Devious marketing of M2

[Joe Dohn]

That list is misleading (BG3 is native, but is listed as needing Rosetta2).

It could be that the game was not available when the list was first compiled.

Either way, even today no matter the list, you'll see that you won't have many options for native gaming on Apple Silicon desktops.

 

[maflynn]

Either way, even today no matter the list, you'll see that you won't have many options for native gaming on Apple Silicon desktops.

I don't think the narrative has changed with the reputation that you cannot game on a Mac. I see that all of the time with Yters reviewing products. People on gaming forums, reddit. I can only assume that philosophy carries through to gaming execs as well. With the consolidation of the gaming industry, Sony, MS, and Tencent having most of the marketshare, I don't see the needle moving any time soon.

While the M2 is more capable, it comes at a price - literally the price of the MBP. While gaming laptops can cost the same of the MBP, they have the distinct advantage of being made for gaming, where as people are not going to spend 3,000 to 4,000 dollars for a M2 MBP, with gaming as a primary usage.

 

[Joe Dohn]

While the M2 is more capable, it comes at a price - literally the price of the MBP. While gaming laptops can cost the same of the MBP, they have the distinct advantage of being made for gaming, where as people are not going to spend 3,000 to 4,000 dollars for a M2 MBP, with gaming as a primary usage.

Even so, you'll still rely on virtualization, cloud gaming or Crossover. That's the real issue here. You'll be paying top dollar for something that not only cannot use the hardware to its fullest potential, but that in theory can easily be taken away from you.

 

[diamond.g]

I don't think the narrative has changed with the reputation that you cannot game on a Mac. I see that all of the time with Yters reviewing products. People on gaming forums, reddit. I can only assume that philosophy carries through to gaming execs as well. With the consolidation of the gaming industry, Sony, MS, and Tencent having most of the marketshare, I don't see the needle moving any time soon.

While the M2 is more capable, it comes at a price - literally the price of the MBP. While gaming laptops can cost the same of the MBP, they have the distinct advantage of being made for gaming, where as people are not going to spend 3,000 to 4,000 dollars for a M2 MBP, with gaming as a primary usage.

Apple can fix this by buying a few publishers (EA & UBISoft come to mind).

 

[leman]

No. How many recent native games do you think were released?
https://doesitarm.com/games only lists 5.9% games as Apple Silicon native games (no Rosetta / virtualization / Crossover) required.

I counted here, and it only gives me TWELVE games. That's only TWELVE games TWO years after the release of Apple Silicon. That's not even considering some of the games being counted aren't AAA games (e.g, Stardew Valley, Among Us).

It's REALLY bad.

I don't quite understand why you don't count games running Rosetta. If it runs, it runs, no? Because otherwise there is this weird thing where you say that Bootcamp makes Macs more gaming-compatible but games actually running on macOS without any additional software or configuration are somehow not relevant.


Besides, that list is woefully incomplete. Where are BG3 and Resident Evil (probably the most notable AAA Apple-Silicon games)? Where is Factorio (a hugely popular indy game that sold 3.5 million copies, native since last fall)? Where is Eve Online? Or what about Minecraft, probably one of the most popular games ever? BTW, Humankind has released a native version just this week. And if you count Intel games that are fully compatible with Apple Silicon, you'll get many more newer successful titles that you can play on your Mac. Civilisation, Total War games, Metro, Hades (which won many "best game" awards) and so on. BTW, all of these games work with Metal, either directly or via a third-party engine. And sure, this is just a drop in a bucket compared to the games available for Windows, but it's also very far from your narrative that Metal and Apple Silicon killed gaming on Mac.

 

[maflynn]

Even so, you'll still rely on virtualization, cloud gaming or Crossover.

This is what many in the game industry is pushing for. No longer playing games on PCs as much as paying a monthly fee to play via streaming. this in of itself will create headwinds for the Mac

That's the real issue here. You'll be paying top dollar for something that not only cannot use the hardware to its fullest potential

Look at how popular the Steam Deck is, and how easy that is to get PC games up and running. IMO, this makes it harder for studios to embrace a new unproven platform. Why dedicate hundreds of thousands of dollars, if not millions to embrace Macs when the Steam Deck can be used for little to no money by the publisher. Then there's the consoles, and I'd venture a guess (and its a wild guess) that any mac user who wants to game may very well already own a playstation or xbox.

Apple can fix this by buying a few publishers (EA & UBISoft come to mind).

