Monday, 1 February 2010


Why is Vertex Dispenser taking so long to get finished?
When someone asks how long I've been working on it, the simplest answer is "three years", since I first started some version of it in early 2007. But it's not like it's been three years of solid work (even solid hobby-project level work), because in that time (apart from writing a Masters thesis, moving to England, starting a PhD, spending periods of time failing to achieve anything much due to illness or depression, getting married, and having a life) I abandoned it for a while then took it up again, made a few other smaller games and started several others that I didn't finish, so it's more like two years, if that. Even so, that's much longer than I'd expected. The difficulty of completing a project seems to increase disproportionately with scale, and the saying about the last 10% being 90% of the work seems about right. Plus I just haven't planned on selling a game before - it was in a state I'd have been happy to release as freeware some time ago. I do think it's quite hard to finish full games when not working on them full-time.

But it really is quite close now. Here's my schedule of what I'll be working on once I get my computer back:

- The campaign levels are too hard, so right now they're more of a challenge than a tutorial. I have a plan for what I think needs to be done for the first part at least (need more playtest data for the second part still). When I've made these changes, I'll need to playtest again on some fresh victims, and probably will have to go through another iteration if it's still not working right. (I should have playtested on non-mathematician gamers sooner, but I just didn't realise there'd be such a difference.)

- Multiplayer has a couple of bugs still. Sometimes it detects as being out of sync when it isn't, and sometimes there's a crazy spike of lag for no apparent reason. (Also, it's a primitive system where you connect to the host by typing in their IP address, and the host has to forward a port on their router, etc. It'd be nice to improve on this but I don't know how and probably won't; current system is perfectly usable.)

- Sort out sales, distribution, etc. Figure out what to put in the demo. Then release it.

(There are some other aspects I could improve, like UI and audio, but I'm not going to because I'd rather get it finished and start on something new.)

So this isn't actually much work, when I have the time and hardware to do it. I'd like to give an estimated time it'll take, but the main bit is the playtesting iteration, which is partially out of my hands, and will take an unknown number of steps. Upper bound of two months, I'd say. Looking forward to it.

No comments:

Post a Comment