In 3 years, 50% of all computers capable of playing AAA games will be Macs

[jeanlain]

You're giving me some very vague methodology to try but without actually giving me any concrete reasons or numbers other than Apple's revenues. Honestly, that doesn't interest me, because it doesn't say anything about how much I'll earn when I write a game.

But I can tell you this much: my games have earned me far more on iOS (and then Android) than the Mac versions did. Ultimately, that was why I dropped the Mac version. And now that Mac can run the same games I've written for iOS, I have even less of a reason to maintain a Mac-specific version.

But we're talking about a completely different type of games, which won't run on iOS devices.
A typical AAA games has a PS version, an xbox version and a Windows version. They usually don't have a Mac version because it has been determined that revenues would not compensate development and support costs.
The question is whether this may change in the future. Even if the Mac version generates much less profit than a console version, say 1 million instead of 30 million, why not do it? That's 1 million $ that you would not earn if you didn't have a Mac version.
However, it's important to ensure that the "lesser" (Mac) version is not detrimental to the most profitable versions, given your limited time and resources. If developing a Mac version means that less effort is spend to the console version, which may reduce the profit down to 20 million, the net result is not good.
But AAA games studios can use porting houses if time/resources are limited.

 

[bill-p]

Yeah, maintaining the Mac version isn't worth the effort. I don't want to upsell myself, but my day job sets my expectation for a per hour dollar amount that I must be getting back from my app sale, and the Mac version is... let's just say consistently less than desirable based on that.

This problem is exacerbated with a AAA game studio where they have a lot of costs to factor in... so even one hour of dev time costs them hundreds of dollars, if not thousands. Is it worth it to maintain a Mac version at that point if they're not getting hundreds of dollars worth of sale per hour? The answer would be a big flat no.

It's not just a matter of gaining $1 million. It's a matter of how long it'll take to gain that. Note that software engineer pay is on average about $100K/year, so if it takes a team of 5 developers and 5 testers to work on a Mac version for a year, it's already not worth it.

P.S.: even bigger game studios like Blizzard can probably "absorb" this cost because their model (online gaming) allows them to gain revenue even after the games are out. But not too many AAA games follow this "pay-to-play" model for good reasons. So... we're still looking at revenue as a problem here.

 

[neinjohn]

If Apple care about gaming, targeting the AAA games' folk using their huge visibility to show enticing, uncomplicated hardware and gaming on their 4K iMac 21,5'' is a golden opportunity since GPU prices skyrocketed. Key is the redesign and presentation, what options they take with CPU-GPU cores balance and mini-LED refresh rate.

Presentation they have to work with 2/3 key studios to port newer games to native and optimised to play on medium in the M1 Macs on the next 4-5 months. Basically what they're doing with Baldur's Gate.

Since they will use mini-LED displays on the next iPad Pro release and the actual model support 120hz, they could/should expand higher refresh rates for all newer mL models.

Give enough GPU cores to the iMac and they have a very simple, very expensive, 4k-capable console like experience at the $1,499/1,899. You open the box, connect the power outlet, download the game and you're ready. You can find it for sell everywhere, no DIY, no Christmas-RGB design.

Use Apple normal modus operandi on the iMac 21,5'' non-4K, make it fanless with a upped M1.

Casual gaming is more fragmented depends more on "enthusiasm" from a lot of developers looking at the Air, eyeball number, interest on the Apple Store/Arcade as business paths I guess.

 

[JouniS]

Tldr: Within 3 years, basic math suggests Macs will be 50% of all computers sold yearly capable of playing AAA games. It will finally make financial sense for AAA developers to port games to MacOS.

That may be possible on the hardware side, assuming that the future low-end Apple GPU is at least 50% faster than today. With M1 hardware, Apple would need 12 GPU cores to match Xbox Series S and 32 cores to match Xbox Series X and PS5. Also, because Series S has 10 GB memory and the other current-generation consoles have 16 GB, Macs with 8 GB will not be AAA capable for long.

In the normal cycle of AAA gaming, hardware requirements jump shortly after the release of a new console generation and then remain almost stable until the release of the next generation. Or, in other words, the games that run on previous-generation consoles will no longer be considered AAA after a year or two. Such games feel limited and look bad next to the true AAA games designed for current-generation hardware.

On the market side, there are two main obstacles to the adoption of macOS as a mainstream AAA gaming platform. The first is backward compatibility. Gamers often return to their old favorites, but Apple can't guarantee even 10-20 years of compatibility.

The second obstacle is pricing. Thin laptops will never be a cost-effective choice for AAA gaming, and Apple is making it even worse by restricting the use of higher-end GPUs to the most expensive MBP models. Unless you already have a high-end MBP for other reasons, you won't get as good gaming experience as with PC hardware.

I would expect that macOS ports of AAA games become more common (but not widespread), as low-end Macs now have decent GPUs. Those ports would be played by people who already have a Mac for other reasons. People looking for new AAA gaming hardware would still skip Macs, because PCs and consoles offer better GPU performance for the same price.

