January 18, 2013 in Editorial
The word on the street seems to be that Linux is set to be a commercially viable gaming platform. And the way it looks right now, this might actually prove to be true. There’s still some uncertainty amongst both developers and gamers though. At the core of a lot of it is this: Developers are hesitant to make games for a market as small as this, and gamers are hesitant to adopt it as a gaming platform because there are so few games for it.
I asked Ethan Lee (@flibitijibibo) if he had anything to add to what I wrote, so I have included his view on the various topics.
Right then, here’s a list of things to consider, in no specific order:
Linux desktop users might already be Windows gamers.
Yeah, so? This is actually quite a loaded little bit of info. It requires little research to verify and says a lot about marketing strategy. With so little to choose from, many Linux desktop users have kept a Windows partition around for gaming. For some it’s the only reason that Windows partition exists.
If you want Linux gamers to buy and play your game on Linux, then make sure to release it for that platform on day 1, or at the very least giving them a clear idea of how long they have to wait for a native version. Because if you don’t, they likely just end up buying it on Windows.
What’s so bad about that? Well, for one, you get a poor representation of what the market split is. Your sales info will tell you that the Linux version sold hardly anything when you finally get around to releasing that.
You might think that you are making and selling games, but indirectly your task is also to nurture a market.
Case in point: I have been keeping an eye on Crusader Kings 2 in Steam/Linux’s Top Sellers list. The newly-released “Republic” expansion seems to be almost consistently higher in the list than the base game. To me this says that many Linux players of CK2 already own the game, previously bought on Windows, but now that they can play it in their operating system of choice they just pick up the new expansion/DLC for it.
Flibit: So here’s something I’ve found in my experience: When you sell games via a cross-platform network, particularly Steam with SteamPlay available, a Linux/Mac user never thinks of buying that Windows version on Steam as buying the Windows version; they’re really looking at it as an investment in the future port. Like right now: The Cave probably got a ton of sales when they announced OSX/Linux support, even though right now it’s not marked for Linux.
Communication with your customers is important, but it’s a two-way street; you cannot have genuine communication in any aspect of game development unless the customer is able to provide feedback.
Whether or not you release a Linux game on day one, it’d be great if there were a system that allowed customers to explicitly mark which platform they intend to play it on. Whether it’s “Linux natively, when they port it” or “screw it, I’m playing it in Wine,” having this information would be really good to have.
We kind of get these stats from Steam, but only when the platform support is there. I don’t think we have _any_ Mac/Linux stats for Waveform before July 22, even though we were openly talking about port work as far back as April. The best I’ve got is people telling me they bought it so they could download it as soon as it was out, or so they could use the content to run my test binaries.
Talk to the users as much as possible, but provide a system to let them talk back.
Steam dedicated gaming hardware.
This is a key element in market growth. It’s Linux gaming for non-Linux users. However, much like the release of any new console, potential customers are going to care less about what OS it runs and more about what games they can play on it.
There will be some confusion. People will think that “Steam” means that their entire Steam library will be available for it. Far from it. At the moment there are just shy of 100 available games, with a similar amount on the immediate horizon. It’s a good list, but not good enough to convince many potential buyers.
Despite Valve’s popularity and track record, it looks like many studios are still hesitant to jump in. We’ve heard very little from Valve about what their plans are to get more studios/publishers to jump in, and we can only hope they have something up their pipe.
I suspect most people won’t mind if a subset of the games run via a WINE wrapper, tailored to run on the system, as long as their playing experience is not any worse than they would have had playing on a rival console or their own PC.
I can think of a good reason to buy this hardware regardless of how much you can game on it though. It’ll likely be subsidised. It can be used as a regular PC and you can install any OS on it, so if you want a decent small sized PC at a discount, this is it.
Sony used to let you install alternative operating systems on the PS3, but they did a 180 on that and have made this very difficult. This is money talking. People buying PS3s to build super-computing clusters won’t be playing games on it, and therefore Sony won’t be recouping the money of the hardware they sold at a loss.
