2012-12-30 - Cancelled

This project was conceived as a tech demo, though obviously I had hopes I’d be able to make a real game out of it. But this project has lain dormant long enough to qualify as dead. I started rewriting it initially with hopes I can make something smaller out of it but this is just a pie in the sky project. The scope of a game like this doesn’t fit into weekend hobby development by one guy. So this is it for Metal Crusade for now. The demo landed me a job so I guess it served its purpose, however awfully cynical it sounds.

I’ve started a new project called Lethe which is something that I can conceivably finish on my own.

Edit: Yes I can and since yesterday it even runs on Android.

2011-09-18 - Mental Crop Rotation

I’m working on this for now:
I’ll be finishing that and doing a test campaign to see how it plays. Once that’s done with I’ll try to come back to Metal Crusade.

2011-06-05 - Working out where I left off

I’ve gotten to the point where every time I get some time to work on metal crusade (usually over the weekend) I have to waste a lot of time trying to figure out where I left off (I do leave myself notes and they do help but not much).

This is because things now have to interact with a lot of existing systems in the game and every time I have to reacquaint myself with them. It’s my failing at encapsulation I guess. I need to figure out a better way of carving out some continuous time for development.

Currently I’m working on the reputation system that allows factions to create policies. If you attack a group belonging to a faction it will turn it hostile and possibly also turn its parent faction hostile towards you. Since I allow faction hierarchies and factions may switch sides this needs to be reasonably robust. Lots of big picture thinking hence my problem outlined above.

But what I’ve been thinking about most recently is controller issues. The game is being designed assuming you have a mouse and keyboard to use but has a facility to accept any input given it can be translated somehow into an absolute position on the screen and has lots of buttons available. Most of the functions are top level and there’s little nesting – this requires 1 to 1 key to function binding. What I’ve been thinking of is trying to come with a function nesting scheme that would allow you to use something with only a couple of button available. Think the menu system of I-War which lets you use a controller to navigate and toggle functions which would normally take up a whole keyboard worth of bindings. This however obscures the view and adds time to reach every function. Other games do away with nesting and make buttons behave intelligently depending on the situation – the action button approach where you press X to do anything that is appropriate here. I can’t see this working given the multitude of options available to you all the time.

Some solutions seem obvious. Can’t have dedicated speed buttons? Use only two buttons and a continuous press will go full speed, single presses will go in steps. The input translation is done per device so this can be isolated from the game allowing the input scheme to be separate from the keyboard one.

But the functions are just too numerous. I’m leaning towards nesting menus but having as much controllable automation as possible. If I can’t have a button reserved for releasing coolant then maybe setting simple rules for automatically releasing it would solve the problem.

2011-04-04 - Falling foul of 90% of indie game cliches

Pixelated graphics – check, chiptunes – check, generated content – check. Only now do I realise I’m a living cliché.

tigsource forum thread about indie dev clichés

Some of the clichés are just a way of cutting corners. OK, chiptunes are just an aesthetic decision but generated content and stylised graphics are also ways of cutting down on work. I’m not going to kick anyone out the door if they produce photo-realistic models for me.

In a fantasy world this game would run on voxels and play jazz improvised at runtime.

In the end though, this is hardly an indie game (not least because it’s not a platformer ;) ). Just something I work on for fun when I’m not slaving away on those “totally sold-out soulless mainstream games”.

2011-04-03 - Terrain generator

The terrain is currently just a premade height-map but I am working on a proper terrain generator. It assembles the terrain randomly from prefabs mixed with detail maps. It’s quite a substantial chunk of code and I can’t commit it until I fully integrate it otherwise it would just break what’s on gitt.

speed dials Basically a 32×32 tile will be blended with other tiles (with 8 pixels of overlap) to fit the layout of the map which may be a custom map or a randomly generated one (or a mix) and then depending on the terrain type a random detail heightmap from a pool for that type will be added on top along with a matching terrain type detail map. This is all done once when entering the map, it generates a conventional texture which is then used by the arena class to create the terrain mesh as if it was a premade one.

I also had some fun trying to redo the HUD in a different style but I didn’t like how it was going so I stopped. I think I went too far with limiting the resolution and the palette (64×128 with 32 colours). I think I’ll try doubling the resolution and create a common but bigger palette for the whole HUD.

2011-03-27 - Targeting improved; I need artists

I small commit this weekend – I’ve fixed the targeting so that when you try and shoot someone close to you the angle will try to converge on him rather than being parallel to your direction.

I feel a bit stuck in a catch 22 situation. To attract artists you need to somehow show your project and its progress visually but without artists it’s hard to do. I need artists to get artists. Actually the catch 22 in the book was less of a cyclic dependency problem and more a mutual exclusion paradox. How about we go with a song title “Taking Drugs To Make Music To Take Drugs To” by Specemen 3 instead.

I’m thinking of redoing the HUD in pixel art and producing some placeholder level props but I’m more keen on getting more functionality in first… back to pushing bits instead of pixels for now.

2011-03-13 - Flipbook animation explosions

Added a separate particle class for flipbook style effects. This will be useful for burning flames, petrol-like explosions etc. The explosion is a single 128×128 sheet of 4×2 frames.


I need some artist’s help to create the art assets. I can keep at it myself but It takes time away from coding. Also, I’m not exactly a pixel artist, though it’s surprisingly fun chiselling sprites pixel by pixel.

2011-03-05 - Back on track

I’m getting back to work. Right now I’m down with a cold or possibly flu so I can’t really sit at the PC for too long but I’m working on ideas on paper and as soon as I stop shading my eyes from the LCD glow I’ll get back to shifting bits around. I probably should fill in bits of the design doc regarding required art assets in case I manage to find an artist to join.

2010-12-27 - Away for christmas holidays

After numerous cancelled flights and a few days of travelling I arrived at my parents house in time for the christmas dinner but not without catching a cold (or flu, or who knows what). Since I’m still the only person working on the project this means a lull in production but I’m adding bits and pieces to the design doc whilst recovering in bed. This might stretch to late January since I’m also starting work at a new place in a couple of weeks. But soon, regular updates will resume.

2010-11-23 - Release 0.1 - revision 1

I’ve fixed some bugs and added more info to the pause screen. I’ve set up a simple scenario – the favourite amongst all gamers – an escort mission. The binary packages are up and the source package will be updated soon (I’ll also add Visual Studio project files to it).