Morten Monrad Pedersen
I was recently interviewed by Guilherme Felga from the Brazilian 1 Player Guild. The interview can be found here, but for those who like me can’t read Portuguese I’m posting the English version here on my blog.
The focus of the interview is my Automa Approach for making artificial opponents for multiplayer games. So if you’re interested in that, then please read along.
Guilherme: What are your gaming habits? Do you prefer solo games over multiplayer ones?
Morten: Being a spare time challenged introvert, I’d say I prefer solo games over multiplayer games, but I also love multiplayer games and have been playing multiplayer games all of my life, while solo games was something I got started on a few years ago.
Solo games are much easier to get to the table than multiplayer games. If I have half an hour in between other things, I can get in a game. That would never work if I had to get the gaming gang together first.
Apart from being an easy way to scratch my gaming itch, solo games give me time to recharge my introvert batteries.
Guilherme: Your Automa Approach is a very streamlined way of devising a solo variant for multiplayer games, even some which would not be entirely compatible with it. What are the main advantages and disadvantages you see in your method?
Morten: I think that the main advantages are that the method in many cases replicates the feel of the multiplayer game – you still have that feel of competing against someone, you retain the tension of a win-lose criteria and the core interactions of the game. At the same time the Automa Approach attempts to limit the amount of work it takes for the human player to run the artificial opponent.
There’s a point of diminishing return, where you could add more of the multiplayer game to the solo experience, but doing so would raise the complexity and workload more than it was worth. Similarly, there’s a point where you could lower the complexity and workload, but the cost in gameplay lost would be too great.
The Automa Approach tries to aim for the sweet spot between those two.
On the other hand, there are of course also disadvantages to the approach. First of all, it requires work from the human player to run the Automa, and most people would rather be playing the game than doing bookkeeping for a cardboard opponent.
The Automas also add complexity to the game compared to just playing for a high score, so there are more rules to learn for the player. For some games, such as Scythe, it’ll add a lot of rules, while in the case of Viticulture, it’s a rather simple system and that solo mode can actually be used as a way to learn the game in a fun way before you teach it to your gaming group.
A shrink wrapped deck of Automa cards for Viticulture. Image credit: Ender Wiggins.
In Between Two Cities we could easily have made a very simple high score mode for solo play. That would have been much easier to learn than the Automa system we made for the game, but to me that would quickly have turned into a samey optimization puzzle.
Finally, it’s worth noting that there are games, where it simply doesn’t make sense to try and make an Automa system, e.g. I dare anyone reading this to apply the Automa Approach to Werewolf or Charades and come up with something sensible :-).
Guilherme: The multiplayer solitary games seem to be the best type of game to respond to you approach. In the age of cooperative games, this is a very critical point, indeed. Do you have any ideas on how should game designers think about their games so that solo variants are imbued within it and strategies such as the automa are not necessary?
Morten: I think that in most cases it will be very hard to do something like that for competitive games, since if you make a competitive multiplayer game that works as a solo game without modification, then you’ve likely made a game with no interaction, which I think would turn off a lot of players. Even games that are labelled multiplayer solitaire has some kind of interaction – I can’t think of a single example of an interesting multiplayer competitive game with absolutely no interaction.
That said making a game with low interaction generally lowers the amount of changes you need to make to a game before it can be played solo.
The only viable way I see to make a competitive multiplayer game that works unmodified as a solo game is to sneak the Automa into the multiplayer game as you allude to in your question. This could be in the form of an “environment” for the game that does player-like actions. E.g. it picks up some of the available resources, makes raids on human players, blocks territories. This environment would then make sure that there’s an interesting challenge even if you play the game solo.
As far as I know Snowdonia is a clever example of doing this, but I’ve never gotten around to try it out (I really really should try it, though), I’ve just read the rulebook.
Other than that we of course have cooperative games, which in many cases can work unmodified as solo games. The Pandemic games are examples of this.
Guilherme: What other kinds of AI do you find interesting? Could you give a few examples of games which adopt these mechanics.
Morten: Well, that depends on what your definition of an AI is :-)
My favorite solo game is Dawn of the Zeds and you could say that you’re playing against an AI in that game, but it’s not an AI that’s taking the place of a human player.