 

[Mcdevidr]

Has anybody seen the Notebook check review of the M1. The real world performance in gaming is not at the level of a 1650 it just does not show that at all. The performance is impressive for what it is ill give it that. I played 4 hours of a game in multiplayer on a fanless computer on a game that can cause over heating on a ton of laptops. So while its impressive the performance in gaming seems more like MX350 and maybe at topping out and native support a 1050 mobile version.

 

[SegNerd]

I don’t think game developers care about processing power as much as they care about demand/money/potential to actually make money. So even if Macs have ten times better graphics performance than PCs, if they are still only a small percentage of the market, game developers will still focus mainly on Windows.

Having said that, Apple Silicon does increase the incentive to develop software in a different way, which is that it can run iOS software. So if you are counting mobile games, then Macs have indeed just become a much larger gaming platform.

 

[quarkysg]

From what I read from forums, news articles, etc., most entry level Macs (i.e. MB Air, MB Pro 13", Minis) are intended for education use. These group of users are the ideal audience for games. The cost of a game is probably cheaper compared to a Friday night out with friends/classmates. In 3 years' time, all Macs sold in that year will be capable of running graphic intensive games, including M1 Macs launched last month.

iOS developers are already familiar with the Metal APIs due to the large market iOS provides. So tapping that pool of developers for macOS would be an avenue for development resources. And from what I can see from Apple's WWDC videos, the development tools are quite advanced, especially in terms of performance optimization.

I don't think Apple wants to be king of the hill in terms of performance (be it CPU, GPU, ML, etc). They probably have a product vision that current offerings from Intel, AMD and Nvidia could not provide.

Apple's Mac offerings targets different markets. IMHO, the entry level Macs' target audience are the group that would care to play games, but unfortunately these Macs does not have the necessary grunt to play games. Apple Silicon Macs would likely change this. iMacs/Pro, higher end MB Pros and Mac Pros are intended for folks who uses the Macs for serious work, so the majority of folks buying these Macs will not bother with games.

I for one am optimistic that the gaming landscape for macOS will improve going forward.

I guess we'll have to wait and see how it evolves.

 

[Argon_]

If Apple care about gaming, targeting the AAA games' folk using their huge visibility to show enticing, uncomplicated hardware and gaming on their 4K iMac 21,5'' is a golden opportunity since GPU prices skyrocketed. Key is the redesign and presentation, what options they take with CPU-GPU cores balance and mini-LED refresh rate.

Imagine if Apple drops the price of the M1 Mini once the M2 releases to $599. Pair that with a 120Hz monitor, and that's a budget gaming rig, if there were the software side support.

Currently, marketing the Mini for gaming would make the most sense, because of high refresh rate monitors.

 

[JMacHack]

Has anybody seen the Notebook check review of the M1. The real world performance in gaming is not at the level of a 1650 it just does not show that at all. The performance is impressive for what it is ill give it that. I played 4 hours of a game in multiplayer on a fanless computer on a game that can cause over heating on a ton of laptops. So while its impressive the performance in gaming seems more like MX350 and maybe at topping out and native support a 1050 mobile version.

Running through Rosetta will do that. A native port would have more performance.

Imagine if Apple drops the price of the M1 Mini once the M2 releases to $599. Pair that with a 120Hz monitor, and that's a budget gaming rig, if there were the software side support.

That's a pretty big "if" I'd say. I don't see the Mac gaming landscape changing no matter how powerful the M-series chips get. Frankly, the core "gaming" audience is Mac-hostile anyway, so who cares?

OP is the same guy who made crazy predictions about the sales of Macs after the M-Series switch.

 

[Ethosik]

Apple computers have ALWAYS been capable of playing AAA games... the issue is that developers write exclusively with DirectX in mind. So even if you could prove that Macs had the fastest CPU and the most powerful GPU on the planet, you're still not going to convince developers to write for the Mac.

Hell when Apple transitioned to the INTEL chip, it and all PCs still had lots of games being writing with OpenGL. It was at that point in time when the potential for Apple being able to get the same games for the Mac as for the PC was a reality.

Guess what. It never happened. OpenGL depreciated. Apple switched to Metal. And the divide between Macs and PCs have never been greater than it is now from a developer's stand point.

It's not the GPU or the CPU... it's the underlying code being used. Microsoft will never abandon DirectX and Apple will never abandon Metal.

You want to play PC games... buy a PC. It's that simple. And that is exactly the mindset of the developers out there too.

This. Its all about Direct X. And Direct X is far more efficient. Playing WoW on my macOS vs Windows on the same hardware, Windows wins.

 

[Gnattu]

Direct X is far more efficient

Not quite.
Lots of games ported to mac still uses Direct X API, but introduced some translation layer to translate DirectX calls to APIs supported by macOS. And that translation layer is the reason of performance loss.

 

[Ethosik]

You forgot to add the Apple tax in your equation.

PCs are cheaper and way more available globally. Then there's NVidia. While the M1 may compete with lower to mid range AMD offering, they're just not there yet when it comes to NVidia GPU.

