Author Archives: Daniel

In the spirit of giving deeper insight into the game development process, I’m taking/stealing a page out of Sid Meier‘s play book and sharing with you guys one of my bad ideas.

In an earlier iteration of Dungeons of Dredmor, spells were learned from spellbooks found on the ground, of various “magical strengths”, which had different statistical probabilities of yielding spells.  Strong spellbooks would likely yield multiple spells, or higher level spells (which were harder to get), but you were never guaranteed a spell.  Not only that, but spell-casters had a better chance to learn spells than fighter-type characters, and so would end up with more spells, but there was always that chance that a fighter could “go for it” and get a really sweet spell to use when they had no more options.

It turns out that this was a terrible idea.  People hate the concept of failing to gain spells and instantly are less inclined to submerge themselves in the “Dungeons of Dredmor” experience.  We then were struck with a bad situation because if we guaranteed that spells would be learned, people would always go for the higher level spells in the book and always be above the power curve.

The solution?  We got rid of spell books and implemented a wizard leveling mechanic which rewarded players with the next level spell of whichever spell school they chose.  It combines limited choice (always a good thing) with automatic advancement (something which keeps players interested and rewarded for their zomby killing) …(and yes, that’s supposed to be “zomby”).

Next Friday, another exciting look into another one of my bad ideas =)  See you then!

Hi, I’m Daniel Jacobsen (or sometimes Citizen) and I am the gameplay programmer and biz dev guy here at Gaslamp.  My posts here will generally have to do with designing systems for our games that are both fun and intuitive (or at least attempting to) as well as the hurdles that our indy game company is going through in order to support our lifestyles making games that people enjoy playing.

Our current project, Dungeons of Dredmor, started out life well before the Gaslamp team got our hands on it, as a project that Nicholas was working on while doing his undergrad degree.  We initially picked up the project as something that was “near completion” and just needed to be taken from a system of dungeon levels and monsters to a compelling experience that someone would enjoy sitting down and playing through.   Like most things, this was way harder than it seemed at the time.

Since we were dealing with a complete system without a game, one of the first things that we had to decide was how to entertain people with what we had. There are a lot of ways that people go with this sort of thing, but as a throwback to the “Rogue-like” origins of Dungeons of Dredmor, we really wanted the player to tell their own story through interesting solutions to problems and unique quests. Essentially, our goal is to allow players to create their own unique story every time they play through the game.

This, unfortunately, is much more difficult than it sounds, but using some interesting and slightly insane mad-lib style “procedural content generation”, we’ve developed algorithms to generate dungeons, monsters, items, and even quests that have such a large number of possible combinations that I can practically guarantee that you will never play the same game twice. While the ways in which you interact with things always stay the same (we thought about generating that too, but it would just be weird =P) and the end goal of defeating Dredmor is going to be the same, you can always find a different way to get there.

I mentioned earlier that the origins of Dungeons of Dredmor are in the traditions of Rogue, as it was initially intended to be a fully graphically represented Rogue-like with a single button interface, which displayed no numerical information to the player. After countless hours debating, arguing, and collaborating upon whether Rogue-like games have a place in today’s gaming world, we decided upon a less… erm, deadly approach to dungeon diving than Rogue loyalists. We also ditched the idea of having a single button interface. While we still feel it is important that everything *can* be done with just one (okay two) button(s), given the level of complexity that most gamers are comfortable with these days, we’ve augmented the system so that hotkeys that people are familiar with exist in our world too.

Finally though, we did stick to the “no numbers” philosophy. In part because not having numbers for players to focus on requires instead that they pay attention to other things in order to decide how well they’re doing in a game, but I’m sure also at least partly because of the challenge associated with rewarding players not by a higher damage number, but in other less common ways which has really pushed us in the direction of trying to understand exactly what players want, and how to give that to them without playing a game seeming to devolve into maximizing damage and hitpoint numbers.