Another year of development has passed like a lighting! As promised, here is a short summary of what has been going on in KeeperRL land.

There were only 3 major updates this year, compared to 5 in 2015. On the bright side, they contained big gameplay improvements: campaign mode with retired dungeons controlled by AI, adventurer campaign, free furniture and floor placement, and manual production. They all required tons of architectural changes, which is why the updates took quite long.

In addition, I managed to complete a lof of technical work, like moving graphics and audio from the SFML library to SDL + OpenGL. Besides getting rid of some bugs, it allowed some nice UI improvements, like the pretty mouse cursor, and smoother text scrolling (ready to be released in Alpha21). The framerate has also improved drastically, although optimization is more of a continuous battle, and it’s not over yet. 🙂

At Alpha19 the game started gathering anonymous statistics, which gave me a lot of insight into how people are playing the game. Below is a breakdown of what types of games players choose. It seems that adding the campaign mode was a good idea, although a number of players still like the single map mode.

Note that single map games are not broken down into keeper and adventurer types, due to my oversight in the tracking system.

KeeperRL’s difficulty can be read from the percentage of won games. The graph below also shows that over half of keeper mode games are left unfinished, which is a bit worrying. Presumably these are games where the player got bored or was unhappy with his or her dungeon. I’m going to add further tracking to figure out why this is hapenning.

Finally, the chart below, courtesy of Steam, shows how much time players spend in the game. About 25% of players have only played the game for less than an hour, and the median time played is 3h:38m. I’m not happy with these numbers, and they suggest that KeeperRL doesn’t have a lot of replay value. Most likely it needs big additions of content. Adding a tutorial should also help with keeping more players engaged with the game.

In terms of sales, 2016 has been good, with around 7000 units sold, compared to 11000 in 2015. The previous year saw the release on Steam Early Access though, so there is no surprise here. What’s important is that I have a stable income and can continue working on the game full time without any distractions. 🙂

Time to talk about plans for 2017 🙂 . With most major gameplay features more or less finished (and a few more coming in early 2017), I think that I will have quite a lot of extra time to add more content. I don’t think that the length of individual games will change much (maybe with the exception of the new endless mode), but I want to add a lot of random events and encounters, which should make invidual games more engaging, and more varied between playthroughs. And I will also add more secrets 🙂 . In other words, the kind of things that make roguelike games very replayable.

I don’t think that KeeperRL will leave early access in 2017. Perhaps it will reach something close to a beta stage, after which I’ll focus more on polishing the gameplay, bugfixing, visual and audio improvement, etc. But it’s a very wild guess, as the project lives its own life, and I’ve never managed to predict its future as far as a year ahead 🙂 . All I know is that it looks bright, and I can’t wait to see what’s going to happen.

I wish all the best in the new year to all the fans of my game, and I hope you have loads of fun with KeeperRL, and a little bit with other games, too. 😛