless Apple simply just using "hand me down" iPad Pro SoC parts, are they really going to save anything over the "intel tax" if they have to fund a distinct R&D path for the Mac for a narrow subset of the Mac line up?


The fate of the Mac Pro ( and MBP ) really is a very critical point here. If Apple's 8-10 year path they want to be on is to walk away from Mac Pro , iMac Pro , top end iMac 27" , MBP 15 ( and perhaps thin bezel 17" ) systems and just simply go with "bottom" half of the Mac line up then going ARM would make sense. Get rid of a signficant fraction of models to track with the ARM SoC ( MacBook , MacBook Pro 13, iMac , and maybe a mini that is a headless iMac/MBP ). Probably could one SoC run at different internal function unit utilization levels (turn off some cores in CPU/ GPU) and vary clock speed for performance segmentation. Then completely toss x86 all together. ( don't chase the higher end multicore benchmarks as hard and just cover what they can with GPGPU for multicore. ) .

that would make sense because not particular tracking what Intel/AMD will be covering with about half of their line up.

However, if Apple is highly serious about MBP 15 , iMac Pro , Mac Pro space then splitting the already small space that Mac product line is in doesn't much much sense at all. If they want to quixotically chase the thinnest possible laptop space they could do that just as easily with iOS over a 2-3 year period for far less investment risk. Fragmenting the Mac product line by chopping off 1-2 laptops off the bottom doesn't do much if both AMD and Intel aren't shooting themselves in the foot with large caliber weapons.
 
Last edited:
  • Like
Reactions: barmann
The point was did you have 128GB drives to go with Fat binaries and double digit megapixel cameras ?

I was there. The binary is a fraction of the total app size. I know people who did strip their binaries, but most didn't and never noticed. The system itself was normally the largest contributor, but these days Apple could easily ship a single binary OS.[/QUOTE]

Apple's inflationary pricing policy on SSD drive capacity cost puts it on the table. Especially at the lower end of the product spectrum.

The problem is, as I think Apple has found... barely anyone at the low end runs apps. They use Gmail and Netflix and Facebook and Twitter. The only third party app most people probably run is Chrome. So I don't think they're really worried about low end customers and fat binaries and ARM releases. And Marzipan is their attempt to shove low end users back into the walled garden of apps, which already comes with ARM binaries and MAS distribution.

I don't expect them to bring fat binaries back. Far more likley if ARM macs do come it will be "get you apps from the App store" solution. ( another reason to make a big deal about Microsoft Office getting there. If they can push Adobe and a few other large resistant players in for mainstream apps. That seems likely they 'cross platform' tactic. ). The folks with mainly high end apps just won't follow is all Apple prunes off is some super lightweight to point of crippling Mac laptop off the bottom. ( essentially same thing that is currently going on in Windows space in relation to ARM. )

I think they'll have fat binaries. But as you've pointed out, that won't apply to low end users anyway because they'll be getting their stuff from the MAS.

Apple chucking 32 from iOS and macOS space was a in part a space saving move. ( at least on iOS side).

iOS already stripped the 32 bit binaries from iOS App Store downloads. It wasn't to save disk space. It was to save RAM from having to load the runtime for 32 bit applications. That's why you got a warning on 32 bit app launch, not 32 bit app download.

I don't know if that really will affect an ARM macOS unless they ship a Rosetta like thing for Intel apps, and even then that can go away separately.

Apple is upset because they don't have a $400-500 Chromebook competitor product in the Mac's line up ?

That's exactly what it's about.

Microsoft almost went ARM for their Chromebook competitor, and the only reason they didn't is because Intel panicked and at the last second gave them a sweet deal on a.... Pentium Gold processor.

Apple feels like they could walk in, build a Mac on the iPad hardware platform, and sell it for $500-$600. It's not a bad idea, and they don't have to use Pentium Gold. Intel doesn't have a good CPU in that space. And losing profit margin to Intel is a lot more critical. And a Retina display ARM Mac with trackpad at keyboard could slot in right where the MacBook is, for $600-ish.

Heck, they could just take iPad Pro, throw macOS on it, add a trackpad to the keyboard case, and go head to head with Surface Pro while beating Microsoft on price.

Mean old Intel makes Apple price their SSDs about twice as high as every else. *cough*. Intel is screwing up Apple system price points ... ROTFLMAO.

