A clear problem of being a one-man-band game developer is that I don’t get access to rich game testing resources. As much as I play the game to death whilst developing it I just don’t get the right kind of feedback to help push the game over the line.
Indeed one of the issues is that I get just too close to the game and probably accept its problems and limitations.
With Spy Chase I resisted adding guns and missiles to the player’s car since I’d always wanted it to be a game about dodging obstacles. But the more I played it the more I realised that I needed to offer some effective resistance to the chopper. The chopper is a big feature in the game and I didn’t want to lose it. But it was proving a real pain and disrupting the natural flow of the action.
So I implemented the auto-firing missiles and without giving it too much thought called it “done”.
The trouble was I’d instantly undone a ton of hard work. The game became so simple that you could just leave the controls alone and still capture the spies.
My last post generated its fair share of inbox activity with numerous suggestions for how I might improve the game.
Now I’m not your average developer when it comes to this sort of thing. Many developers that I know would quite arrogantly dismiss these suggestions but I actually read them through and valued the input. If you take the time to type it I’ll certainly take it on board.
Several suggestions were not relevant but some were excellent. So I took the excellent ones and played with them.
One of the first things to change was the auto-firing weapons. I love auto-firing weapons in mobile games and had got it in to my head that it was the only way to implement weapons. But of course it isn’t. Weapons should be a bonus. As such I made them collectable and further set them apart by changing the firing rate on bullets to be much faster than missiles. Missiles naturally deliver a bigger punch.
The biggest change I made though was to the effect that the obstacles had on the player.
Previously I’d penalised the player for hitting an oil slick or road cone by decreasing their speed. But this was wrong. It just felt bad that the player had to protect his speed. The largest part of the fun in the game is the high speeds that can be achieved. I wanted to maintain that.
So instead I hit the player in the other area that he cares about – car damage.
This was actually a huge improvement since it meant that the player could no longer just leave the controls alone and progress.
For every collision with an obstacle I hurt the player’s car. This really counts since the further in to the game you are the more likely it is that you will face such obstacles.
Finally to help give the player a fighting chance I reduced the maximum speed of the road from 32 to 28. Believe it or not the extra 4 pixels of movement in the road makes a huge amount of difference and can be the difference between deliberately aiming for something or being able to steer out of the way.
What I liked about this process was that a) people felt like they could offer suggestions and b) I was able to consider and act upon them with ease since I’d structured my code to be flexible enough. In short this is an exercise in creating a rich array of properties and variables such that at the end of the day I am simply playing with numbers.
A valuable couple of days.
The final game is here: http://www.wilfscorner.co.uk/games/spychasemobile/