That will move the needle to be sure, BUT (and its a big but)

Sorry that song came into my mind, so I had to create a meme.

Anyways the big but is that Apple has not bought any game publishers. We can might've should've, could've ourselves till we're blue in the face but Apple in the 3 years of M series processor has not made any explicit moves to embracing gaming in a serious way. I'm not saying they won't in the future, but as it stands The reputation that Macs are not gaming machines has not diminished with the M1/M2

 

[Gudi]

"RISC" and "CISC" were simply two labels for how CPUs used to be designed forty years ago.

No, 40 years ago was when the development of RISC and CISC processors separated forever to form their very own branches of CPUs. Nowadays as Apple Silicon are also 3.5 GHz, 64-bit, multi-cores, it becomes more apparent than ever that RISC chips are fundamentally more power efficient than CISC chips at every performance level.

Nowadays these labels are probably much less useful, as CPU design has evolved and became much more complex.

On the contrary, because of RISC and Metal it became less complex, more efficient and way faster and cooler.

Contemporary concerts are very different from what was important in the 70-ties and 80-ties.

Yeah, ABBA songs are some of the most complex compositions ever, whereas a lot of contemporary music is often stupefyingly simple, created by people without any musical education in their mom's basement.

I believe that instead of relying on imprecise terms like "RISC" and "CISC" to characterize architectures, one needs to study the design details and discuss what the instruction sets have in common and how they differ.

No you don't. Before you understand the difference between birds and mammals, we don't need to talk about what makes animals airworthy. All bird bones are filled with air instead of marrow, which makes all the difference regardless of penguins and bats. You need to understand the fundamentals before you bore us with useless details.

 

[Gudi]

Maybe it's because I just might be a little bit knowledgeable about these things?

What you posses is called "gefährliches Halbwissen" (dangerous half-knowledge). You're so obsessed with the little bit of computer science in which you have a tiny bit of expert knowledge, that you outright deny the rest even exists.

What you are writing are platitudes found on user forums quoting long outdated books and wisdoms about CPU design. None of this stuff has been relevant for the last two decades.

All math is build upon the Peano axioms. We don't mention them every time, because we assume that someone who went to kindergarten has acquired basic knowledge of what natural numbers are. But every now and again someone doesn't understand that you can't have your cake and eat it too.

Nobody is talking about OpenGL. Current industry standard for graphics is Vulkan.

And by industry you mean non-Apple products, nobody on MacRumors cares about.

BTW, as a former active member of the OpenGL community who has been following these things very closely for many many years I have my own theories why we ended up with the current API fragmentation. Short version is: committees suck.

More dangerous half-knowledge. There is no API fragmentation, it's an all-Metal platform from the iPhone mini to the Mac Studio Ultra.

 

[Joe Dohn]

I don't quite understand why you don't count games running Rosetta. If it runs, it runs, no?

For two reasons:

* Rosetta is a temporary solution. As soon as the last Intel Mac is discontinued, Rosetta will be on a countdown for deprecation. And when it IS deprecated, the games and applications that rely on it will stop working. It's similar to what happened to the first Rosetta (what you call Rosetta is actually Rosetta 2).

* Rosetta translated x86 instructions to ARM instructions, similar to Windows ARM's translation layer. Which means it is not native.

"If it runs, it runs" is not good either because it relies on solutions that usually involve using some processing power to translate the game to something that can be processed by the hardware, which means you never use it to its fullest potential.

 

[leman]

No, 40 years ago was when the development of RISC and CISC processors separated forever to form their very own branches of CPUs. Nowadays as Apple Silicon are also 3.5 GHz, 64-bit, multi-cores, it becomes more apparent than ever that RISC chips are fundamentally more power efficient than CISC chips at every performance level.

Humor us then, please. When you write "RISC" and "CISC", what is that you mean exactly?

 

[leman]

* Rosetta is a temporary solution. As soon as the last Intel Mac is discontinued, Rosetta will be on a countdown for deprecation. And when it IS deprecated, the games and applications that rely on it will stop working. It's similar to what happened to the first Rosetta (what you call Rosetta is actually Rosetta 2).

Ok, maybe. But as things are currently none of this matters to the end user. The software starts and runs, and performs well. When (and if) Rosetta is dropped, that software will be unusable. But right now it is perfectly useable.

