One of the most common questions people are asking us is what engine/tech we’re using for AirMech. Well, yesterday we nuked last closed source library from our code base and as of today all the code we’re using is custom in-house or open source. We now have a complete list of open source software we’re using for our game development. This list is complete for all software we’re using on the client. We’ll also update it with open source software that we’re using in code we’re not distributing: for development, game servers, web, etc. Without listing all this we might not realize how much we rely on open source. This project wouldn’t be happening if it wasn’t for this open source or at least our progress wouldn’t be as rapid as it is. We are not against licensing or using proprietary middleware, but so far on every step it showed to be an obstacle. An obstacle in terms of lock-in onto specific platforms, or if a middleware vendor is not interested to support some platforms we are considering, or just requiring signing complex agreements. To us as an indie developer, freedom to pick any platform we want and port our code base to, is really important. Other benefits of not having middleware is that we don’t have to annoy players with splash screens which are often required by middleware providers. We want players to get into the game as quick as possible, without waiting and staring at product or company logo for some time limit defined in some licensing agreement. Based on questions we are asked I’m under the impression that, for some reason nowadays, people think that is impossible to make a video game without using some commercial middleware. Yes there are great products that enable a lot of creative people to make games without investing too much time into technology. But there is also a lot of great open source software that can be used to build games. We are also old school, and we come with a lot of background in building engines and tech for games. Also we’re living in the time when game development has never been easier. Graphics hardware is matured, there are no huge changes, or at least for the target market we’re targeting there are no huge changes. In the 90ies supporting whole range of hardware was such a mess: many manufacturers, many standards, some complete crap graphics cards and you have to write software rasterizer for players without graphics cards. Nowadays there are only a few GPU manufactures and not many graphics standards. There are still issues with players have really old hardware, and there are still graphics cards that can’t even run DX9, but we set our minimum specification to DX9 class GPU and we’re going to optimize for it. Just because it’s open source doesn’t mean it’s good by default. When choosing an open source library, we have a few considerations. We avoid bloated libraries that are trying to do everything. Ideally library is only doing one thing and it’s doing it really well. It’s easy to integrate and maintain. Also even thought we’re using C++, we find it easier to adopt a library written in C than in C++. Usually C++ libraries tend to be more complex than their C counterparts. This complexity usually comes from requirement to include other libraries or just from the feature set. Often C++ libraries look like someone was trying to impress the C++ committee with their metaprogramming cleverness than to create something useful for other C++ programmers to use. When we choose a C++ library, the library cannot require the use of exceptions or RTTI functionality of C++ language, since these are always disabled in all our builds.

I know in the past I have been a big fan of cocos2D and have talked extensively about the functions. I have in the last two weeks written a side shooter an a box2D pinball game nether of the two could have been done without cocos2D. Now I am learning how to build a rapper for UIView class for displaying video for OpenGLES layer. This software is amazing the box2D stuff is complex and time consuming but a huge improvement over writing a physics engine for 2D. With the tools such as vertex helper it has shorten the time building vertex models for your box2D powered app. Here they now enter open source C++. Its a little bit more involved then the Object-C version but a huge improvement over trying to build it from the ground up. The other benefit is the ability to develop for both iPhone and Android. So great platform cross portability for Android developers are racing it seems to get in on the cocos2D fun. I myself am jacked about the idea of being able to build on one platform and with only small amount of changes port it over to the other. I am still working with Objective-C its just another option for people to port over to move their games over to another platform. cocos2d-x - News - cocos2d-x.org "A lazy man had to write something because there's 25 days ago since the last news posted here. I'm lazy so I need the multi-platform framework to reduce my jobs, and left me time for playing video games At first, many useful blogs & tutorials are written by our kind developers, in the last 2 weeks How to compile and run cocos2d-x in linux (by laschweinski) cocos2d-x linux and android project in eclipse (by laschweinski) How to Make a Catapult Shooting Game with cocos2d-x and box2d (by clawoo) Performing a selector after a delay in cocos2d-x (by clawoo) Creating a cocos2d-x box2d android project (by clawoo) Platform versions supported by cocos2d-x How to enable Multi-Touch About device orientation And then, we have more funny cocos2d-x games, ios top10 games Cut Cut Boom, Ninja Chicken 2:shoot'em up, "3 Kingdoms TD: Legend of Shu" are ported to android based on cocos2d-x. Please navigate to "Games" tab for more new games. Fishing Joy HD android version, the super star of cocos2d-x community, has reached 3.4 million android users & more than 1.1 million active users so far. Currently, cocos2d-x games have more than 700,000 download on ios devices & more than 4.3 million download on android devices." Fishing Joy HD is the most popular download in China it seems. Ok looking game but it would drive me nuts!

Through a recent email to Alex Warren (a key developer behind Quest). The LGC found the newly popular Quest 5.0 game development software, does not have a native Linux edition. In part of supporting the linux game development and open source communities, the LGC is asking any eager developers who are "keen to make their mark" to join the effort. Those who are serious can contact Alex directly Quest 5.0 is a brand new system for creating modern text adventure games. It is a complete rewrite of Quest, and lets you create sophisticated games complete with graphics, sounds and videos, all without having to know how to program yourself - a full visual editor is included. - Visual script editor. A plain English approach to programming, where you can choose commands from searchable lists, so you don't need to remember any syntax - Highly customisable - most functionality comes from a Core library, which is written in the Quest game language (ASLX) itself - Games support unlimited undo - Games are fully translatable - there is no hard coded text within a game. You can translate the English template to create games in other languages - we already have support for Spanish, German and Dutch. - HTML game output allows embedding sounds, pictures, YouTube, Vimeo clips - XML game file format is easily editable outside of Quest if you don't want to use Quest's own editor - Games can be played online by uploading the .quest file to http://www.textadventures.co.uk/, or you can use the WebPlayer project to host them yourself on a web server with ASP.NET 4.0 - Ms-PL licence (like MIT) means you're free to create commercial Quest games, and use Quest in closed-source projects Todd B. LGC Administrator, Linux Game Consortium