WME2 developmentIncoherent info about WME2

As I hinted before, WME2 source code is now accessible at Bitbucket under the MIT license. I don’t think it will be useful to anyone in the current heavily-work-in-progress state, but if anyone’s interested, they can dig through it. Also you can now follow code commits (I know odnorf probably will ) Next milestone: the roadmap.

https://bitbucket.org/MnemonicWME/wme

While getting wme2 source code ready for uploading to bitbucket, I’m using the opportunity to reorganize and rename the project files a bit.

I apologize for the lack of communication lately. For now let me just quote what I posted in the WME forum:

Thanks for your interest. The project is alive, although a little stalled at the moment.

I’m kind of burned out because of my daytime job and I’m failing to kick myself into more programming in my spare time. But I’m sure it’s only a temporary setback
Realistically, though, there are tons and tons of work to be done. For that reason I’m considering changing the way the development is organized and making the development more open, in the hope of luring some contributors to the project. I’m not thinking necessarily just programming, but perhaps some artistic stuff (demo assets, 2D art for the app, shaders…).

I hope to be back later with more news! Thanks.

I managed to hack SDL 1.3 into displaying in landscape mode on iOS devices (it’s a quick&dirty hack, I’m still hoping for an “official” landscape support), so I was able to do some more tests.
While most things seem to run and perform nicely in the iOS simulator, it gets more complicated when running on a real device.
I tested on my 3 years old 1st generation iPod Touch, i.e. the oldest and least performant iOS device you can get. Hopefully I’ll be able to test on some more modern device to see the difference. But anyway, here are my findings so far.

Continue reading →

So what I’ve been up to lately. I took a small break from wme2 (don’t worry, I’ll be back soon) and turned my attention to the good old wme1. I was thinking, what would it take to port wme1 to other platforms? I knew it wouldn’t be trivial, but the main reason was the 3D part (it’s a bit too much dependent on Direct3D). However, perhaps it would be worth the effort to only port the 2D portion of the engine, wouldn’t it? And that’s what I did and thus “WME Lite” was born. WME Lite is a fork of WME1, with all the 3D functionality and non-portable bits removed, the renderer is using SDL and the sound system is based on Bass.

Currently WME Lite runs on Windows and Mac OS X. A Linux port should be fairly straightforward, I think, but I’m not a Linux expert.

Admittedly, my main aim were mobile platforms, namely iOS. It’s still a big unknown, though. Unfortunately SDL currently doesn’t support landscape mode for iOS, so unless it’s added soon, I’ll have to hack the landscape mode in somehow. Also my early tests show serious performance issues when simply running an existing WME game on my 1st generation iPod Touch. I don’t know what’s the performance difference between the old iOS device and the newer ones, but in any case I’ll bang on things some more.

It’s a fun little project and I’m sure the experiences will come in handy for wme2. Stay tuned.

I mentioned on several occasions that I’m using the Qt framework for developing the Armitage Editor for WME2. Qt is an amazing C++ library stuffed with functionality, available for free under LGPL license, it allows easy porting to other platforms, and it’s backed by a powerful corporation (Nokia). The future looked bright… until last friday.

Continue reading →

I’d like to address some comments on earlier blog posts, where it’s being suggested to go open source with WME2.

Firstly, I believe open source can be very useful. When using Ogre and Qt while developing WME2, I relatively often check the sources or trace into their code in the debugger when something doesn’t work as I’d expect. It took some time and effort to get familiar with the structure and philosophy of those projects to be able to do so, of course. Ok, but that’s the point of view of a C++ programmer using open source *libraries*. For applications, personally I don’t find open source as much useful. I don’t think I’ll be ever inclined to dig into Firefox code, for example. And I’m a programmer, for normal users it’s out of the question anyway. WME2, even though it will be essentially a developer tool, is going to be closer to the application side. Most of its users shouldn’t need to deal with the sources. Or at least that’s how I see it.

Continue reading →

So what is this “serialization” thing I mentioned several times? Simply put, in context of games, it’s when the player saves their game (serialization) and later restores the saved game (deserialization). It may look simple, but it’s not. A typical running game consists of hundreds or thousands of various objects, connected randomly together in a huge graph. The purpose of serialization is to take this graph, somehow flatten it and save it to a disk file, and later to restore the file into an exactly the same object graph.

Continue reading →

In preparation for serialization support I’ve been doing some consolidation of existing functionality. You know, I have those little pieces of functionality scattered all around the place, temporarily glued together for testing, and now I’m joining them together in a proper way (content manager to viewport layout to scene content to game entities to scripting engine to user input to…). It’s almost a miracle when things start fitting together. I can attach scripts to objects, the scripts react to mouse clicks and trigger actor animations, soon hopefully the scenes will be able to save/load themselves to/from disk.
Don’t get too excited, though, there’s still a lot of missing functionality (even for a minimalistic initial release), not to mention missing editors, but I’m getting there

What, it’s 2011 already? So where are we standing with wme2 development? I’ve been mostly working on the 3D scene editor and related tasks. I’m rather pleased with the result, if I may says so myself. I’d like to say the editor is nearly finished, but since software is never really finished, let’s just say it’s nearly in the state I want it to be for the initial release. It’s probably the most complex editor of all the editors Armitage will contain, so from now on everything should be a breeze. Right? Right?!?!

After finalizing the scene editor, I intend to get back to the engine itself and revisit the scripting engine. I’ll hook it to the game objects and finetune some performance settings. Then I need to implement one last major functionality – serialization (meaning, saving and loading the game in progress). But I’ll blog about that separately.