Hostage Negotiator is a similar example of a very nice game, where you play against an artificial opponent that doesn’t play the part of a human player. In that game you negotiate with a hostage taker, which has a rudimentary personality, a shifting mental state, and takes actions.
The PnP solo game Maquis, is a worker placement game that has an AI that places blocking workers just like Viticulture Automa does (Viticulture Automa is actually inspired by Maquis), but it’s not specifically replicating another human like Viticulture’s Automa does, instead it’s presenting an obstacle to carrying out the two missions that represent the game’s win-lose criteria.
At the Gates of Loyang is another game that I love. It’s a 1-4 player game where there’s a rather neat mechanic that simulates that another player is taking away cards from the pool of available cards and there’s sort of a hand of cards that you can steal from. Other than that there’s no opponent to play against. So the solo game replicates some of the main interaction points of an opponent, but you don’t compete against it and instead play against your own high score.
On the other end of the spectrum there are games that not just tries to replicate the interaction points of a human opponent, but almost fully replicate one with a rather advanced set of rules for the cardboard opponents. Examples of this approach is the COIN series of games. I’ve never tried such a game, since the time investment required to learn those games are simply too high for me currently.
That kind of AI seems interesting to me, but by going that route you make the game inaccessible to most players and personally I prefer to spend my game time taking my own turn instead of managing an AI. Please read that while remembering that the criticism is coming from someone (me) who hasn’t played that kind of game, and thus the criticism should be taken with a huge scoop of salt :-).
A Distant Plain with AI flow charts. Image Credit: DoM Lauzon.
Guilherme: What do you think will be the next innovations in board game designing for solo gamers?
Morten: I’ve considering doing adaptive and/or Legacy-style AIs that evolve over time depending on what happens in the game and what playstyle you have.
So, if you for example attack a lot, then defensive or counterattack cards could be added to the AI deck, and if you win a game by razing the AIs home town, then some cards would be added to the deck for the next game that would make it go harder for your home town to get revenge.
Other than that I’ll refer you to a couple of posts on my blog that deals with possible innovations in AIs:
1) What if a game could hate you? https://boardgamegeek.com/blogpost/42307
2) Blowin’ in the wind: Trying to breathe life into your cardboard opponent – a guest post by Jan Schröder. https://boardgamegeek.com/blogpost/42638/
Guilherme: How do you see the adoption of apps, tablets and cell phones to the analogic games? Do you think this could pose a threat for the Automa Approach?
Morten: I can definitely see, how digital board games could pose a threat to The Automa Approach and solo board games in general. I think that solo board gaming is more threatened by digital board games, than multiplayer ones, since a digital game can replicate most of a solo board game except the tactile experience, while for multiplayer games the digital versions are also missing the social aspect, which is very important for many players.
I can see someone playing a digital version of a multiplayer board game and then buying the physical version to play with his friend, but for solo games I think that’s going to happen more rarely. I, for example, played the app version of Levée en Masse so many times, that it seems unlikely that I’ll ever get the physical version to the table. Similarly, there’s a good chance that I’ll never buy Sentinels of the Multiverse, because I have the app version.
For me, though, the tactile feel of the game makes a big difference, and more importantly a board game takes me away from email, BGG, facebook and all those distractions. When I play a game on my iPad I tend to switch back and forth between the game and checking email etc. When I play a physical board game, I put away electronics, which means that I get to unwind much better and get more engrossed in the game.
Thus, I don’t think that digital games will replace physical solo games, and interestingly we’ve seen a rise in the sales of board games and the number of solo playable games in the past decade, where we’ve seen tablets, smart phones, apps and digital board games take off. So luckily there’s nothing that indicates that digital board games will replace physical ones, no matter whether they’re solo or multiplayer
Guilherme: In your text, you were very critical towards games in which you are constantly trying to "beat your own score", such as Agricola. How do you think this could be changed for those games?
Morten: I might have come off more negative than I really am. I for example actually like Agricola solo, but I think that it could have been much more fun with an opponent and I have a secret dream of being allowed to make an official Automa for Agricola :-). The is a prime example of a game whose solo mode could be improved by having an artificial opponent and I think that it could be made in a manner fairly similar to that in Viticulture.
