The technical considerations are uninteresting here: all of Blizzard's titles are optionally built to use OpenGL, letting them avoid the golden handcuffs of DirectX & Friends. So if porting is cost-effective, why would they do it when virtually all of their revenue comes from Windows systems?
I say: gatekeeper-free subscription gaming appliances.
There's no evidence the total Windows PC experience is getting any better or cheaper for the uninitiated, there's no evidence that Playstation or Microsoft will bend over and allow third party platforms on their consoles, and all of the new-era Desktop App Store UXes force significant technical and billing restrictions that would eat at Activision Blizzard's subscription and in-app business model and reduce their control over the game experience.
And I don't think this is a bet that desktop Linux/x86 is going to be relevant for gaming. It has no inherent advantages over classical Windows.
So I think this is a gamble that new-era quasi-consoles based on Android+Linux/ARM+x86 are going to succeed to some degree (be it the Ouya or Steam Big Picture or some Samsung Generico-Colossus that can spit out WiDi or whatever).
A keyboard, mouse, and controller capable Linux platform with no overarching gatekeeper and adequate gaming horsepower is going to be extraordinarily cheap in the near future. If Blizzard Activision gets their ducks in a row now, they'll be ready to jump into the first one that offers ten million users, a reasonable hardware target, and a bullshit-free content delivery mechanism.
Right, they obviously don't want to wed themselves to any particular "quasi-console", but they want to be ready if the "quasi-console" thing really picks up steam. That's how I see it.
I don't, but let me talk louder and slower in case I'm being unclear.
The goal is to sell subscriptions and in-app upgrades to game softwares of exceeding depth and breadth that are operated primarily with a keyboard and mouse.
The Windows PC has arrived at an unsubsidized price floor of about $300-$400 and is unlikely to go lower with Microsoft charging perhaps $40-$80/unit depending on who's licensing. This market sells ~350M units a year with a total installed base of 1.25 BILLION, yet unit sales are declining, desktop Linux is doing nothing to change this, and Blizzard's flagship title is in decline.
The Linux PC cannot go lower than $300-$400 despite essentially $0 licensing costs because virtually nobody is selling or buying raw Linux PC hardware. So the platform is drafting on boutique sales, whitebox sales, and repurposed Windows PCs while still offering no benefits above and beyond standard Windows PCs. At best, it's (currently) worse.
The classical consoles are off limits. Nintendo has no coherent Internet strategy evolving and a severe lack of onboard storage. Sony has allowed a single third-party platform which has no billing support, subscription or otherwise. Microsoft has no third-party platform support. All of these consoles come with significant licensing expenses and do not support software subscriptions. All of these consoles require major rework when porting existing PC titles due to custom CPUs, GPUs, and input frameworks.
Mobile as we understand it is either completely off limits (iOS) or unaccustomed to keyboard and mouse input (Android).
So if you are in the business of selling subscriptions and in-app upgrades to game software that requires keyboards, mice, and untrammeled billing pathways, what are your options for addressing new markets?
There is one: quasi-consoles. Stationary (TELEVISION OR MONITOR OVER HDMI) Android (LINUX) Consoles (MOUSE, KEYBOARD, CONTROLLER SUPPORT) with untrammeled billing pathways (BATTLE.NET) at dramatically lower price points than the existing options ($100) that will enable less-engaged gamers (ME) easy one-click access to pre-existing titles (WORLD OF WARCRAFT).
And in order to do that you first need your titles running on a Linux stack, addressing OpenGL hardware.
Which is what Blizzard Activision is (allegedly) doing.
I'd imagine that Valve would use their own Linux Distro for their Linux Steam client. When they make the GabeCube or whatever, it will run Linux with the Steam Client for Linux and play in TV mode for TV sets. I'd imagine it would also have Netflix, Hulu apps and other stuff added as well to compete with other video game consoles.
Since a lot of classic video games are DOS based, they'd just have to use the Linux version of DOSBox to run them, which shouldn't be too hard. In Windows games like Doom, Dark Forces, etc run in the Windows version of DOSBox when bought from Steam. I'm imagining Steam would also port their Sega Genesis emulator to Linux and other things as well.
I'd really like to see the Linux based GabeCube or whatever they call it (SteamTV?) that doesn't require a Windows license and can play most video games from Steam.
It has been nicknamed "SteamBox" by the rumormill.
Also, I hope they don't roll their own distro. Customizing an existing distro seems like a much better idea. Considering Steam Linux is currently targeted at Ubuntu, I'm guessing that's the distro of choice.
Good analysis and a good explanation of why I am so bullish on Android. It's going to be much bigger than phones and tablets because it's going to become the go-to application stack for anybody that wants to build a new special-purpose device.
Although they might not use Linux (a mistake in my opinion) rumors say Sony will use full native OpenGL for PS4, and coupled with x86 hardware, it should make it relatively easy for game developers to make a PS4 game and then port it to a Linux machine.
I say: gatekeeper-free subscription gaming appliances.
There's no evidence the total Windows PC experience is getting any better or cheaper for the uninitiated, there's no evidence that Playstation or Microsoft will bend over and allow third party platforms on their consoles, and all of the new-era Desktop App Store UXes force significant technical and billing restrictions that would eat at Activision Blizzard's subscription and in-app business model and reduce their control over the game experience.
And I don't think this is a bet that desktop Linux/x86 is going to be relevant for gaming. It has no inherent advantages over classical Windows.
So I think this is a gamble that new-era quasi-consoles based on Android+Linux/ARM+x86 are going to succeed to some degree (be it the Ouya or Steam Big Picture or some Samsung Generico-Colossus that can spit out WiDi or whatever).
A keyboard, mouse, and controller capable Linux platform with no overarching gatekeeper and adequate gaming horsepower is going to be extraordinarily cheap in the near future. If Blizzard Activision gets their ducks in a row now, they'll be ready to jump into the first one that offers ten million users, a reasonable hardware target, and a bullshit-free content delivery mechanism.