I think the thing most people are forgetting, as i read these comments about x86 is... the entire industry is moving to RISC computing and they are becoming so fast, that in addition to their 30 percent advantage they are now starting to outpace CISC on raw power alone. So it doesn't really make since to still support x86. That being said, by the time we get down to the 3nm range on RISC, there will be so much extra room on the die, they might just allocate a portion of the extra room to a hybrid module that can perform basic x86 or convert at a fairly significant loss, but still fast enough to perform basic to intermediate workload. Also keep in mind that Apple is now designing (with outside help in a lot of cases) their own chips. It is possible by creating a larger die for desktop systems, their new chips could carry two independent systems. This wasn't possible before as the die would have to be so large, but apple is already sporting the largest dies in the industry, as we shrink these things below the 3nm, there really are no rules on what can or can't be done.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
TSMC to Remain Sole iPhone Chip Supplier, Could Provide ARM-Based Chips for Mac in 2020, Apple Car Chips in 2023
- Thread starter MacRumors
- Start date
- Sort by reaction score
Wow, I guess I don't really agree with you on that. Yes there are far more "apps" for iOS due to the iPhone and iPad popularity, but OSX is a robust hard working platform used by millions in ways neither the iPhone or iPad can go. The iPad is getting closer in functionality with the "Pro" versions, but it isn't there yet.
You must be a solid iOS user to say such a thing. Not that there's anything wrong with that.
It's a solid platform, but come on, it's hardly flourishing compared to Windows, iOS or Android. The App Store is a few decent apps (many if which are still 32bit) amongst a barren wasteland of off-brand versions of more popular Apps. Bringing the iOS developer base into contact with the Mac will be the best thing that could happen to it. And I'm a very contented Mac user, thank you very much
I didn't quote the whole thing but you are correct - the issue is that we have been on Mac forever. We need bootcamp to open some specific apps that have no counter part on Mac, to do conversions, etc. Great to not to have to have another machine just for the times we need to run those programs. I certainly see in a large organization that bailing on Apple is probably the smart move.
What Apple has done with the Finder interface is what keeps us in their thrall. The quicklook feature, the copy and pasting between all different apps. The clean window/folder relationships, the lack of really bad viruses and malware. They have a really nice OS for a Creative Pro to work with. Nicer than windows. Switching would be unpleasant. It is so frustrating that the sole focus is on consumer gadgets. Apple could have it all if they would just split the Mac computer/MacOS into a separate division. It wouldn't hurt them and they could build a massive ecosystem. Cars and Trucks as Jobs so rightly said. It is just bad management that has kept this from happening. Cook is just riding the inertia that Jobs created. He won't have a clue if that inertia starts to fade.
Also being a consumer gadget is ultimately dangerous as consumers are super fickle. I remember Motorola phones were number one, then Nokia was the end all be all, and what happened to blackberry? Consumers will dump you in a second for the latest flavor.
Having a solid, Pro division in the company would maintain a dedicated, long term user base that Apple shouldn't take for granted. That group saved their bacon in the late nineties before the iPod appeared - losing that group - when they are easy to keep makes no strategic sense.
Because they will release it with Intel, then release update it later ? (or just defer it)
If Apple did release the Mac Pro with Intel they could make more money as people would then upgrade to ARM-based.
Also, if Apple DID eliminate Bootcamp. at least it had a good run from 2006 when Apple jumped to Intel chips. Since virtualization has become more popular, and you could argue you can run 3D gaming just fine with VM's, the only thing you really need Bootcamp for is today is the 'latest' games.
The reason for keeping it, that list is getting shorter.. The 'pro's' to ditch it, are outweighing the odds of VM's ever time. and cpu-cycle based apps work better in BC.
Last edited:
You do realize that conventional RISC is not RISC anymore.
X86 is not longer the x86 of the 486/586 generation either.
The AMD Ryzen is an all new architecture and so is the Xeon.
For Apple to compete with these companies, it's not a single processor but a family of processors.
Ampre computing has a 32 core ARMv8 processors with a peak speed of 3.3GHz, 32KB L1 insrecutionb and data cache per core and a shared 256 KB cache per two cores, 42 lanes of PCIe, 4 SATA Get 3 ports done in TSMC 16nm FinFET
This is the class of processor they will need to compete with in a "PRO" desktop machine.
This is not just an A12x on steroids.
You talk about 3nm and nobody is currently taking about a 3nm node, when 7nm processors are just hitting the market.
Apple isn't anywhere close to the largest dies in the industry. They are nowhere near the reticle or die size limit. Current AI and neural networking chips are pushing reticle limits not anything Apple currently does.
If you want to look at CPU beasts, see Ampere, Qualcomm Server, Cavium, Ryzen, Cisco Networking among others.
An A12x is small peanuts in the ASIC/SOC game.
I don't think Apple is going to dump Intel anytime soon.
Rumors had Apple doing it's own LTE modems and processors two years ago and they still buy Intel modems.
