Apple moving to their own chips

Support Mac Technical Support
142

It’s less “can it be done on ARM?” and more “can it be done well enough to be more passable than a MacBook Air on ARM?”. The jury’s still out on that. More and more over the last three years Mac users have been “why is X game running so badly on newer machines vs. the older ones?”

90% of Apple’s Achille’s Heel right now is APFS. It’s just that crappy a filesystem for gaming. It isn’t even a good work filesystem if you make tons of edits or patches to files. SSDs are propping it up. But no other filesystem on the planet dies so horribly on a standard HD than APFS. You can bet Apple will put in huge honking letters “hard drives are not supported” on their ARM Macs. It would effectively be a death knell to the burgeoning platform to have really bad reviews, and Apple won’t go for that. They’ll do everything they can to make it seem like only an SSD can run an ARM Mac (sadly, with APFS they’re spot on there).

The rest is going to be the harder question to answer: Is it worth it for Blizzard to add a third effective platform to the game, only to kill off one of the three (Intel Mac) quickly afterward? Because Apple will not update feature sets or fix bugs for the Intel Macs once the ARM Macs have been transitioned to. You saw it happen with the 68k to PPC and PPC to Intel transitions. It’s worse now. Apple isn’t fixing stuff on current architectures, let alone one that’s a dead end after the fact. That has to weigh fairly heavily on the bean counters at ActiBlizz.

Remember, Mac fixes in game code only get pushed with Windows fixes in most cases unless the fix is for a catastrophic bug. See: Diablo 3 still having pixelated textures for a year and a half or WoW being unusable on Intel IGPs w/o essentially disabling nearly all particle effects. Blizzard can only fix so much without Apple’s help, and Apple hasn’t exactly shown a willingness to do much of anything on their end regardless of how many Radar reports are fired their way.

Here’s another scenario to consider, which I’m sure is on Blizzard’s radar: Apple is designing everything to be one platform. Essentially once the ARM transition is complete macOS will be just a rebadged iOS, complete with lack of user control. Now, being that they’re all so intertwined and that Apple wants everything to work on all of their devices - just what do you think is going to happen with a game like WoW? Yep. People are going to want to be able to play it on their iDevices. Sounds stupid until you realize just what crowd Apple is trying to draw in here, and it isn’t the brightest bulbs on the planet, I can assure you. Does Blizzard really want to have to tell players “sorry, no can do on your tablet even though it’s ARM like your desktop is”? It’s something to think about.

This transition isn’t anything like the previous two. Apple’s going all in, and not giving developers a whole lot of incentive to follow in their footsteps. Short of having nVidia level performance out of their SoCs, which I highly doubt is going to happen, we’re not only losing 95% of the games that already exist (32-bit), but also being forced to settle for less than optimal yet again, just like we have had to with LCD displays coming off of CRTs (not-black blacks, easily measurable input lag, burn-in [early on], and too many competing panel technologies, most of which had worse color than even NTSC 1.0 did).

And if their interoperability with Windows computers, or lack thereof currently, is any indication, Apple’s going full bore into just their own ecosystem. Not exactly a rousing reason to follow them.

1 Like
143

I’ll probably end up buying an arm mac for imessage and iphone interoperability in general. Plus idea of running a few iOS apps on desktop is cool. Probalby also still use it as BBEdit coding computer because I’m just used to it and don’t want to change to a different UGLY windows IDE.

However I’d ALSO have a full blown windows PC for gaming and such.

144

Sorry, but I think this is a serious misreading of reality, at least as Apple sees it. From their standpoint, Intel is now where Motorola was with the PowerPC in its last couple of years: underperforming and not delivering technically. And unifying the architecture, at some level (obviously the GPUs in Macs are not going to be the same as in Apple Watches) saves Apple OS and library development money. As we know from decades of experience, Apple is all about the money.

Apple was perfectly happy with Intel as long as it was delivering better performance at a reasonable price. Apple has to realize that bringing CPU development in-house brings some degree of risk, but they’re comfortable enough with the ARM architecture after 13 years of iPhone experience that they’re willing to make the jump. And while they may call the new SoCs “Apple” silicon, the fabs are TSMC’s, on whom Apple has found they can rely far more, and for far larger numbers of processors and other components, than they ever could Intel.

