Designer Interview: Jeroen Stout of IGF Nominee “Dinner Date”

[Editor's Note: The following is a design-focused interview with Jeroen Stout of the fascinating Independent Games Festival Nominee "Dinner Date".  If you haven't heard of the game, be sure to check out the Youtube video above for a primer.]

The Game Prodigy: The concept for Dinner Date isn’t something that’s normally explored in games.  What was your process for coming up with the original concept for Dinner Date?  What core experience did you want the player to have?

Jeroen Stout: I was much impressed with seeing the games The Graveyard and Dear Esther. After seeing these games I suddenly felt I could rethink my entire approach to games. It was not directly the content of either of those, but rather that they did not act as-if it was needed that the player would do big, bold actions. They offered short stories which you could experience by doing a small action to progress through it. This really enticed my imagination, but it took me a long time and a few projects before I could really work on it. When I was studying under Dr Pinchbeck in Portsmouth I thought it interesting to make a game in which you could not even walk and rather only sit at a table. The original concept came into being initially almost as a joke but soon became quite serious. The most important step was realizing it was sensible to do harmonic actions – to allow the player to do things which correlate with the story – and to treat it as-if it is completely normal. I wanted to make with the story more of a literary experience – to pick an interesting intellectual theme for the story and explore something of the character and to back this up with interesting actions to do while listening to him. In a way it is playing like an actor does – taking on the role based on a script and doing things which allow him to pretend to be a character. This type of interaction is in some sense a bridge between acting and observing. This is an under-represented part of games, sadly, but it has been my interest ever since reading Rogier Callois and discovering mimicry to just be a part of games – and it is interesting to focus primarily on makings games doing exactly this.

During my production I wrote theoretical work on the experience of being in someone’s mind by doing his actions, which correlated with the experience I had thought of for Dinner Date. It revolves around the sensation of a game influencing how you perceive your actions – you may feel strong if you play a strong character but you may also presume yourself to feel happy if you play happy character. Thoughts on how to achieve stronger sensations of this kind and a search towards a theoretical cause can be read in a paper, which is available on my website here.

TGP: What was your design production cycle like?  (Did you incorporate frequent playtesting, redesigns, or prototypes, or just follow your original doc?)

JS: There is a good lesson in this, because it was a little bit of a confusing development. It started as a prototype I would ‘perchance’ release, so I made some very basic visuals and gameplay. I wrote the story under the impression this was an experiment so I indulged myself in my then emerging deep love of literature and wrote the story as a piece of romantic-realistic fiction. When I let some people play an original version and they liked it I considered the game really ought to be released as part of the experimental scene. So I worked further on finishing the game in terms of animations and visuals. But I came to a point where I concluded that doing the game with ‘experimental’ stamped all over it would not embody my ideology – that the game is a medium for making something playable, rather than drawing attention to its ‘odd nature’. The story already had been written as romantic-realistic and I felt I should do that honour by going the extra step of making the whole game in this style and go to the length of selling it. So the kitchen is ‘sort-of’ realistic with many appliances taken from real existing models, and I started adding things such as the main menu version of it and a more detailed version of the outside world. I contracted a very good composer, Than van Nispen tot Pannerden, who complimented the game with a very apt soundtrack which certainly adds to it.

So looking back on this development I can see that once I decided to sell the game it was simple because all I had to do was improve it to a point of ´good enough´ realism and I could already confidently expect some people to like it. But during the various phases I repeatedly fell into the trap of being overly cautious and doing many ‘try-out’ versions. In some ways this is good, but I think the lesson is to decide whether or not to make this a ‘real’ game sooner rather than later. For instance, my next game will have a very lengthy prototype phase because I am doing something new of which I first have to convince myself it will work – but this time around I know eventually I will release it and I can lay groundwork for the eventual production phase and try to ensure I will have to re-do as little as possible.

TGP: Can you share a major design challenge did you faced while developing the game, and how you worked through it?

JS: One of the main problems was a relative lack of context. For instance, I am aware of resemblances between Heavy Rain and Dinner Date, but at the same time Heavy Rain is a lot more intense, whereas Dinner Date is subdued. Everything in Heavy Rain hinges on a crime plot, in Dinner Date everything hinges on a character-centric plot. In terms of writing, the Dinner Date type of story was far easier for me because I quite simply prefer to read this type of stories myself so it comes naturally to write in this manner. But in terms of visual style and overall feeling it is very difficult for me recognise when it is ‘good enough’. For instance, perchance the game would have been better if I spent twice as much time on modelling real-life kitchen objects. But perhaps this would have not been noticed. Perhaps I should have aimed for more realism in lighting and texturing like in Heavy Rain but perhaps I have been right in saving that for a game not set in a single room.

I think the design challenge which thus still exists for me is that I cannot always see how others do it for some content beyond the abstract. The ‘8bit’ style is great for some because it is a cultural acceptance of extremely barebones, cheap graphics and allows experimentation of gameplay at low visual cost. At the other end you would have to look hard not to notice trends in how genres like role-playing games and shooters look, so doing a new story within that genre means competing with the established ones.

