Arpad invited his friend, Laszlo Gerecs, to help with the programming aspect of Beyond Beyaan. Laszlo is a Linux developer, and we're currently discussing what direction we should take with Beyond Beyaan. I believe that we will convert it over to C++ for easier crossplatform development (Linux, Windows, and Mac), but the discussions are still ongoing.
We've also decided to move the source over to git, so from now on, you can find the source code and assets from here: https://github.com/Beyonders. We will clean up the source files and assets so they match open source format (Assets will be in their own repository, with source code separate in another repository). You may notice that we have "Mono". This was an attempt to see if Mono is feasible for our crossplatform goal, but turned out to be difficult to set up. Hence our discussion if we should go the C++ route.
The code will still be on SVN at google code, but it's depreciated and won't be used anymore. It's there for reference (I still have the UI XML and other nifty features that I ripped out in there).
I also worked on Beyond Beyaan a bit, and finished up the Fleet Specs window, the one where it displays if you're trying to add a new design but you have six designs already. Here's a screenshot (For reference, the old one is seen here):
Moving over to Git + GitHub is *great*. That will really make it tons easier for potential contributers.
ReplyDeletePossibly switching over to C++ is interesting as well. Porting BeyondBeyaan over to other platforms was never going to be an easy task.
Converting the logic to C++ would be quite straightforward, but replacing Gorgon never was going to be and would be necessary even if you migrated over to Mono. (Obviously I'm not telling you anything that hasn't come up in your discussions already, I'm sure).
If you are commit to other platforms then you're due for a major rewrite in any event. The only way it could be avoided is to continue to be Windows-only.
The standard industry advice is "ship the game", and do your rewrite for sequel. That's good advice.
Of course, this is all a bunch of software developers talking, yes? Of course we all relish the thought of chucking out old code and starting with new shiny, perfectly engineered code. Well honestly starting over very rarely pays off. But we all want to. I read the post and I go immediately "oh yay!" because I'm a C++/Linux dev myself. So, if you really think you can get away with it....
We discussed that in order to attract more developers, it has to be multiplatform. When we looked at the amount of work required to re-do the rendering code, we felt that "If half of the game has to be re-written to work on multiplatform anyway, why not take a little step further and convert to a more popular/accessible language?"
DeleteIt's not about "starting over with new shiny, perfectly engineered code". It's about getting more developers on board. C++ is the most popular open-source language. So we're killing two birds with one stone (multiplatform and C++).
I will continue to develop C# version, then when we've got C++ project up and running (we're trying to determine which 2D engine we should use at this point, as well as frameworks and other stuff that we want to use), we will start converting stuff over (mostly copy-n-paste whenever possible, with conversion to C++ code as needed). It's very possible that we'd continue work on C# version as our UI prototype while we convert the game logic to C++.