That's part of the issue. But again. Even when OpenGL is used, performance on OSX in games is generally significantly behind performance on a similar hardware configuration with Windows using OGL.
OpenGL on Mac - 2010 spec OpenGL 4.1 with a few 4.2 extensions.
OpenGL on Windows - OpenGL 4.4.
It has potential, but I'm not putting too much into that until we see some high profile games using it.
Metal has only recently become a technology cornerstone in OSX since October 2015 the gestation period for "high profile" games is a few years so it's not going to happen overnight. But I assume you saw Epic on Apple's stage showing Fortnite running on Unreal 4.3 running on Metal so getting support into engines may make it happen sooner. Apple can only provide tools an frameworks - developers need to make it happen. The Metal API will gain support and traction just from iOS - Apple were smart to leverage this to push OSX forward.
Blizzard are still committed to releasing on OSX, but they seem to be starting to lose interest. Whether due to lack of sales or difficulty of porting.
I don't want to get too O/T but what have Blizzard said or done (or not done) to make you think this?
Valve are more interested in OSX as a cross platform target in addition to Linux, so they are unlikely to use Metal. Vulcan will have potential here, but we'll have to see how that plays out. Of course, that all depends on whether Apple will support it or not. And current indications are that they either won't be supporting it or aren't in a hurry to support it.
Valve's priorities are to make games accessible to as many people as possible regardless of the platform to the extent of building their own platforms: SteamOS and SteamMachines. Valve are also pushing Vulkan which will likely never make it to Mac unless AMD or Nvidia do this work themselves (no chance) although it's not inconceivable that Steam would port Vulkan to Mac to help developers because Valve are pushing Steam as a one-stop-stop for developers and customers.
I don't see Valve (or any other developer) investing in a dead technology (OpenGL) over a developing technology (Metal). Metal, Vulkan and DirectX12 (and Mantle, RIP) are intended to solve the same core problems and most high profile games in development will be using close-to-the-metal APIs. This development is actually the best thing that could happen for non-Windows PC gamers - and this is another opportunity for Microsoft. People want to play Windows games but they don't want Windows. Even if you give Windows away for free, you won't find many people who like to switch operating systems to play games. What people really want is Direct3D.
The problem there is that Microsoft would still be reliant on Apple to provide performant drivers etc. for the virtual OS and/or Microsoft provide them for hardware that doesn't have a Windows driver.
That's a strawman if ever I saw one. Everybody using or developing on Mac is reliant on Apple providing good drivers but if you're rendering graphics in a native Direct3D virtual machine then all you need OSX to do is take that target rendered rectangle and put it onscreen. Everything else is done by Direct3D - that's the whole idea.
It also abstracts things farther away. Without as much control performance will suffer.
What abstraction? There is no abstraction because DirectX is running native in a VM. This is so much faster than Wine-wrappers deconstructing DirectX calls, emulating a fat API pipeline and translating everything into ancient OpenGL on the fly.
Microsoft are unlikely to feel the investment is worth it. iOS and Android are worth it due to the huge install bases.
iOS and Android are not running on the same hardware that desktop Windows is. How would that even work?
OSX makes sense for high margin products (like Office), but lacks the install base to make Microsoft want to invest significant money into low margin products.
Where is the significant investment? Microsoft own a host of virtualisation technologies which already support Mac. This is a sunk development cost. Microsoft will continue to develop DirectX for Windows. This is a sunk development cost. Microsoft already have an App Store and server infrastructure. This is a sunk cost. How much do you think this would cost? Much of what we do at work is virtualised environments and while our environment will be nothing like Microsofts, virtualising workflows and pipelines in virtual machine environments and tying them together is kind of our bread and butter work. It's also how Azure is intended to work. Microsoft have all the pieces.
Even though a hypothetical Xbox Store on OSX would isolate them somewhat from the high risk associated with releasing a game on the platform, I just don't see the ROI being there to prompt investment into it.
Well if Microsoft aren't up to the task of doing it then they shouldn't. But there are a fair few gamers on OSX and Linux who do want to run Windows games on their hardware without having to switch operating systems at levels of performance beyond Wine. Taking a cut from digital sales sold through app stores is like taking candy from a baby. And hell, who says Microsoft have to take the industry standard sized piece of candy. If they're providing a service which makes publishers games accessible to Mac and Linux who otherwise wouldn't have access to them without Windows or translation technology, take a bigger cut. And as I said before, this virtualised environment is where they were heading with Xbox One's architecture - this is the next step.
Perhaps it'll become a viable option if the Windows/Xbox store takes off in spectacular fashion and becomes hugely profitable and greatly desired by users thus making a foray into OSX potentially profitable. In other words, if it becomes a serious competitor to Steam, then maybe, but even then? Doubtful.
Microsoft sitting on it's arse only trying to expand after other markets become successful is arguably a big part of why they are where they are - in markets with little growth potential. They lived off Windows and Office licensing for more than a decade and you don't become successful by being complacent and entering a market if it looks profitable after its been established.
Direct3D remains one of Microsoft's most valuable gaming technologies but Microsoft's lack of incentive or willingness to share and license it means it has begun to become marginalised. Things are only of value if you allow people to buy them and increasingly in this day and age, on their terms. Bless AMD for kicking this off with Mantle but the avalanche of thin-graphic-APIs is now unstoppable. The thinner you make the API, the less there is to differentiate (and add value) to how you plumb this game code into this GPU hardware. I don't see this trend reversing.
Games
are a growth market. The division of where dollars are spent shifts but every year the revenue from gaming rises to crazy new levels which is why Microsoft should want their licensed gaming technology be on as many platforms as possible and investing in technologies that make it easy to jump to new platforms. The sooner you get in and start doing it, the more experience to gain to work out the kinks. If you launch a new product with kinks in an already-developed market, you're going to get eaten for lunch. Azure is a good server platform but it's never going to displace AWS and Google. Windows Mobile is a good OS but it's never going to displace Android or iOS. Zune was a good music player but it was never going to displace the iPod. I would have hoped by now that Microsoft will have learned their lesson about leaving things until later. Steam is currently the 800lb PC gaming gorilla but there are things it can't do in terms of serving customers, particularly Mac and Linux customers, what they want. There is money there and Microsoft already have the tech to help them collect it.
It would also be the logical next step to the architectural technologies chosen for Xbox One. Architecturally, the Xbox One is a solid platform.