My doubts have absolutely nothing to do with MacOS running on ARM, that's not even a consideration.
NeXTStep, i.e. first MacOS X, ran on 68K, 88k, PowerPC, Sparc, HP-PA Risc and X86. MacOS is portable and there is nothing to see here.
My doubts don't come from the ability of Apple to buy or hire the people to make an ARM server class/desktop class processor.
My doubts have everything to do with ROI and indications and statements from Apple that iOS and MacOS are different markets and they aren't trying to merge them.
There is no monetary reason for Apple to make a processor to compete with x86.
Performance isn't a reason because they don't use the fastest Intel/AMD chips already.
It's not pricing because Apple gets the best prices and early access if they like, to any Intel processor made.
It's not features, because they partner with Intel to get any feature they want. They have done it before.
They are going to need Verification engineers, RTL design engineers that have background in cache coherency and CPU architecture, additional back end engineers, test insertion, chip fabric engineers, more signal integrity people, etc.
They will need to ramp and you cannot hide that kind of ramp from people in the industry doing chips.
So as someone that still works in the chip industry everyday, I'll say, what I said months ago; the people I know and I'm one of them, designing high end ASICs and processors haven't gotten any calls for jobs for the type of ramp that Apple would need to do processors.
I just attended ARM TechCon and funny thing nobody doing ARM designs even mentioned Apple in the last couple of days in the context of ramping to do anything high end. The community of people doing high performance computing in the CPU space isn't that big and Apple is anything but a small stealth company.
So we'll see. But I was right 3 years ago when they first started talking about Apple making chips for laptops in 2017 and 2018. Nothing appeared as of yet.
separate kernels. They have similar messaging designs, but are separate beasts.
You can - today - put apps on the App Store with "bitcode" enabled, which means you don't even have to recompile yourself.
[doublepost=1539935151][/doublepost]
It's not "your iPhone app to Intel code". It's "your app running on _any_ processor as long as the operating system is available." If I can run apps on iOS, no matter what the processor, why on earth would there be any problems running an app on macOS, no matter what the processor?
[doublepost=1539935390][/doublepost]
The processor in the iPhone XS is perfectly capable of running a MacBook, including a MacBook Pro, or an iMac. Not the high end versions, but nobody expects Apple to switch _all_ Macs. An iPhone isn't capable of running at highest clock speeds for extended periods of time, but that is just a matter of cooling. No need for re-designing anything for high end.
[doublepost=1539935737][/doublepost]
There is no reason for Apple to switch _all_ Macs.
One place that I worked, nobody used a MacBook with Bootcamp. Lots of people used Macs. Windows users with enough pull in the company would get a MacBook, throw out MacOS and install Windows with our company license. No bootcamp. I have no idea how many "Windows PCs" Apple is selling that way, but there's no reason to give up that market.
[doublepost=1539935902][/doublepost]
MacBook Pro: Both ARM and Intel variants. Intel for the highest spec.
Last edited:
I don't disagree that for consumption you can be just as productive on an iPad as a traditional computer.
But for production, I can't see it. I am far more productive on my computer than I am on my iPad.
I can't see me being as productive at writing code or editing video on an iPad. There are just some tasks that are more suited to keyboard shortcuts and precise movements that a cursor offers that touch only does not offer.
Do you prefer excel on a computer or iPad, I can't stand excel on an iPad and only use it to view files.
[doublepost=1539936827][/doublepost]
Calm down...
You were responding to someone who said that it wasn't a case of flipping a switch.
You responded by saying that it was for most software, its a bit disingenuous to imply that moving to ARM is mostly solved by flipping a switch.
Rosetta isn't exactly magic. Do you seriously think nobody at Apple knows how Rosetta worked? Compiling x86 code to code for other processors was done in the mid 1990's. It's twenty year old technology. And having to implement this not for the operating system, but just for a 64 bit application on a little endian processor makes it a lot easier.
Apple is also _big_ in compiler technology. So if they want to make a fast x86 emulator to run old apps, that's no big problem. I doubt they will do it, but they could.
calm down
how long do you think Amazon was not making a profit for? Tesla is not all doom and gloom.
Anyone’s ability to use an iPad as their work computer always comes down to the tasks they are trying to accomplish. For me, I’ve been able to adapt my workflow easily and it’s been a much better experience than any computer I’ve used. Nothing to do with consumption. I really only use excel to review reports so I haven’t had a problem with it. Not ideal for any sort of operational work though.
Last edited:
It still is. I've been running a Windows 10 build since the OS was in beta. When it first came out, it was an awesome upgrade from 7. Since that point, it's become so much worse. I can't stand using it anymore. Recently I reinstalled the LTSB version and that made it more bearable.Of course, it helped that the alternative was horrible back then
Needless to say I mostly just use my MacBook these days.
When people say consumption they are usually referring to the simpler tasks on a computer.
