I am usually an Intel person but am cautiously optimistic on the Apple Silicon move precisely because it is Apple Silicon and not actually ARM per se. Remember Apple SOCs are made under their license, which is an Architecture license for the ARM 64 instruction set. They use that set plus a lot of addons of their own design and the cores are a completely Apple design.
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.
Anyone else used to be very pro-ARM but now having second thoughts?
- Thread starter Spungoflex
- Start date
- Sort by reaction score
I also use YNAB, but am looking for an alternative. Their subscription fees are just too much for my simple needs.
TBH, i could probably just use a spreadsheet in Google Sheets, or Apple Numbers
I am only pro-anything when it adds value/benefit to me and what I do. If ARM brings that I will be interested, if not, don't really care.
The geek in me will be getting an ARM device for personal curiosity.
The geek in me will be getting an ARM device for personal curiosity.
@johngwheeler If you are a Microsoft Office 365 user, Microsoft just announced a free financial tool add-in for just this. I don't know anything about how useful it might be to you or how good it really is, I just remembered reading about it. You do have to have a Microsoft account and be a subscriber to MS Office 365, but if you already are this looks pretty good from the review I linked to - it even downloads transactions from your financial institutions.
I am eagerly waiting to see just how much of a performance jump we will see with these new Macs before I get too excited. My dream would be to be a few years from now and Apple Silicon is blowing x86 out of the water and not only that but Apple's GPU's on the SOC have also become insanely powerful to the point where we start seeing major gaming companies also making Mac versions of their releases. I think it could definitely happen as long as Apple Silicon GPU makes a massive jump from current laptop GPUs. If the power is there, I think major gaming companies and indie devs will see the potential bring their games over to reach all the Apple users, and on most games (apart from e-sports/MMO/FPS) could also easily have an iPad or iPhone version rolled out too through the same code. That creates a gigantic user base they can reach with their games.
Totally excited for the arm64-based Apple Silicon future!
Really interested to see how Apple handles the Mac Pro desktop line-up. Interested to see if the iMac Pro makes the transition to Apple Silicon or gets the axe? Expecting some high-core-count high-performance APUs from Apple Silicon, but also wondering if Apple intends to move away from third-party GPUs? I would expect Apple to continue with a large chassis for the flagship Mac Pro, even if they can match AMD & Nvidia GPUs with their APUs, there will still be a need for PCIe slots (Afterburner cards, audio DSPs, video DSPs, etc.)?
I could see a new Mac Pro Cube, which would (ironically) be sort of like the trashcan MP; HPC APU, DDR5 RAM, NVMe storage, & everything else via USB4. Exact same set-up could be in a hyper-expensive 32" iMac Pro that uses the XDR display. Cube would fit for folks who want a replaceable monitor, Apple will also offer new monitors in sizes / panels matching the iMacs. Maybe the Mac Pro Cube also offers integrated audio / video DSPs for a fully equipped multimedia workstation?
I am hoping this Apple Silicon transition will lead to Apple Pro machines being the Silicon Graphics of the 2020s for DCC workstations! Mac Pro Cube, 27" Apple Monitor, & 12.9" iPad Pro (with an Apple Pencil 2) running in Sidecar mode would be a great entry-level workstation set-up!
Really interested to see how Apple handles the Mac Pro desktop line-up. Interested to see if the iMac Pro makes the transition to Apple Silicon or gets the axe? Expecting some high-core-count high-performance APUs from Apple Silicon, but also wondering if Apple intends to move away from third-party GPUs? I would expect Apple to continue with a large chassis for the flagship Mac Pro, even if they can match AMD & Nvidia GPUs with their APUs, there will still be a need for PCIe slots (Afterburner cards, audio DSPs, video DSPs, etc.)?