I’m sorry about your bad experiences with servers. When I was still working, we did nearly all our file service over NFS. While Apple frequently changed configuration methods for that in the earlier versions of macOS, it eventually settled down, and it always worked. I have a hard time understanding why anyone uses anything else.

And I have to say my experience has been that Thunderbolt (all versions) has been terrific for desktop storage. Not cheap, but worth the premium pricing (says the guy who actually bought an optical Thunderbolt cable so he could daisy-chain two storage towers ~ 5 meters apart).

1 Like
145

Sorry, but I think you’re seeing this from a purely games perspective, which hasn’t been an Apple priority since… always.

Apple wanted Metal for its prosumer customer base (Adobe, video editing, &c.), not for gaming. OpenGL had been tolerated for a while not because of gaming or its usefulness to the scientific/medical community, but because they had to have something to serve that prosumer market segment. Devs may not want to wait for every major bug fix and missing feature they need, but that’s always up to the whims of the OS developer. I suspect Blizz has had to do a workaround or two on what Microsoft was offering at any given time in their (Blizz’s) product dev cycle.

I believe what you say about macOS and iOS developers’ perceptions of Apple’s responsiveness. Money talks, and Apple is not a charitable organization.

I have no idea how long WoW can last on Macs. My pessimistic view is that Blizzard will simply say, “Meh, not worth it at this point,” as it has for all its newer games. and pull the plug when the last Intel Mac is replaced with an Apple Silicon model. Or they may be generous and continue low-level support until Apple declares the last Intel Mac “vintage.” Since only Blizz knows what fraction of their subscription base is macOS-based, I don’t know how the arithmetic works out.

146

This was planned long before Intel started having its Skylake refresh refresh refresh refresh refresh slump. Apple wants total control. That’s why nVidia is no longer in the game. They refused to cave to Apple’s demands for control and Apple just flat out blocked them (in a very anti-trust way to boot).

Of course it is. Why do you think we’re still paying the Mac Tax™ despite Macs having lower end features than gaming PCs and laptops costing less?

There’s some irony here in that nVidia is potentially looking to buy ARM. Two guesses what happens to Apple’s ARM licensing if that happens?

The servers are running Windows Server, not linux based. For smaller businesses like a dental practice, forcing the staff to have to deal with NFS is usually not practical. It’s the most “agnostic” of the protocols, yes, but hardly intuitive. The fact that OS X can’t even see Windows computers on the network should be an indication of just how far Apple has gone to skew things toward their own ecosystem. I could see every iDevice on the network instantly. But not one of the workstations in the operatory and hygiene rooms, nor the server itself, which is required in order to map the network drive for Open Dental (or Carestream Dental in the case of my primary dentist).

Thunderbolt’s problem has been one of fragmentation more than anything else and the confusion between it and USB-C (either being a subset of the other makes for frustration if one isn’t implemented properly or at all at the device level). And a lot of us have limited real estate. Having internal storage space for drives means no wall warts, no cables snaking everywhere, no USB-Cwhat?, no extremely short distance limitations outside of optical cabling, and no bloody mess of enclosures everywhere. TB3 docks, RAID, and eGPU enclosures make sense, but having to store a mess of other crap just for storage that could easily be internal? No. That also adds points of failure, I might add.

The issues are almost never in the OS itself. Both MS and Apple do a fair bit to make the OS as problem agnotic as can be (Apple being more successful here as we’ve witnessed in Windows 10’s version 2004 release this year). The issues with games are nearly always in the graphics drivers. Apple controls these for both Intel and AMD. They have in-house engineers that they demanded from the two. AMD gladly gave the engineers because they don’t want to have to dev for OS X anymore on their own. Intel is the same. nVidia’s out of the equation because they refused to do the same and preferred to dev their drivers themselves, almost always more quickly than Apple does with the other two vendors. Remember the Desert Sands crash in Diablo 3? It was fixed six months earlier on Windows than on OS X because Apple didn’t bother to roll out the change until the next major OS version drop. Nobody in their right minds is going to trust Apple with timely fixes after years of seeing them drag their feet every time.

