I’m going away for a little less than a week and I’m not going to have any internet access, so no working on Caverns of Underkeep and no updating the blog. For those of you testing the game, please still give feedback, I really do appreciate it, you can leave comments on the blog if you wish, I won’t be able to reply but I will be able to read them when I get back.

As for Underkeep news I’ve added the first mini-game, a very simple jumbled-up word-puzzle used to unlock chests that are randomly placed about the dungeon. It was pretty easy to add and I think it’s quite effective. Although I’m not sure about the hint system I’ve developed, because you can just brute force solve it - I’ll probably change it when I get back.

Another thing that I’ve been thinking about recently is the direction I want to take the game in - And I’ve come up with two main objectives.

1. Short playtime (approx 1 hour to beat, 2 would be absolute maximum)

2. High replayability

The way I see replayability working is that you get a different experience almost everytime, this is somewhat handled through random level generation and different classes\challenge modes but I think I can take it a little further.

For example, I want to introduce ecologies, meaning that a level might be infested with spiders or undead or Fire and Lava creatures. But you wouldn’t know until you played the game as to which you got and each ecology would have it’s own special strategies that work best against it. The level would have all the decals and features for the creatures inhabiting it of course.

That way I can still keep the depth of the dungeon small, but increase the chances that the game will be different each time you play.

Another thing I want to do is add heaps of secrets and easter eggs, I really liked it how in Kings Quest II there was some ridiculously small chance that you could see batman driving the bat-car in and out of a cave. A friend told me about it once and I thought he was lying because I’d never seen it - Turns out he wasn’t lying, just that the probability of seeing it is really, really small.

Edit: I’d just like to point out how much adding ecologys would effect random generation. Say I have 6 levels that each have 2 possible themes, then I have 2^6 = 64 unique combinations of a 6 level set. This number grows at an exponential rate. So if I have 3 possible themes for each level then I have 3^6 = 729 unique combinations. That means there would be a < 0.14% chance that you would get the same ecology combination as your last playthrough.