Macs are not really more expensive. People play spec wars but you also need to factor in build quality, battery life, screen quality, resolution, if it maintains power without being plugged in, macOS itself and more. Its not just Intel vs Intel and RAM vs RAM when you pay for Macs.

 

[Ethosik]

Not quite.
Lots of games ported to mac still uses Direct X API, but introduced some translation layer to translate DirectX calls to APIs supported by macOS. And that translation layer is the reason of performance loss.

Uhhhh Mac doesn't support Direct X. And you just proved its not efficient because its being translated? What?

WoW on macOS gets less FPS than WoW on Windows on the same hardware. Due to Direct X and efficiencies and optimizations with Direct X.

 

[Gnattu]

Uhhhh Mac doesn't support Direct X. And you just proved its not efficient because its being translated? What?

Because Mac doesn't support Direct X and developers don't want to rewrite all their codes to support Mac, they will use translation layer to translate Direct X API calls to macOS native API calls like metal.

Direct X + translation is always slower than Direct X native, it is easy to understand.

I'm not "proving Direct X is not efficient". I'm proving "games on the platforms are not equally optimized"

By the way, DirectX is closed source and is not officially supported anywhere but Windows, it's not Apple's fault to not support DirectX.

 

[Ethosik]

Because Mac doesn't support Direct X and developers don't want to rewrite all their codes to support Mac, they will use translation layer to translate Direct X API calls to macOS native API calls like metal.

Direct X + translation is always slower than Direct X native, it is easy to understand.

I'm not "proving Direct X is not efficient". I'm proving "games on the platforms are not equally optimized"

By the way, DirectX is closed source and is not officially supported anywhere but Windows, it's not Apple's fault to not support DirectX.

So you are saying Direct X is not efficient when you are mentioning examples that use a translation layer? That is not how it works. Look at Windows and Direct X there. It is efficient. FAR more better than ANY Mac game port as booting into Windows with the same hardware usually always produces 20% more performance right out of the get go.

I am not understanding your disagreement with me that Direct X is efficient? Throwing in translations layers does not prove Direct X is inefficient.

 

[Gnattu]

So you are saying Direct X is not efficient when you are mentioning examples that use a translation layer?

Come on, I'm saying that on macOS, lots of games runs slower BECAUSE they are using a translation layer instead of write code for APIs like Metal directly. You can not compare the results using native api vs translated api, this looks like saying "arm64 native codes runs much faster than rosetta". Correct, but meaningless.

 

[Ethosik]

Come on, I'm saying that on macOS, lots of games runs slower BECAUSE they are using a translation layer instead of write code for APIs like Metal directly. You can not compare the results using native api vs translated api, this looks like saying "arm64 native codes runs much faster than rosetta". Correct, but meaningless.

Does every macOS port do this? Last I checked, WoW offered Metal I believe, or one of Blizzards game did.

 

[Gnattu]

I am not understanding your disagreement with me that Direct X is efficient? Throwing in translations layers does not prove Direct X is inefficient.

I'm disagree with you because your observed fact may not be a result from a more efficient API, but the penalty of a translation layer, even less optimized code.

 

[Ethosik]

I'm disagree with you because your observed fact may not be a result from a more efficient API, but the penalty of a translation layer, even less optimized code.

Are you saying every single macOS port uses the translation layer? If so then I agree. I play a lot of games from Tomb Raider, WoW, Starcraft, Heroes of the Storm, Diablo, and many more that have Mac ports. And every single one of them gets about 25% improvement just by running in Windows.

 

[Gnattu]

WoW offered Metal I believe

WoW on macOS should be at least on par with Windows since 8.0 due to the Metal 2, and I've seen claims that macOS outperforms Windows after 8.0.

 

[Ethosik]

WoW on macOS should be at least on par with Windows since 8.0 due to the Metal 2, and I've seen claims that macOS outperforms Windows after 8.0.

I am still getting 25% better performance on Windows.

 

[Gnattu]

Are you saying every single macOS port uses the translation layer

I can not verify each title's implementation, but from Apple:

more than 148,000 applications use Metal directly, and 1.7 million use it through high-level frameworks, as of June 2017

This translates to less than 10% of the Application supports metal is using it directly. I don't know for each "specific title", but the chance to have an App writes for metal directly is very low.

 

[Gnattu]

I am still getting 25% better performance on Windows.

The conflicting results means there is something more than graphics API to determine the relative performance.

 

[Ethosik]

The conflicting results means there is something more than graphics API to determine the relative performance.

I definitely hope you are right! I prefer macOS over Windows 1000% and if Metal is almost as good as Direct X that is a good thing! The ONLY reason I use Windows for is gaming and Visual Studio. Visual Studio for Windows is still light years ahead of Visual Studio for Mac.

 

[Gnattu]

Visual Studio for Windows is still light years ahead of Visual Studio for Mac.

I have to say, the Windows version is not "ahead", it's a "different thing". The one for Mac is a reskined Xamarin, and only useful for .NET and nothing else. I don't even think Visual Studio on Mac will ever get the functionality that Windows version has because they are developed for different audiences. There is no doubt the Windows version is way more powerful, but I don't think Microsoft has a plan to let the Mac version "catch"