Microsoft announces XNA for homebrew, score 1 for my prediction skills

Over the last few months I’ve pieced together various snippets and hints from the web to come to the conclusion that Microsoft’s forthcoming XNA platform, specifically the XNA Framework version, would be available to home-brew developers and let them develop on the Xbox 360 — the first official home-brew since the PlayStation 1’s Net Yaroze!

Indeed I’ve been telling people in IRC and on forums that I believe this to be true. One friend, after messing with Managed DirectX2, told me if I was wrong about this I was in trouble. I went as so far to claim that it’s already tested and GarageGames Marble Blast Ultra was an XNA Framework title.

Today Microsoft announced at their Gamefest conference the XNA Game Studio Express — a free download due this “holiday season” (a beta is out 30th August) — that lets anyone create XNA Framework based games on their PC which other people can run, on their PC.

The real icing on the cake however is that for $99 USD a year you can join the “creators club” and run home-brew XNA titles on your 360 — your own or other peoples.

Check out Microsoft’s XNA Developer Center and their XNA FAQ for the low-down.

Phew, seems my reputation is in tact.

[)amien

4 responses to Microsoft announces XNA for homebrew, score 1 for my prediction skills

  1. Avatar for

    Information is only used to show your comment. See my Privacy Policy.

  2. Avatar for Ben Garney

    MBU was not developed on XNA. We just ported it over to make a nice demo. Sorry. :)

  3. Avatar for Steve

    I don’t think language / native code is really a security issue — if it was every online PC game under the sun would be hacked to death by now, and there would already be 360 pirate bootloaders out there — the fact that MS doesn’t supply the tools doesn’t stop people writing code for it, as proven with the original Xbox. The real key to security is signing — in the online universe you never blindly trust what the user sends you, and that is ubiquitous. Same goes for bootloaders & cheats — if they retained control over how signed / unsigned code was launched it wouldn’t matter what it was written in.

  4. Avatar for Damien Guard

    They can’t give people C++ and uncontrolled access or you’ll see pirate bootloaders, cheats, savegame hacks and Live! point exploits run amock.

    By using C# they can sandbox it thanks to the VM nature of the CLR runtime.

    There’s no announcement yet on the availability, pricing or functionality of the Pro version but there’s a small chance that’ll support C++ for indies although probably not running on your own 360 unless signed by MS at a guess.

  5. Avatar for Steve

    Definitely good news, I really can’t believe no-one has done this before given the constant interest in homebrew development.

    Am I right in thinking you’re only going to be able to use .Net, rather than native languages? If so, this isn’t particularly attractive to a lot of developers, particularly given that the industry is 90% native C++. I realise MS would like to catch more fish from other industries in their .Net, and believe that they can convert the whole world over to it, but there are plenty of people who don’t want it. Besides, there are simply far more game-oriented libraries available for C++, and no matter how good GG / MS’s tools & libs are, people are going to want more — in the same way that most people prefer using extra libs rather than coding straight to DirectX, and why not everyone likes Torque. Cutting out access to C++ really decimates the field of good code you have to choose from. Sure you can re-write those libs in .Net, or wait for someone else to do it, but that kind of defeats the object of having tried & tested libs available to use. I’m sure they’ll get lots of converts from people who already use .Net and want to make games though, I just wonder whether it will just turn into another ‘subculture’ unless there’s a direct path to native development — one subculture already exists for people using Managed DirectX for example, but there’s relatively little direct transition between that group and the more numerous / mainstream native code developers.

    I was also amused to notice they’re pushing Autodesk’s FBX format — most of the momentum over portable media formats these days is behind Collada, with an awful lot of people rejecting FBX mostly due to Alias (and now Autodesk) exerting far too much control over the format, which sort of led to Collada in the first place — it’s open source and now a Khronos standard. But of course Collada was invented by / is promoted by Sony and even though they’ve cut it loose now, it’s now controlled by the same group that manages OpenGL, so I guess Microsoft doesn’t want to get involved in that (for the same reason they chose to reinvent with DirectX rather than pitch in with OpenGL). Still, I expect third parties will add Collada support pretty quickly no matter what MS choose to promote.

    I also note they’re keeping strong control over who can play the creations, so it’s mostly going to be a developer-to-developer exchange programme I guess (I’m not sure many non-developers will sign up for the annual fee), with end-users only being able to play those that get officially promoted to XBL. Good for their quality control I guess, but it does lead to the question of how this is any better than current ‘unofficial’ homebrew scenes. Developers who are savvy enough can already develop on consoles using homebrew kits and exchange their code, they just can’t give the code to anyone else who isn’t into the scene because of the need for hacked firmware or workarounds. So your $99 is just making that same setup ‘official’, it’s not actually adding anything new — except perhaps that you can talk to MS about your game without having to shuffle your feet about how you developed it I guess.

    On balance though, definitely a step in the right direction, even if it’s not quite as far as some of us would like, it’s better than nothing. I only hope it encourages Sony / Nintendo to go one better and come up with a more open version of this programme, allowing non-proprietary toolsets.