The holiday break was rather dramatic for Salvage. We took a couple of weeks off, sure, but as I suspect isn’t that unusual we spent that time stewing over some pretty big decisions about the direction of the game design. The result is design heartbreak, reduced project scope and a fair bit of progress on the encounter gameplay.

Ranged Combat System

We’ve implemented the core of a decent combat system that’s driven by actions. Actions are essentially attack types (think ‘spells’ if you’re RPG inclined) that are currently ‘granted’ by the weapon you have equipped. Different attacks might have different AP costs, different base accuracy bonuses or damage modifiers. They may also have a chance of applying a status like ‘stunned’ or ‘pinned’ to the target – those aren’t a factor in the game yet, but will be soon.

Initiative Ladder

Instead of the whole-team turns more commonplace to this type of game, we’re experimenting with individual character turns, utilising an initiative ladder. We feel this is worth exploring because we want our combat to be brutal, with very brittle combatants, which in a whole-team turn system could mean one team devastating their opponent in a single turn. Initiative ladders also add the concept of time and coordination to a player’s decision-making burden.

Tactical AI

Where previous prototype AI I’ve written has been driven by simple conditional logic, I’ve now implemented a location-weighting system into Salvage that assesses all tiles an enemy agent could reach and picks the best, weighing such factors as the combined CTH (Chance to Hit) of every target it could see from that location and the defensive value of it against the risk posed by the combined CTH of every target that would consider the unit the most tempting target at this location. In this way the AI decisions are now far more nuanced and less predictable. I anticipate explosives changing everything.

Layered Animation

We’ve long known that we needed to reconstruct our animation controllers in Unity to handle animations in multiple layers, but we’ve only just had the time to do it. I highly recommend doing it sooner rather than later; it can save you from the nightmare of animations like “RunningWithRifle”, “RunningUnarmed”, “RunningWithPistol” and so on. Now we simply have a layer to handle what the legs are doing, another for the torso and soon we’ll split the arms into individual layers so we can have dual-wielding weapons without all the hassle of special animations for them.


Our current priority is rebuilding the ‘new’ campaign so that the game design is whole, but not complete. The goal we’re aiming for is to have an Alpha ready by the end of the month that’s playable without explanation or instruction, so that we can begin gathering feedback from gamers and developers we know to aid us in our development towards the Beta.

Until next time.