This is the question everyone wants an answer to. I suspect it’s going to be support for Intel Macs until the features Blizzard wants can’t be implemented and/or bugs can’t be worked around anymore. If they were still on the Mac bandwagon, we’d likely have seen an Overwatch 2 or Diablo 4 announcement stating support for the Mac platform. That hasn’t happened. And as evidenced by a growing number of hackintosh users like myself that are decidedly, and rather lopsidedly, in the Windows camp at this point, the sentiment is that Apple is in it for themselves and nobody else. Power and true Pro users aren’t Apple’s goal anymore. It’s the iDevice users. The money is in the mobile space, so making Macs “home mobile” devices is where it’s headed. And I’m not buying a Mac just to play mobile games on my TV. Sorry. If I wanted that I’d get a Shield TV and have a better experience (and linux to boot).

1 Like
147

Well whatever happens, screw using Windows as my main OS. I’ll set up a separate winbox for gaming purposes only or pull myself through the endless-tweak-hell that is desktop Linux if that’s what it takes, but I refuse to use Windows as my daily driver.

1 Like
148

I guess my biggest question is will this allow users to play wow on mobile devices like the iPad and iPhone since iOS games will now work on Mac OS X devices? Doubt it thought it might be a cool idea to explore considering the power behind the iPad Pro.

1 Like
149

Or do what I’ve done and lock Windows 10 to version 1903 where it’s stable and everything works. You can disable updates until you find one that works properly. Thankfully Windows’ restore points work better than Apple’s version of that feature. I still prefer OS X though. This just sucks all around. :frowning:

It’s possible that it will, but that would also mean that Blizzard would have to add in touch control to the binary as well. By this I mean explicit touch control, as regular iDevice binaries do not have this, much the way that iDevice binaries would also need to add explicit keyboard and mouse control to work with non-iDevice controls on desktops. iDevices can handle mouse and keyboards just fine, but not by default.

As I noted before, this isn’t like the previous two transitions. This time Apple is trying to make a gigantic mobile-desktop amalgomation out of their platform. While there are portability benefits, there are a lot of other detriments as far as being a programmer is concerned. Apple may have their iOS/macOS compilers on Windows now, but that doesn’t mean it’s going to necessarily be worth it as a developer in the long run.

150

I’m sure that improves the situation quite a bit, but I have a laundry list of things about Windows that drive me up a wall, parts of which have been standing for over a decade and may never change. Honestly I think the only way I’m ever going to be happy on a desktop OS other than macOS is if I write my own desktop environment for Linux, which isn’t at all practical even if it’s technically possible.

151

The grand irony of Apple’s transition is that it may end up giving Linux an unanticipated boost. Not a large one mind you, but a noticeable one. And anyone that learns linux at anything more than a cursory level gains some fairly valuable skillsets in the process. Still, I wouldn’t want to have to rely on Proton for my gaming needs. :wink:

152

Apple has a perpetual license to the ISA.

153

I had planned to buy a 13” MBP a couple months back when the Apple Silicon rumors hit. Now I’m waiting to see how they turn out. I expect I’ll buy one, because all this talk about walled gardens and old game libraries and so forth is far less relevant to me than a few more hours of battery life would be.

Ultimately, I think that’s what Apple’s real goal is: to build better machines than they can build with current processors, where “better,” for most of us, means some combination of faster, lower power consumption, and thinner/lighter.

Which is not to say other people’s priorities aren’t valid, of course, but I’m personally not going to feel cheated if I gain those qualities in my next machine at the cost of games.

I don’t think it’ll come to that, though. If I had a nickel for every time someone told me a change Apple was making would kill software development for the Mac, I could afford a machine for every room in the house. Hell, even just John Dvorak’s predictions would be sufficient.

1 Like
154

If things play out that way, it’d be kinda like a modern parallel to the Powerbook G4 era, where Mac laptops weren’t really gaming machines in any sense of the word, but were still pretty great laptops compared to the Windows laptops of the era. Windows laptops have improved a lot since then (hell the Razer Blade is practically just a black MacBook Pro), but if Apple can achieve a substantial margin on performance, power consumption, weight, or even design (custom SoCs open up designs not previously practical) they might put themselves in their own category for laptops once again.

2 Likes
155

They might be in their own category, but it would be for the wrong reasons. Interoperability being primary amongst them. The real world uses Windows for most computer tasks at the business level, and iDevices out in the field for mundane tasks (Apple’s tablets are far superior in ease of use to Android tablets, sadly). So unless Apple reintroduces good networking discovery and transfer capability, it’ll still be their own ecosystem they’re in, not the bigger whole of everything. That’s fine for mundane tasks, but when you need to get stuff done in a mixed environment, it makes Apple’s units dead weight.