I don't know of any way Intel can provide hardware at that price point. Like I said, Microsoft went that way and all they ended up with was a sh*t sandwich and a sweetheart deal from Intel on one of the worst CPUs on the market.

Apple's margin strategy isn't really relevant here. Even the PC vendors haven't figured out how to build good hardware at that price point without going ARM.

The notion that we are going to get more affordable Macs if Apple goes to ARM SoC is mostly delusional. Similar the notion that Apple needs to wrestle away INtel's profits so they can just add a deeper layer to their own money Scrooge McDuck money pit. That too isn't a value add for customers and would like bite Apple over the longer term.

If Intel had any products that were competitive at that price point, this would be a different conversation. They don't.

I think in theory they COULD have a good solution for price conscious laptops. This isn't an x86 problem. But it looks a lot like when IBM couldn't get their act together and ship higher efficiency CPUs. And we know what happened when IBM fell behind on PowerPC.

Second, unless Apple simply just using "hand me down" iPad Pro SoC parts, are they really going to save anything over the "intel tax" if they have to fund a distinct R&D path for the Mac for a narrow subset of the Mac line up?

Can't evaluate price competitiveness on something from Intel that doesn't exist.

However, if Apple is highly serious about MBP 15 , iMac Pro , Mac Pro space then splitting the already small space that Mac product line is in doesn't much much sense at all. If they want to quixotically chase the thinnest possible laptop space they could do that just as easily with iOS over a 2-3 year period for far less investment risk. Fragmenting the Mac product line by chopping off 1-2 laptops off the bottom doesn't do much if both AMD and Intel aren't shooting themselves in the foot with large caliber weapons.

iOS is struggling with adoption. It's not at all a solution. That's why the iPad is getting it's ass kicked right now by Microsoft and Google. It's another core problem here. All Apple would end up doing is turning iOS into macOS, and then iOS would eat macOS, and the pro line would be stuck in the same place as if Apple just put macOS on ARM in the first place. The pro line might even be in a worse place if it gets eaten by iOS instead of macOS on ARM.

Apple is also in a place to dictate dual x86/ARM compiles. Even if x86 is in the minority in the Mac market, realistically no one can get away with not compiling for it. And with their push into Metal they've already decoupled performance acceleration from the host CPU platform.
 
Last edited:
You lose Boot Camp/Parallels at the high end... Even if they can force all Mac apps to recompile/rewrite for ARM, one of the advantages of Intel Macs (in the more professional space) is that they do a pretty good job running Windows CAD/GIS/3D modeling/scientific packages while still giving you MacOS for e-mail/Safari/office use/graphics. The big PC packages will never recompile for ARM unless high-end ARM PCs start appearing. If HP, Dell or Lenovo shows up with an ARM workstation line, rather than just tablets and ultralight laptops that sort of run Windows, then those guys will think about supporting ARM.
 
  • Like
Reactions: Aldaris and Nugget
Windows 10 will work on ARM CPUs. If Apple ever goes ARM, you will not lose BootCamp capability, because Windows will work with this architecture.
 
Windows 10 will work on ARM CPUs. If Apple ever goes ARM, you will not lose BootCamp capability, because Windows will work with this architecture.

People use BootCamp because they want to run Windows applications.. Windows for ARM is competing with Chromebooks and light, browser-centric usage. It’s not a realistic alternative to proper x86 Windows. The apps aren’t there at all.
 
  • Like
Reactions: pl1984
People use BootCamp because they want to run Windows applications.. Windows for ARM is competing with Chromebooks and light, browser-centric usage. It’s not a realistic alternative to proper x86 Windows. The apps aren’t there at all.

"Windows for ARM" includes an x86 emulator, no?
However: it's presently limited to running only 32-bit Windows apps, AFAIK.
 
Fat binaries don’t worry me too much. In the classic days we had 68k/PowerPC. Then we had PPC/Intel. Then we had Intel 32/Intel 64. Now Intel 32 is dropping out and ARM will take it’s place.

And regardless of what happens on the pro line, I think ARM is most certainly happening at the entry level/low margin products. Technical considerations aside, it makes sense to not want to pay an Intel tax on those products. ARM is coming to macOS regardless of what happens to the Mac Pro.
Didn't Microsoft try something like this with their Suface / Surface 2 models only to switch to x64 processors in the Surface 3? IMO moving to ARM on the Macintosh would be problematic as software would no longer run natively on the platform.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.