Frictional Games Interview - PAGE 1

Not long ago we got a good, long look at Amnesia: The Dark Descent, an "interactive horror game" on the PC from Penumbra creators Frictional Games. It didn't take long for it to become apparent Amnesia was not like other horror titles or even other games, and with that in mind we had a chat with company co-founder and programmer Thomas Grip via IM.

Read on for a discussion about horror and innovation, and a deconstruction of what the term "game" is really all about.

I guess I'll start off heavy. (laughs) You've talked about building Amnesia and games in general around the experience first and mechanics second. How is this harder and easier? Can you describe this process at pre-production, mid-production, and post-production?

Oh! Heavy indeed.

Gonna try and sum things up.

Take your time.

Me and [co-founder] Jens [Nilsson] actually talked about this a few days ago and we said that we perhaps make games a bit different than what normally is done. The "normal" way is to start out with some game mechanic and then build the game with that as the base. A "normal" game might first try out this mechanism in simple prototyping setups, iterate and enhance until it is good enough and build the rest of the game.

For us it is almost the complete opposite! We start out with a certain experience in mind and then from there we build the game, changing the mechanics as we go along. I say "almost" because we generally start out building the basic interaction and character control and then go from there. Once we have this ground we then design levels on paper and write story. We usually start doing 1/3 of the designs first and then start building the levels according to design. So design and level building happens at the same time so that we can change design if we feel that something does not work out.

An important point here is that we almost never make prototype levels. For Amnesia we tried making levels in simple geometry, but we did not get much atmosphere so it was hard to judge the experience and instead made them something like 80 - 100 percent complete. This is also how we made levels in Penumbra.

Now during this process we also tweak the gameplay alot, which means stuff like enemy AI, item placement and function, general systems (like sanity in Amnesia), etc. Puzzles pretty much remain the same at the core but we tend to iterate them a bit, adding more solutions and so on to make sure we have the proper flow.

Post production phase is not really existant. Once the final level is built, the game has to stay like it is and we just polish some minor things. This also means that we tend to play the full game very late in development. For example, I played Amnesia from start to finish a couple of weeks ago.

I think that gives a decent overview on how we work!

Also worth nothing is that we actually tried to not do it like this in Amnesia. We started out the project trying to come up with a good basic gameplay mechanic and then build from there. This approach did not work very well though and we found ourselves sacrificing alot ambience and feel of the experience just to keep the gameplay working. So we scrapped that and went back to how we made Penumbra. Acually, the Amnesia production turned out even more atmosphere and horror focused than Penumbra and we changed almost every mechanic during the production.

Quite a mouthful. Good stuff.

Aside from that methodology, what did you learn from Penumbra that you're bringing to Amnesia in order to "evolve" games, as you say? And in the same sense, what have you learned from Amnesia that you'll bring to future games? Or "works", or whatever you prefer. Going to talk about that shortly.

I think that most of the learning from Penumbra has come from more basic design things, like how to build story, how to place puzzles in levels and so on. Very broad and general stuff, that is hard to get right by pure iterating and polishing since it covers so much of the game. This has given us knowledge on what kind of puzzles work, pacing and so on.

I hope that this knowledge can then be used in order to enhance the kind of experience we are after. By this I mean that we can add mechanics and design levels in that work towards certain emotions and themes.

It is really hard to point out specific things, but the at the core it is a about finding ways to make the experience engaging. At least I hope we have succeeded in this!

With Amnesia we are trying to reduce the more gamey elements even further and put more focus on the atmosphere and immersing the player than what we made in Penumbra. I think it would be hard to do that without having made the Penumbra games as we did a lot of testing in trying out what works and what doesn't.

In the end, this will hopefully enable us to create more varied kinds of experiences and to involve a wider range of emotions. At least that is what we are striving for!

Finally, from a horror perspective, we have learned a lot on what kinds of encounters, scenes and events work and what doesn't. Building a good horror experience is a very holistic thing and everything needs to come together in order for it to work. Again, hard to point out specifics, and it is more about getting a general feel for things.

It's fitting that it's abstract.

The lack of combat seems to help a lot in making the experience so immersive, which is maybe unexpected.

I found it unexpected how much it changed the game and how players approached it! Without weapons players seem to view it very differently. If you have a weapon in a game, (gaming) experience tells you that you should be able to use it on all that you see. But when the weapon is removed and you face a monster unarmed, it is a totally different thing.

I believe much of the experience is up to the player and some part of roleplaying is needed. Removing weapons seem to give more room for this.

Yes, it's disarming literally and figuratively.

Yeah, exactly!