I could see a new Mac Pro Cube, which would (ironically) be sort of like the trashcan MP; HPC APU, DDR5 RAM, NVMe storage, & everything else via USB4. Exact same set-up could be in a hyper-expensive 32" iMac Pro that uses the XDR display. Cube would fit for folks who want a replaceable monitor, Apple will also offer new monitors in sizes / panels matching the iMacs. Maybe the Mac Pro Cube also offers integrated audio / video DSPs for a fully equipped multimedia workstation?
I am hoping this Apple Silicon transition will lead to Apple Pro machines being the Silicon Graphics of the 2020s for DCC workstations! Mac Pro Cube, 27" Apple Monitor, & 12.9" iPad Pro (with an Apple Pencil 2) running in Sidecar mode would be a great entry-level workstation set-up!
That’s a new take on the Mac Cube angle. Thanks !
Most audio professionals don’t need great GPU power and will surely use multiple external monitors.
Apple still may offer a bigger trashcan or a cube trashcan !!!
The thermal corner for the trashcan MP is right up the alley of the less power hungry Apple SOC. Add to that no need for a dGPU and you have a new desktop option for Mac users !
Most audio professionals don’t need great GPU power and will surely use multiple external monitors.
Apple still may offer a bigger trashcan or a cube trashcan !!!
I don’t think we will see an ARM MacPro for a long time, I’m sure that it will be the last to transition if it survives at all, I don’t think that ARM will be able to match the Xeon in the next two years but I could be wrong. As far as GPU is concerned, I’m sure that Apple will still be able to have AMD graphics. They had ATI (AMD) and Nvidia graphics during the PowerPC days, not entirely the same but still possible. As far as the iMac Pro is concerned, I think it’s going to get merged into the new iMac with BTO options.
I’m super excited, but I will admit I do have one reservation now that I didn’t consider before.
I just bought a 2018 15” MacBook Pro a little over a year ago and it was the most I’ve ever spent on a computer or device. It still runs like new and I was hoping to use it about 5 years at least, which is how long I used my old MacBook Pro. However, now I’m concerned about how long intel support will last. I really don’t feel like shelling out a lot of money again anytime soon and the fact that my resale value for it will tank makes it worse.
I just bought a 2018 15” MacBook Pro a little over a year ago and it was the most I’ve ever spent on a computer or device. It still runs like new and I was hoping to use it about 5 years at least, which is how long I used my old MacBook Pro. However, now I’m concerned about how long intel support will last. I really don’t feel like shelling out a lot of money again anytime soon and the fact that my resale value for it will tank makes it worse.
Intel support should be anywhere from 3 to 7 years after the Apple Silicon transition is complete. I would say half of that as system updates, the other half as security updates. But just because the unit is not receiving system updates does not mean it will not work with the OS it has! So use what you got for a handful more of years, then get in on a third-gen Apple Silicon machine!
If Apple is true to its word, then we should see a new Mac Pro in 2022, but I agree that it will be the last to transition. It may still use a discrete GPU on a PCIe card (ARM designs support PCIe 4) , but it remains to be seen if this will be an Apple designed GPU or a commercial model from AMD.
[automerge]1594075045[/automerge]
If historical trends repeat, then your 2018 MBP should be able to run new MacOS releases until 2024/25, after which you'll get security upgrades for 2-3 years. I would expect Rosetta 2 support to disappear in about 2025/26. You should get a reasonably good life out of your Mac, but yes, it's resale value after this will be negatively affected.
Actually there is a reasonable chance the late Intel Macs will hold some value as the last ones capable of x86 Windows.
Agree. If your Intel MacBook is not a version that has keyboard issue it should hold the value well.
I think this is a reasonable worry. Not that intel support will disappear immediately, but I think one can assume that a bit down the line Apple and developers will abandon Intel based Macs. Apple, unlike Windows, actively eliminates compatibility for legacy apps. It is a strength and weakness: you get apps created with relatively new APIs and frameworks, but older abandoned may no longer work. This is exacerbated when moving to a new processor family.