With any luck, Valve will remain true to their initial promise of a platform that’s not locked down.
Flibit: This is one I was thinking about earlier this week, and while I think the technical stuff can be talked about to death (read: “Sure, the boys in Ryan’s lab can make it hack-proof. But that don’t mean we ain’t gonna hack it”), but I found the business end to be far more interesting. Not really the “omg the implications of the Steam box will end the universe!” part so much, but how the non-technical people running the industry will look at it.
Here’s the thing: when a console comes out, a publisher yanks the leash that’s carrying their developer and they go “Hey, you: you’re making that game thing for this magic box thing. I dunno a thing about how it works, you figure it out.”
We, the customers, or just gamers, know it’s Linux. But as far as the people in the board room are concerned, you’re just programming for another console.
360 SDK? PS3 SDK? `sudo yum install gcc-c++`. Same crap, isn’t it?
Because of how GNU/Linux works (in a legal sense), it’s the only desktop OS that we can take and re-fit it for consoles. Valve could never _ever_ get MS to greenlight something like this, and obviously Apple would rather ship Apple TV or the Mac Mini themselves while pushing the App Store.
So here we have a console that devs build for like any other console, except that the OS is actually a desktop OS too! This is how Valve intends to get people making games for Linux. They don’t have to convince people to develop for the OS, they just have to get people to develop for the console. Same thing, hugely different impression to their potential partners.
Just do. Don’t play wait and see.
It’s like a party full of introverts. You want to dance, but nobody else is dancing. So you wallflower along with everyone else. It’s a shit party. But if someone makes a move and just gets out there shaking their arse, then you will get others following. The more people see a reason to be there, the more they will dance. And the more people dance, the more people will feel comfortable doing so.
Flibit: Even more important than being the brave one is simply not giving a damn about what the others think. Sometimes you’ll get on the dance floor and begin to walk the dinosaur, and nobody will join in. They may very well point and laugh. But if you end up having a better time than they do, what does it matter? “Oh, sorry, I can’t hear you laughing through the soundproof walls that I was able to make for this money fort I built!”
Serve the environment, and the environment will serve you.
A game like Crusader Kings is not likely to be very big on the Steam console, being very much a “PC” game. Some games will do better on the console than on the desktop. Both sides are Linux, but they are two distinctly different groups with a bit of crossover. But both groups need to exist, and the more of one group exists, the more it makes a case for the gaming platform to exist, and by extension feeds the other group.
It’s the party again. Some want to slow dance, some want to flail their arms about like the kraken rising from the depths. It’s important that there is a dance floor, that there is music and people participating.
If your game is on Steam/Linux, but perhaps not selling too well, either because it’s a game that has been around for a while, or just because it has not had much exposure, it’s still valuable for it to exist within that ecosystem. It keeps the party going. And later on when a sale comes along and your game gets included in an indie bundle or something similar, you’ll be glad.
Flibit: I think my favorite ports during the time that I was getting all these ports done were the ones that had already been released for Windows ages ago. Blueberry Garden and Eversion were released in 2009 and 2010 on Steam, respectively. Really old in the world of games! However, I liked them both enough to want to port them anyway… they weren’t huge sources of income at all, but my customers have looked at those games again, all these years later, and my clients have definitely noticed this. Like an expansion pack, fixing platform support is a good way to get everyone to pay attention to your game again. Getting that attention all at once with a simultaneous release? Well, look at The Cave again…
If cross platform development is hard for you, you’re doing it wrong.
I almost feel like I don’t have to explain this one. I don’t feel that “making games for Linux” should be the final goal here. The key is adjusting your development strategy so that your third party tech and toolchain does not get in the way of your ability to deliver to multiple platforms.
We’re living in a time of change, and it’s important for developers to remain agile. That is all, kbai!