When people are producing things they are usually referring to more complex tasks.
What I am getting at is that an iPad is not good at typical complex tasks where you need cursor or proper keyboard support.
So when you say not ideal for any sort of operational work, that is what I mean by production
and when you say i really only use excel to review reports, that is what I mean by consumption.
Areas of production where the iPad excels is in artistic graphic work where someone is drawing.
We'll just have to disagree and move on. When I think of consumption, I think of watching videos or playing games, what is typically discussed here as the only good uses for an iPad. On this site, if you don't code, key in data, or any other strictly operational repetitive task - it's "consumption", it's not "real work", or it's not "complex" enough - so it's not productive. I use my iPad to run virtual meetings, draft client proposals, create and share presentations, review and adjust analytics dashboards, manage team files and reports, do one one Facetime meetings, manage and draft emails, collaborate on documents in the cloud with my team, the list goes on. I consider that productive work, most on here don't. In the end, if my company wants to pay me what they do to play with a toy, so be it.
My comments aren't all directed at you by the way, as you have been fairly objective and rational in your discussion, but the underlying message of your comments remains the same - the iPad cannot be used for "real work", which is just a tired old mentality at this point.
Amazon? 10 years in the red; doesn't mean every company will turn a profit after that length of time. Don't get me wrong maybe the World gets hard for Tesla cars next year and pre-orders sky rocket but keep this in mind when talking about the Model 3 from:
http://uk.businessinsider.com/elon-musk-tesla-model-3-we-were-huge-idiots-2018-7
"We were huge idiots and didn't know what we are doing" That's a direct quote from Elon Musk. Fills you with hope doesn't it?
Yes, I'd be disappointed by the fools that didn't learn from their mistakes.
Pre-orders have already skyrocketed before the car was even revealed.
[doublepost=1539950152][/doublepost]
I wasn't meaning it can't be used for real work, it clearly can. Just that there are a class of production work that cannot be done as efficiently on an iPad as can be done on a traditional computer. such as video editing, software development, cad etc
I'd even be suspicious that the full version of Adobe Photoshop on an iPad will be as productive as a computer, but probably falls somewhere in between some of more difficult tasks I mentioned above.
Any die hard photoshop advocates out - can you answer if you would switch over 100% to the iPad?
That's fair, I agree. I have said all along, it depends on the task. Some are great on the iPad, some not so much. I think that will only get better with time. I can't speak to your question myself, but my fiancee used Photoshop for years for her graphic design business. She now has moved onto Affinity apps on her iPad Pro and she has loved it. She does revert to her MacBook Pro though for extensive projects, so I think your point is valid there.
I didn’t say they shipped the same kernels. I said the OS X kernel compiles on arm.
For some niche cases like that an iPad could probably work. My daughter is doing amazing drawings with one. But in my world of being a cpa, an iPad is basically worthless besides showing off financial statements and using to compliment a pc.
And that’s what I want to see apple doing. Making it an even better compliment to my Mac or pc. Sure it can probably replace a laptop for certain people but then again an mbp was probably incredibly overkill for these people with simple needs. Funny enough I see this generation as even less pc savvy vs latter 90s-2000s. They’ve learned iOS Android, chrome book and whatnot through school. Office is foreign to them. Some may not touch a computer before college. Where as I was learning in the 90s PC only. I think having little pc skills is setting up kids to struggle later but that’s public schools for ya.
I’d still prefer to see apple market it as a compliment vs replacement. What can both an iPad and Mac do for you together. But back to chips. Without windows, I can’t use a Mac. And I don’t mean the useless win 10 toy apps. Heck it’s hard enough now to make a Mac work at my firm but worth the hassle.
I manage a team of 60 people using just an iPad Pro. Not sure if I consider my position niche, but your ultimate point is true - it all depends on the tasks needed. As the iPad continues to evolve, more and more functionality will come, allowing more and more tasks to be accomplished using said device. I don't think Apple is marketing it as a replacement (although I get that from some comments/commercials), but more so an alternative. Not everyone wants (or needs) a traditional desktop experience. For some, an iPad is a better alternative, just like a laptop or desktop or tower doesn't work for someone else. It's just another computer to choose from.
Not "fat" binaries again.
Intel just announced the i9 9900K.
8 cores, 16 threads, base frequency of 3.6GHz max of 5GHz, 16MB L3 cache.
It doesn't support ECC so this is a iMac type processor with a TDP of 95W.
So besides completing here Apple would also need a server class processor to compete in performance with the likes of current Xeon, Ampere, Cavium, Ryzen, etc.
And remember it's a moving target so they need to be looking at the i10 one to two years from now.
Also getting from 2.5GHz or so to a 5GHz peak like the i9 is no easy feat.
Like I said, I don't see them doing a processor to replace Intel, but time will tell.
Increasing clock speed is the worst way to increase performance. Having to go to 5GHz is an engineering failure, not an accomplishment. P=CV^2f