You could probably alleviate this a bit by selling earlier than you planned. Macs generally hold their value well, but I for one would not buy a used intel Mac in five years time and I think it's reasonable to assume that it will affect second-hand prices.
Not everyone wants to run an OS through virtualization. I use it for windows on occasion and it’s not bad. But there are certainly Windows programs, especially pro apps, that virtualization just doesn’t support very well or in native speeds. And will Apple virtualization for Linux give the guest OS native hardware calls or will that be virtualized to generic equivalents?
It's way to early to decide yet, all based on a an OS that's not even available to the public yet as "stable"
I wouldn't be deciding based on a beta release. Hopefully Apple will come alone, or VMWare will find ways.
In some respect, it my be, but i doubt everyone NEEDS the latest Mac..... What won't you be able to do on Intel chips that you will be able to do on ARM ? Developers will take ages redoing their app. and perhaps not run as fast if it not native code, just converted (programming is not my specially..) but it will still take time. even if they did have help.
I am more pro ARM than I thought I would be. I think because of the fact that it kind of feels like PowerPC again, and I hold all of my PowerPC Macs near and dear from a time where in my opinion, Apple was at its best. For me its probably just the kind of sort of nostalgia of Apple feeling more exclusive with its inability to run Windows natively. Not that in practice this is better, but I did enjoy how different Macs felt than PCs back then. Also, I don't know if I would even expect 5 years of Intel support. Remember, with the PPC, Apple dropped it rather shortly as far as software support goes. The last PowerPC Macs, I think came out perhaps in late 2005 or early 2006. We had Leopard come out in 2007, which even today is not the preferred OS for lack of Classic mode and poor performance on a lot of the G4 systems. So who knows how the next iterations of Mac OS after the transition is nearly complete will run even on the last and best Intel Macs. 2 years later we got Snow Leopard, which I remember on an Intel Mac I had at the time, most definitely sped it up. However that was that for PPC. Only one OS was released compatible with it after the Intel transition. However, today we have yearly OS releases which from my understanding are not as extravagant as the every couple year OS releases of yesteryear as far as feature set goes. So in theory, we may have a "higher number" of OS releases but the same support of 3 years +/- if memory serves. With that being said, I'd be weary of a new Intel Mac purchase today.
Just be mindful that developers will dump Intel/Universal binary just as quickly as they can. The current/next version of software on your Intel Mac might be the last. As soon as Apple rolls out an ARM based Mac, there will be developers who will skip universal binary as they port to the new code. While Rosetta offers compatibility for x86 to ARM, it won’t run in reverse. Universal binary is a good tool to have while moving into the transition, but Rosetta working x86 to ARM to x86, would have been better. It would have allowed Intel Macs running Big Sur to install iOS apps on the Mac. It also would have allowed Intel customers longevity with developers who will be crappy about the transition.
I went through the 68k to PPC transition and the PPC to Intel transition. Both were desperately needed. Maybe this transition is desperately needed too. Perhaps Apple needs a pro line of computers running AMD chipsets for x86 compatibly and perhaps their pro portables should have active cooling along with Ethernet port. A card slot, and at least one HDMI port, a professional grade machine. I don’t know the answers but Apple has a game plan and it has worked before. But I would find a replacement for the app you’re holding onto. A subscription model is the reason I abandoned adobe for better/cheaper software from their competitors. The moment Apples iWork suite can export reliable word docs or excel spreadsheets, I’m done with office 365, but I digress.
the main problem with a one way translator and a universal binary is that you have developers who will bail & fail. They will bail on the platform and they will fail to offer support. The good developers will port with universal binary. There are many of those. And some developers will release their next new version, replete with new features in universal binary. That version will have many point updates as they squash bugs and optimize their new code and get development processes lined up. That set of developers will not release another version under universal binary. They gave you one last hurrah as they streamlined their roadmap.
