Why Game Developers Should Support OS X and Linux
kevind23 writes "Although Mac OS X and Linux have a small (but growing) market share, Jeff from Wolfire Games argues that supporting non-Windows platforms can lead to a huge increase in game sales. Using their popular game Lugaru as an example, he shows how less-popular platforms, or more specifically, their userbase can be a powerful advertising force. This can lead to a dramatic increase in popularity and exposure, which usually means a large boost in overall sales. The short article is an interesting read, especially for those working in game development and sales."
I remember it being drilled into my head over and over... develop for new hardware instead of old hardware, do everything for the expensive crowd because people who don't spend money on their hardware are less likely to spend money on software. This might be an outdated school of thought, but I'd say it goes double for Mac users. They're really expensive, and especially nowadays they're taking on this image as a trendy status symbol instead of a tool to do work with. Another things Mac devs have going for them, there is a lot less competition. If you would say that Macs don't have enough games out for them, then that translates into a niche to fill for aspiring businessmen.
Why, this is the perfect place to advertise the Linux Installers for Blizzard Products Petition! I believe that if Blizzard supported Linux for its upcoming titles, it would change Linux gaming forever.
But I thought that article trivialised the whole affair and offered very little evidence for the point, bar a spectacularly presented pie chart. One publisher made money from a game. Not quite the smoking gun.
One thing that is true is that there is a lot of respect and word of mouth thrown the way of a good game with native linux client. That would of course diminish if there weren't so few quality games supporting it, of course.
I also find myself wondering whether this game Lugaru is an opengl game, keeping migration costs down.
I record my sleeptalking
DirectX forms a very small part of any well designed game. Everything would be abstracted for portability, you think the PS3 supports DX?
Let's just go through the thought process of porting a game that supports Windows to Linux MacOSX, starting from a DX only codebase.
It would be trivial to support OpenGL as a 2nd renderer as well as D3D because, as I said, games are designed for portability, but as you pointed out that's more maintenance.
But then why keep D3D? OpenGL is portable and runs on Linux, Windows and OSX so the logical decision would be to ONLY support OpenGL, suddenly the game becomes more portable.
Then there are the other things that DirectX does that need to be duplicated for other platforms, for example input, sound etc. The logical choice would be to use, I dunno, some libraries that already took care of the work, like SDL (windowing, input and events) and OpenAL (sound).
But wait. If you use SDL + OpenAL then suddenly the game runs on all platforms... then what's the point of a DX version?
The point I'm getting at is if a game developer wanted to support the 3 main PC platforms they could do with the same amount of development work. The reasons they don't are:
1. They already have a whole DX tool chain built on Windows and with the blessing of Microsoft. It is a risk for them to change their whole process, what if it doesn't pay off?
2. There WILL be more testing required. Chances are things would work the same as all platforms but they'd still have to test that.
There are of course some advantages to writing for more platforms:
1. Compiling your code with more than one compiler is good practice because it flags up bad code that your original compiler allowed erroneously
2. Parts of the code that aren't abstract enough will be flagged up pretty quickly.
Anyway I'm waffling. The point is, the studios won't change until the increase in market share makes up for the change in their development processes.
strange question, shouldnt I know the answer myself since I've been using all three OSen for ages myself? (Typing this on an Ubuntu desktop)
But it's been quite some years now that I last mastered a win/mac CD (it still had OS9) and I never did one for Linux before.
On the other hand my own computer usage has so much shifted to a net focus that I hardly ever install and run a CD myself anymore. And if I do this at all, it's always on win.
So, win is easy, there will be an autorun.inf with a link to an icon and a link to some autorun.exe or whatever.
On Mac, I'd expect the CD to appear with a large friendly icon, a window opening on double click with more large friendly icons that make it very clear what to do (i.e. drag the application onto the application folder alias). No autorun here.
On Linux? I have no idea. From my own usage pattern I don't expect the stuff to be on a CDrom in the first place, it's either in the repositories of my distribution or in a .deb/.rpm dnl'ed from some url or I got a tarball and have to do the ./configure / make / make install - dance. I don't think I ever opened a "commercial" CD intended to be used from Linux (with the exception of install discs). Autorun? - Gott bewahre! Rather a README, may be an install.pl ...
Now there should be sites discussing that question, design guides, style guides, best practices. No way that I'm the first one pondering about how to make a CD look just right on all three OSen - but google drowns me in a bazillion of unrelated pages. Which is why I turn up here with my question, hoping that some of you keep a link or two in their bookmarks to help me find my way.
605413? Yes, it's a prime.