* Rosetta translated x86 instructions to ARM instructions, similar to Windows ARM's translation layer. Which means it is not native.

"If it runs, it runs" is not good either because it relies on solutions that usually involve using some processing power to translate the game to something that can be processed by the hardware, which means you never use it to its fullest potential.

This is something I don't understand. So because it might run faster or smoother in other circumstances it doesn't count? So mediocre console to PC ports don't count either? And certainly, games running on Steam Deck should count either, since you are running both WinApi and DX12 emulation and neither of these things come for free either, no?

 

[Joe Dohn]

This is what many in the game industry is pushing for. No longer playing games on PCs as much as paying a monthly fee to play via streaming. this in of itself will create headwinds for the Mac

Except not really. If I can game on the cloud, it also means specs don't matter, since all the processing is done by the cloud. This means even a 20-year old computer will game at nearly native speeds, PROVIDED IT HAS A GOOD INTERNET CONNECTION AND CAN RENDER VIDEO AT THE REQUESTED RESOLUTION.

This also means I can game on a very cheap and crappy Chromebook, provided meets the two requirements above, and my experience will be amazing. Why should I then use an expensive computer at all?

Look at how popular the Steam Deck is, and how easy that is to get PC games up and running. IMO, this makes it harder for studios to embrace a new unproven platform. Why dedicate hundreds of thousands of dollars, if not millions to embrace Macs when the Steam Deck can be used for little to no money by the publisher. Then there's the consoles, and I'd venture a guess (and its a wild guess) that any mac user who wants to game may very well already own a playstation or xbox.

Correct. But Apple could have taken a dozen measures to be more receptive to gaming. They thought that Metal would be so amazing that developers would be flocking for Apple Silicon, which is a dangerous thought (for Apple!).

 

[Joe Dohn]

Ok, maybe. But as things are currently none of this matters to the end user. The software starts and runs, and performs well. When (and if) Rosetta is dropped, that software will be unusable. But right now it is perfectly useable.

Because the translation layer that makes it possible can be removed at any time. THAT's the point. Apple is not making any promises that you'll be able to run Rosetta software after they discontinue the last Intel product.

Think of it: I'm an end user ready to buy a new PC. I'm considering a Macbook. If I'm aware Rosetta will be discontinued at some point in time and I know the Windows games I like rely on it, why would I consider a Macbook at all?

Or imagine the following scenario:

My game runs buttery smooth now, but 2 years pass, and either through a software update or a system update, Rosetta is removed. Then, I learn that the game that would run before simply stops working.

I'll be very pissed.

This is something I don't understand. So because it might run faster or smoother in other circumstances it doesn't count?

They cannot count as native because native means the games have been compiled with the native system APIs, and will not be removed. It's that simple. They also talk more directly with the hardware, which guarantees better performance and stability.

So mediocre console to PC ports don't count either? And certainly, games running on Steam Deck should count either, since you are running both WinApi and DX12 emulation and neither of these things come for free either, no?

I'm not sure what you mean. As long as they use the native system APIs, they count. Ports can definitely be rewritten to become native. From the Windows point of view, WinAPI and Direct X 12 are native, and they will not be removed anytime soon.

 

[leman]

They cannot count as native because native means the games have been compiled with the native system APIs, and will not be removed.

Of course they don't count as native. But you seem to be completely discounting them in terms of available software today. I mean, it's entirely possible that Apple will remove Rosetta 2 in five years or ten years, or whenever. But today I *can* run Civilisation that has been compiled for Intel Macs, and it runs better on my M1 than it run on my Intel Mac. So for all intends and purposes this is a game I can play on my system today, without any extra steps or hassle. What might happen tomorrow is a different concern. I mean, Microsoft might spontaneously decide to remove DX11 or ban all alternative APIs from their system as well (they already tried this with OpenGL btw). My point is: isn't it a bit restrictive to disregards software because some future OS or hardware update might make this software incompatible?

 

[Joe Dohn]

Of course they don't count as native. But you seem to be completely discounting them in terms of available software today. I mean, it's entirely possible that Apple will remove Rosetta 2 in five years or ten years, or whenever.

Not "whatever". The whole point is that we don't know when Apple plans to kill Rosetta 2. It could be in 10 years, but it could be 6 months from now or tomorrow.

The most educated guess would be 1-2 years after the Mx Mac Pro is announced. Assuming it is announced this year, it would mean 1-2 years from now, I won't be able to use a Rosetta build of a game that runs now fine in my Mac. It's a pretty big deal for a device that has a 5-year cycle.