The last type of developer will release bug fixes for Intel code starting now. If the first apple silicone Mac rolls out in September 2020, that’s when Intel code is gone for them. They will release a final round of patches/bug fixes in the pipeline and they will wait a few months to release the new version of their software which they started developing as soon as the developer support became available from Apple (WWDC). They will stop making new code even though Intel Macs are still rolling off the assembly line. Any hardware that might need a driver or software support- same problem.
Developer support starts dying long before MacOS updates. I’m a little surprised some smart apple Engineer didn’t pipe up and suggest a two-way Rosetta translation layer.
This move is actually different than Power PC. At that time Apple was not designing and fabricating its own chips and now they do.
In fact, the term "ARM" here is a bit of a misnomer. Apple's SOCs use the ARM instruction set but also have a lot of added instructions and the core and chip designs are 100% Apple. It is not the same as someone (for example) putting the latest Snapdragon in a laptop).
In fact, the term "ARM" here is a bit of a misnomer. Apple's SOCs use the ARM instruction set but also have a lot of added instructions and the core and chip designs are 100% Apple. It is not the same as someone (for example) putting the latest Snapdragon in a laptop).
You’re right. It was very disappointing actually. Apple finished their transition sooner than they initially planned. The transition was announced in 2005 with the first Intel Macs planned for June 2006 and the last to ship sometime in 2007. What happened is that they shipped starting in January 2006 with Tiger and the last PPC to be released, the PowerMac G5, announced in October 2005. These shipped with Tiger, but Leopard came out in 2007 as you stated. But Snow Leopard was released in 2009 which is Intel only and the last OS to support PPC software with Rosetta. If you bought a PowerMac G5 in 2005 or a $2,900 PowerBook just prior to the announcement as I did, then you’re pissed when you discover your platform is dead and your Mac is in virtual vintage status at Apple.
But Apple was a different company then. Pro described professional and their pro products came with pro horsepower, active cooling, and ports with expandability. Everything wasn’t soldered on while chasing the thinnest form factor. That’s fine for consumers, but pros need a balance between portability and power and upgradeability. That Apple is gone. And while the MacPro this year was a much needed change it’s yet to be seen how Apple will handle this transition. There aren’t a small number of Intel macs on the market. There are tons. It’s a huge amount of software to boot.
Afraid I can't agree with this. Every Mac user is an intel Mac user (apart from a few PowerPC holdouts). The vast majority are not going to replace their perfectly working older or newly bought machines until there's a compelling reason to do so. Yes, early adopters might shift but these are a small minority.
Developers want to make money. They will do this by supporting the broadest possible range of Macs to get the maximum sales. This will remain the case until Intel users are firmly in the minority.
You’re probably right. Maybe developers figured that PPC would get swept away overnight when they bailed out quickly on that platform. They aren’t going to lose money just because you are one or two versions separated. I’m making the case that some developers will push what they have for the current install base and move forward without looking back. But I hope that’s the case. And it certainly wasn’t every developer either. But there was plenty of software for Leopard that was Intel only and that was 2007, one year after the transition was complete. And at that time the whole install base was still PPC.
But it’s all speculation at this point and I’m willing and hopeful to be wrong.
[automerge]1594246395[/automerge]
I agree. I think it would have been a smart move considering that Intel is a much larger install base today than PPC was in 2005. Apple is roughly 10% of the computer market (if memory serves) and a significant percentage of the upscale computer market. That wasn’t the casein 2005, despite the turn around that Jobs pulled off. I thinkhaving a translation layer that can work forward and backwards might make some users feel less like they’re “the other” in the transition.
I think you might be overestimating the difficulty of maintaining the Intel version. This is not the same as the PowerPC transition before: ARM64 and x86-64 have a lot of points in common. Besides, since Intel platform offers stronger guarantees, if your code correctly works on ARM, it will almost always correctly work on Intel (not via versa, as some bugs in your code might not become apparent when running on x86-64).
And besides, two-way translation would be a terrible idea. ARM version running on x86-64 would be dead slow.