Category: game design

This is not an easy task. It is for most Tower Defense games, because they are simply scripted. The level has a set number of enemies of certain types and times, and they appear as attackers. And of course, most TD games don’t let you play as the attacker, so programming the defense AI is not needed. I need both of these to be non-scripted for Gratuitous Tank Battles.

I need attacking AI that ideally adapts mid battle to your decisions, and this is harder than it sounds. The first approach I took was to have a system with 3 types of placement timing mode (steady placement, or hoard points for a big attack or spam the current unit) and two types of unit selection (random, and select the one that has been most effective so far…with some variation).

The AI would switch randomly between these combinations and it looked pretty convincing. Sudden waves of identical units followed up by who-knows what. The problem was, it obviously did so with little care as to what it was up against.The ‘most effective unit’ code was good, because it meant totally useless designs which charged into a hail of bullets would not see much repeat business, but something major was lacking. That was the anticipating of future events.

If the defender has placed 6 machine-gun turrets (rubbish, but devastating to infantry) you shouldn’t place down infantry, regardless how well they did earlier in the battle. If they have lots of laser turrets, place down shielded units to get past them…etc.

So today, at the end of a long day of video rendering and editing, and tweaking, and bug fixing, I’m starting work on a more generic system for opponent army composition analysis, that can take snapshots of the enemy forces and realise that its 63.2% anti-shield units, and thus we should de-prioritise shielded unit selection etc.

This is annoying, fiddly, long winded code that nobody will ever see, as such, but will make Gratuitous Tank Battles a convincing challenge, which is well worth spending time on.

Life is so much simpler when you bunch up all your hotkeys and toggles for debugging the game into a single menu. I feel like I have a rough idea what’s going on now. I also feel more confident about coding decent defensive AI when you play as the attacker:

The code for defending basically needs to manage three things. The play style (Is it spamming units, hoarding points, or steadily placing), The choice of units (Does the enemy have units weak Vs lasers? Are there lots of infantry needing flamethrowerising?) And the choice of unit placement. I’ve been concentrating on the unit placement debug UI today, so I can see what the AI is thinking when it chooses to place a turret at X rather than Y. More work to do, but currently it evaluates proximity to enemies, their size, victory-point value, closeness to the exit and proximity to the placement, plus it also has bonuses for being able to cover multiple paths from a single location.

It all needs lots of tweaking and balancing obviously, but it’s getting there. Attacking AI is much simpler.

After getting this stuff done, I’m thinking of experimenting with a unit ‘cooldown timer’ beyond a certain level to prevent spamming the same unit everywhere. Essentially there are supply delays if you plonk down 4 heavy tanks all very quickly, as new ones need to be dispatched from your offscreen HQ. I think it might be a good mechanic that encourages variety.

GTB is such a huge game that I seem to veer into certain areas of it for weeks at a time, then veer back into other bits of it and think “did I write all this code?”

One of the features from Gratuitous Space Battles that I was very happy with, was the online challenge system. It was very popular. About 226 trillion billion zillion challenge games have been played (roughly). Obviously I want a similar system incorporated into GTB, and naturally I want to fix the things that were not perfect, which I identify as:

• The challenge browser was not as good as it could be. You couldn’t filter out played/downloaded challenges.
• Any challenge that had content, be it DLC or a mod that you did not have, could crash the game, and not be pre-filtered
• The emotional connection between you and the challenger was limited. (Rarely used messaging, and rating, but not leaving comments etc).

I plan to fix all of this, but the middle one is the current one of interest.

GSB had a binary .gsb format for a challenge, which basically packed in binary data for the fleets designs, deployments and orders, and a little bit of data about the challenge (custom settings etc) and that was it.

GTB will use a new .pak format I’ve written that will behave a lot like a zip file. (no compression yet, sadly). It will be a folder full of stuff for each battle, all the enemies unit designs, the deployment timings for their attacking units (or initial deployments of defending units), and all of the data from the scenario file, down to the location of every tree, crate, barrel, barbed wire etc…

That makes a GTB challenge file slightly larger than before, but I can live with that. They are under 100k. How many custom maps in games are under 100k now? It also means you could move a few trees or change the texture of a single tile, and upload an existing scenario trivially without any inconvenience.

Anyway…. The upshot of this new file format is that theoretically, if I write the code for it, the format could include custom sounds, graphics etc. That means you could do a mod for GTB which included different textures for the terrain, and custom units, issue it as a challenge, and EVERYONE could play it, because the required content gets downloaded with it.

Sadly, there is no way to prevent dupes there, so if you have a custom tank texture, and upload 10 challenges, someone downloading all 10 gets the texture 10 times, but I think that’s not so bad. Only a minority of players are likely to issue or play modded challenges anyway.

The main thing is, my pak file format pakker and unpakker all works fine, so it’s another step along the path to having online play working. Now I need to replicate a lot of the GSB functionality (and improve on it).

Edit: I swapped out my code to use some zip code instead , after finding a zip wrapper that was extremely lightweight.