That doesn't apply to just games, by the way: all productivity software that runs through Rosetta will stop working. So this could mean the wrong choice leads into a financial loss (due to a loss of productivity).

Also, native builds are important because I'm buying a particular piece of hardware for the games it offers without band-aid solutions. The Apple Silicon has few actual native builds, it means developers don't consider it worthy the time and effort to use their budget to port it to the Mx architecture, which is pretty serious.

 

[leman]

Not "whatever". The whole point is that we don't know when Apple plans to kill Rosetta 2. It could be in 10 years, but it could be 6 months from now or tomorrow.

The most educated guess would be 1-2 years after the Mx Mac Pro is announced. Assuming it is announced this year, it would mean 1-2 years from now, I won't be able to use a Rosetta build of a game that runs now fine in my Mac. It's a pretty big deal for a device that has a 5-year cycle.

That doesn't apply to just games, by the way: all productivity software that runs through Rosetta will stop working. So this could mean the wrong choice leads into a financial loss (due to a loss of productivity).

You are building all this elaborate argumentation based on nothing but suppositions and guesses.

Also, native builds are important because I'm buying a particular piece of hardware for the games it offers without band-aid solutions. The Apple Silicon has few actual native builds, it means developers don't consider it worthy the time and effort to use their budget to port it to the Mx architecture, which is pretty serious.

Why would it mean that? The software runs, is tested and supported on Apple Silicon. Sometimes rebuilding for ARM might be trivial. Sometimes it might be more complicated (some third-party framework that hasn't been ported yet or an over-engineered internal high-performance concurrent library that relies on strong memory ordering). The point is, it works. The mere fact of porting a game to macOS already involves a lot of effort for many titles. The relevant measure of developer interest if they do that, not if they do the full way.

 

[Joe Dohn]

You are building all this elaborate argumentation based on nothing but suppositions and guesses.

Apple themselves said it when they released the Mx Macs that Rosetta was a TEMPORARY solution. My assumptions on when it'll be discountinued are based on the original Rosetta. So, it's a reasonable assumption.

Why would it mean that? The software runs, is tested and supported on Apple Silicon. Sometimes rebuilding for ARM might be trivial. Sometimes it might be more complicated (some third-party framework that hasn't been ported yet or an over-engineered internal high-performance concurrent library that relies on strong memory ordering). The point is, it works.

Because the whole point of Rosetta is that you DON'T need to rebuild your software for ARM.

With most games that use Rosetta, once that build is released, that's it – or else they would already have been recompiled to use Metal and the native APIs at this point.

I'm also talking based on the original Rosetta here. Most games that used the original Rosetta just stopped working after it was discontinued. The developers didn't want to recompile the game natively at all.

The exception would be Crossover Office. They said they are using it because porting Wine to ARM was a very complex task, because ARM is a different architecture.

MAYBE Crossover will move to Box86, but it's inherently less reliable than Rosetta is. You would have to ask them what they plan to do after Rosetta is discontinued.

 

[i486dx2-66]

The most educated guess would be 1-2 years after the Mx Mac Pro is announced. Assuming it is announced this year, it would mean 1-2 years from now, I won't be able to use a Rosetta build of a game that runs now fine in my Mac. It's a pretty big deal for a device that has a 5-year cycle.

What makes you think the timeline is so short?

Apple's big investment in Rosetta2 has already been made, and providing backward compatibility is an ongoing benefit that helps sales. Rosetta 1 was available in supported operating systems for over six years.

 

[Joe Dohn]

What makes you think the timeline is so short?

That's the approximate timeline they had originally announced when PowerPC products were developed back then.

Apple's big investment in Rosetta2 has already been made, and providing backward compatibility is an ongoing benefit that helps sales. Rosetta 1 was available in supported operating systems for over six years.

Yes. Roughly around four years before deprecation, and two years as a grace period.
Once Apple announces Rosetta 2 will be deprecated, Rosetta 2 will be on grace period.
Which means you shouldn't count on it then.

Either way, I wouldn't count on something I know is temporary. Apple was pretty transparent here on this matter.

 

[leman]

Apple themselves said it when they released the Mx Macs that Rosetta was a TEMPORARY solution.

They did? Where?

My assumptions on when it'll be discountinued are based on the original Rosetta. So, it's a reasonable assumption.