In fact, you could say that The Automa Approach is my answer to your question. It’s the way I propose to take solo versions of multiplayer games beyond high score modes. That said, it’s hard to give a general answer, since the way to go will depend very much of the specific mechanics of the game.
When I play Agricola in the solo mode I can predict most of what will happen and there’s no risk as I wait for the resources to pile up, because there are no one else to snag them up before I do. Thus there’s little tension and each game feels like a fairly static optimization puzzle.
It’s important for me to stress that I do like the solo mode for Agricola and I ranked it as number 16 on my top 20 solo games a few months back – it just seems that to me personally the solo mode could rival the greatness of the multiplayer version instead of “just” being good, if it had an artificial opponent added.
Guilherme: What should we expect from the solo variant of Scythe?
Morten: Hopefully a fun game experience :-)
Scythe Automa provides a 1 on 1 experience, where you play against one Automa that tries to mimic the key interactions of a human opponent. I’ve also put some rules up on BGG that allows you to play as up to 5 players with any combination of Human and Automa players, but that hasn’t undergone playtesting, so it won’t be included in the rulebook.
The Automa in Scythe is quite a bit more complicated than the Automa for Viticulture, so there’s a definite learning curve.
Viticulture’s fairly simple worker placement system lend itself well to a simple system for picking where the Automa places its pieces on the board. Scythe on the other hand has a much more complicated board with lots of freedom of movement. The board and its freedom of movement represent some of the major interaction points of the game and the positioning of your units in relation to those of your opponent is very important.
We needed to deal with this fact in some way and basically we discussed two approaches:
1) Come up with unit movement rules that had some intelligence.
2) Take the “Horde approach” of AIs by making the opposition extremely stupid, but giving it hordes of units.
Approach 2 would make the game feel very different from the multiplayer game, so we went for approach 1). As a aside, I can say that the system we made was partially inspired by John Conway’s Game of Life (https://en.wikipedia.org/wiki/Conway's_Game_of_Life).
We’ve actually discussed also taking approach 2) and make a coop expansion for Scythe, where you together fight against mechanical hordes sent out from The Factory that at the heart of the game’s board and universe. We’ll likely never get the time to make that happen, but it’s something I’d like to do
Guilherme: What are your favorite games nowadays?
Morten: My favorite game is Dawn of the Zeds. It’s the most cinematic-hanging-on-by-the-skin-of-your-nails-in-barely-controlled-chaos game that I’ve ever played.
My favorite game, Dawn of the Zeds. Image Credit: Me .
Other favorites include Onirim with its combination of simplicity, variability and cozy vibe, Mound Builders for being the mechanically strongest game in the States of Siege series (a series of tower defense games), and finally Lord of the Rings: LCG, which has a ton of variation via the numerous expansions, which gives me a nice sense of adventure.
On the multiplayer front I’m fond of Puerto Rico, Tigris & Euphrates, Warhammer Invasion, and Blood Bowl.
Guilherme: What can we expect from Morten in the near future?
Morten: Not much, I’m afraid - unless you’re following my blog where I sound off something like once a week currently :-).
Between Two Cities and Viticulture Essential Edition were released at the end of last year with solo modes I worked on. The next release I’m involved in will likely be Scythe, where I worked on the solo mode. That should be in backers’ hands in August this year.
I’m also part of a team working on an expansion for Euphoria, but that’s going much slower than planned, so right now, I can’t say when that’ll be out.
There’s also a small side project that’s being playtested by a publisher at the moment, but I’m sworn to secrecy on that one :-) and I don’t know whether it’ll end up getting published or not.
Finally, my two partners (David Studley and Lines Hutter) and I have a few games in various stages of development, but none of those are ready to be shared with a publisher yet, and these are started as projects for learning the craft of game design. If one of them should end up being publishable, then that’s great, but that’s not my initial intention with them. Instead our focus is the projects that we do for Stonemaier Games.
That said, I’m very happy about a specific one of them, which I plan on finding external playtesters for soon, since and I think that it’s the best game I’ve made so far. It’s a simple solo/coop card game that started life as marriage of Onirim and a free print and play game I made back in 2012 called Endless Nightmare.