156

Eh, interoperability isn’t much of a roadblock. I work in state government, and we’ve been expanding our Mac user base for several years now. Not a lot of our material lands in OS-specific applications that can’t be moved back and forth pretty trivially. I imagine engineering shops (Boeing and so forth) get platform-locked more easily, but for the large majority of the world that survives on office applications, it’s just not a big deal to mix platforms anymore.

1 Like
157

Well, with things like Office 365 of course that part is trivial. But for specialized suites or networking it is a problem. It’s mostly a networking issue, since a lot of those specialized suites have Mac versions (like Open Dental does). For non-office document type work, interconnectivity matters.

158

The one elephant in the room everyone in this thread is forgetting is Microsoft’s efforts t have an ARM version of Windows. “Windows 10 Arm emulates x86 applications with very decent performance. x64 emulation is in the works but isn’t expected to arrive until 2021.” - Windows on Arm in 2020: An ecosystem finally worth buying into?

“In a sea change, Apple Inc. and Microsoft Corp. are exiting Intel Corp.’s x86 processor architecture for their personal computers. (…) Microsoft is also investing in Arm-based PC chips and PCs for Windows and is planning to remove the software migration barriers by 2021. (…)
The primary reasons for this transition are lower costs, a reduction in power requirements and a common platform enabling applications to run on smartphones, tablets and PCs. In turn, this universal platform will enable faster adoption of software and hardware innovations and lead to higher productivity for the end-users.” - Exiting x86: Why Apple and Microsoft are embracing the Arm-based PC

Let’s face it, even Microsoft has realized that ARM and not x86 is where the future is. Microsoft likely will have emulation longer then Apple thanks to the huge amount of legacy software but the handwriting is on the wall.

Any developer that ignores this does so at their own peil. Might as well bit the ARM bullet now then goin into panic mode three years from now.

1 Like
159

Regarding Wine this came up on the WineHQ forums and while there is a ARM version of Wine around it will only run ARM programs not x86 ones. Given Microsoft is going to ARM as well I suspect that some form of speedy opnesource x86 emulator will show up before the built in emulators in the MacOS and Windows 10 buy the farm in two to three years (perhaps longer for Windows but it is clear that X86 code is on the way out even on the Windows side of things).

1 Like
160

I’m going to be blunt: Apple can get away with this because they are Apple. If MS tried this, then windows would be dead to the world and people would just run older copies that still ran on x86. To be fair, I know of customers with literal warehouses full of either m68K based macs or PPC based macs. Because the critical software they need to run is still on that architecture. For MS this is largely 16bit software that requires 32bit versions of Windows. MS is not going to abandon those customers. Moreover ARM is not some mythical savior, it’s just an ISA. To get comparable performance to x86 you’ll have similar power, latency, memory, etc. requirements and that will mean similar results. Apple is mostly tired of Intel lacking innovation for the last 5 generations.

In other words… this is about Apple more than ARM. As for Blizz biting the bullet: Any software on an ARM mac is likely to need to be installed only via the app store or enterprise app store. Meaning blizz would have to give apple a 30% cut. It would also mean supporting Apple GPUs, something that Blizz has never done themselves and always farmed out to other companies like Unity etc.

Qualifying a new platform is incredibly expensive, so unless you’re the Unity/Unreal Engine/ etc. and are targeting mobile… it’s probably not worth it for the less than 10% of customers that will actually be on ARM by 2030. It’s definitely not worth doing fiscally in 2020.

Oh and waiting at their own Peril: Hardly, you stay where your customers are. If you chased every upstream plan of a vendor they’d have made an Itanium version. Itanic is a well known joke in the industry.

161

I probably will get an arm mac for my iphone/ipad integratino/utility to replace my hackintosh when it goes windows only for gaming. However, I won’t get a first gen one. if i’ve learned anything, first gen apple hardware is buggy. I mean heck the first gen intels (not apple tech persay but still rushed was 32 bit then literally the next year was 64 bit with less bugs. my 32 bit macbook pro intel had coil whine and had to have logic board replaced…twice)

Never again with first gen. First generation users can beta test for me and I’ll get refined less broken second gen.

1 Like