Original Rosetta was a very different technology which Apple has licensed from a different company. Rosetta 2 is an LLVM based transpiler which is essentially maintenance-free. The only cost Rosetta 2 is incurring is in form of hardware TSO support.

What's interesting is that every year Apple improves Rosetta 2 support. Like recently they added the ability for virtualised guests to use Rosetta to run Intel software. Given that Apple Silicon is a popular development platform, the ability to run x86 code with the same semantics as an x86 processor can very valuable. I have no doubt that over time the new APIs will be deprecated for Intel builds (and Intel builds will probably be deprecated as well). But there is not much motivation for Apple to remove Rosetta 2 in short term.

With most games that use Rosetta, once that build is released, that's it – or else they would already have been recompiled to use Metal and the native APIs at this point.

Most games that use Rosetta also use Metal. What exactly is your definition of "native API"? Intel programs use the same APIs as ARM programs (API availability and version might vary by ABI or course).

Besides, a bunch of Intel games had issues running under Rosetta (mostly because of different behaviour of Apple GPUs which exposed bugs in the code), and many of those were fixed. So developers are still committed to supporting their games on Apple Silicon, even if they do not provide a native version.

I'm also talking based on the original Rosetta here. Most games that used the original Rosetta just stopped working after it was discontinued. The developers didn't want to recompile the game natively at all.

Sure, and if Apple removes Rosetta 2 Intel games will obviously stop working. There is no incentive for developers to put in extra work. It's just we seem to disagree on the timeline or even the inevitability of this.

 

[i486dx2-66]

You also have to consider that the PPC to Intel transition happened quite quickly - just the seven months of January through August of 2006, and the whole line had been switched over. Apple Silicon, by contrast, has moved much more slowly. Starting in Nov 2020, we're over two years in, and Apple is still selling Intel machines. Just weeks ago, they had Intel CPUs in multiple products still for sale. x86 software (and the need for translation compatibility) aren't two years in the past, they're still very current needs.

 

[Joe Dohn]

They did? Where?

M1 Apple keynote speech, 2020.

 

[leman]

M1 Apple keynote speech, 2020.

I'm looking at the transcripts for both the WWDC and the M1 event and I don't see any mention that this is a temporary solution. I mean, sure, it probably is, but it's not like there is a clear expiration date.

WWDC 2020: Complete Transcript of Apple's Keynote with iOS 14, macOS Big Sur, and More

Apple's first-ever all-online Worldwide Developers Conference (WWDC) starts today with the traditional keynote kicking things off at 10:00 a.m....

www.macrumors.com

Apple Event Live Coverage: Apple Silicon Macs Announced! [Full Transcript of Event]

Apple's virtual "One More Thing" event kicks off today at 10:00 a.m. Pacific Time, with Apple expected to debut its first Mac models...

www.macrumors.com

 

[Joe Dohn]

I'm looking at the transcripts for both the WWDC and the M1 event and I don't see any mention that this is a temporary solution.

WWDC 2020: Complete Transcript of Apple's Keynote with iOS 14, macOS Big Sur, and More

Apple's first-ever all-online Worldwide Developers Conference (WWDC) starts today with the traditional keynote kicking things off at 10:00 a.m....

www.macrumors.com

Apple Event Live Coverage: Apple Silicon Macs Announced! [Full Transcript of Event]

Apple's virtual "One More Thing" event kicks off today at 10:00 a.m. Pacific Time, with Apple expected to debut its first Mac models...

www.macrumors.com

Everyone expects it to be removed at some point. Apple didn't call it "Rosetta" out of nowhere. This was discussed here in this very forum: https://forums.macrumors.com/thread...nary-2-exist-for-and-other-questions.2242369/

I would only trust Apple not removing the API after the discontinuing of the X86 Mac Pro if they explicitly state otherwise.

 

[leman]

Everyone expects it to be removed at some point. Apple didn't call it "Rosetta" out of nowhere. This was discussed here in this very forum: https://forums.macrumors.com/thread...nary-2-exist-for-and-other-questions.2242369/

It is a difference however between "they said its temporary" to "everyone expects it to remove it", isn't it? And it still does not give any concrete or motivated timelines.

Not quite sure what you mean by the "didn't call it Rosetta out of nowhere"? Rosetta is a reference to the Rosetta Stone, which is an artefact that has been used to decipher Egyptian script. You can see it in London. It's very impressive.