So we spoke about being lucky enough to still have all the assets and lore from the original Shallow Space project, one thing we lost however is the source code.

There has been multiple iterations of Shallow Space, the original project was a single instance RTS with a lot of UI clutter, when it became clear that we simply couldn't do what we wanted with that and because we had a whole new ship lineup we then rewrote the game as a multi-instance RTS (2.0.)

We have the source from the original project, but not the more useful revised project (the one currently on Steam.) This sounds like bad news but it's more of an opportunity to get a few things right from the start - the biggest issue with both of the previous game projects was that they didn't have multiplayer.

One doesn't just add multiplayer to a game, the game actually has to be designed around it. It affects (and restricts) decisions on gameplay design.

This time around we started by writing the multiplayer components first, so if you've been wondering what's been happening for the last 6 months, the screenshot above should help.

The engine has been designed to be light and performant so it'll run on cheap servers in the cloud. This means we can leave it running and have super cheap bills and scale it horizontally if the whole world turns up to play.

In that rather dull looking source, we've managed to capture the essence of what made the original Shallow Space and the revised 2.0 version so exciting and cool, multiple sectors with concurrent activity and the potential for detailed ship customisation.

But none of that matters if the game is as much of a buggy mess as the previous two. This is where the real time-sink lies, writing the bloody tests!

It's important we do this, because with tests we can be sure every component works as intended and as we roll in the features and functionality (this time in a planned and controlled fashion,) we need to ensure what works already doesn't break.

This was a key failing of both of the previous Shallow Space projects.

Fortunately we have much more knowledge of software creation this time around, and we have experience building and supporting apps that scale.

So with all of this we can ensure that we create something that is more reliable right from the very start. So losing the source code wasn't such a bad thing after all...