But in Dinner Date you eat bread while listening to a soliloquy – whether bread or soup need physics in this context, to name a deceptively complex question, is something I still do not know. It differs for people as well – there was a strange tone in a review which suggested I had polished too much whereas I myself regret not adding little touches like proper fluid physics to the soup.

A large part of this is just trying things out, which is resource-intensive solution. The idea with me remains that if one brings new content one also has to be confident to take a step in an unknown direction – and one does not have to reinvent the wheel, just reapply it from other media. Paintings have meaningful glances, books have character-driven monologues. There is a wealth of classical knowledge which we may apply again in making games which are truly emotive. Gaining the mindset of an author and painter while retaining the knowledge of a game-designer is so far my favourite ideology for making games.

TGP: What big design successes did you have for the game?  Anything that worked that surprised you?

JS: There are certain parts people have commented on that it gave them a first-time experience of being in someone’s head. It is interesting because my theoretical interests lie in this area but for Dinner Date I had not actually planned it. I think it mostly happens with the animations I made later in production. For instance, in my original script Julian mostly sits at his table and I felt I had exhausted all the actions on the table and I had to do something fresh. The resulting actions became some an establishing moment in the game, a sort-of confirmation that it is real.

It is an interesting aspect that playing with the amount and type of actions has an even larger emotional effect than I thought. If I would do another game with the exact same Dinner Date technology I would include a lot of more personal and emotive bodily actions – fatigue, joy, getting tired while running to catch the train but also things like brushing your teeth and stroking a cat and work far more on the rhythm of actions – like giving very lacklustre versions of already existing actions when a character is suddenly disappointed or very expressive ones for when a character is exalted.

TGP: What reusable design lessons did you learn from this game?  Did you learn any new design tactics from this unusual genre that may be applied to other games?

JS: Something which has rather struck me is that there is a serious fault with judging the inclusion of actions based on whether they are ‘essential’. Rather we may select actions based on whether they will add to the overall experience. For instance, there are plenteous games where there is some quick time event which prevents you from being killed. A monster jumps at you and you press ‘e’ repeatedly until you see yourself kill it. But the same game during a relaxing walking section will not allow you to sit down on a couch while listening to a character, even though that is something quite sensible, or to pat a friend on his shoulder when he tells some something which upsets him. It seems doing those things is less essential: after all patting a shoulder does not prevent yourself from being killed. But I believe that doing these things will add a lot of humanity and realism to such a game.

I think an interesting perspective is to look at non-interactive segments and ask oneself: why is this not interactive. For instance, having two ‘playing with hair’ animations allows one to give minimal interaction to a character lying on a couch with his lover. All the while plot exposition can be given in the dialogue while just letting the player switch between these animations at will until the scene ends. The question is: why would this scene not be interactive? Why would I choose to remove the option of playing with hair and force the player to watch what essentially is a video which is merely rendered in real-time? Why not rather increase the amount of choice and add subtlety to the motions which indicates that they both fear this may be the last time they meet before they part?

I believe the experience of making Dinner Date has taught me to stop thinking about actions in an ‘essential’ way and rather wonder about how to make anything interactive. For me that means making games attempting to recreate the experience of life for a person as part of a larger story – in essence it is for me an absolute requirement you feel you are in their body and do their things, especially if that is something so intimate as brushing your teeth and cuddling your children. While I understand many games have no such goal or do not even remotely feel interested in doing these things, I still think it peculiar that when it comes to humanizing characters in any given genre you may give me a ‘talking with your buddies’ scene, set it in a pub, and then proceed to refrain from giving me control over drinking the whisky. I will think it a great first step if I have to do something to see myself sip. Things may go from there, but it can simply be a good first step for many games to start opening up this area of interaction.

In essence, I think it is best to prefer crude interaction which gives me a hint of being that character over watching a painstakingly produced 3d cut-scene which may as well be pre-recorded for all the interactivity it gives.

TGP: Do you want others to make games like Dinner Date? How will you use your Dinner Date knowledge in your next games?

JS: Something which I am moving towards now is avoiding granulation. It is a question of design: I want the player to have the freedom to do things but also have to combine it with a narrative and actions with other characters without having a clear sense of causality. In reality many permutations of the same event are possible because the world is highly complex, so one less quickly starts using game theory (in the social science sense of the word) on the world, whereas technology forces us to make far more coarse-cut triggers in our games and people are tempted to start playing on causality rather than mimicry. Currently I am working on a way of letting the player control an AI which in turn controls the character. Such a no-direct-influence scheme maintains my interest in characters doing what they are naturally like and giving players a chance to view things from their eyes while not making it impossible for players to contribute.

It would be interesting to see other tries at such things, truly. I certainly want to see others make games like Dinner Date, I think it would be difficult for me to advance if I never see any games like it. In a very limited way this already happens, but I would be very interested in character-centric games using novel interactive forms. From a developer’s perspective it will save time because I will not to have to try out brushing teeth and see how it works in a different game. But from a more intellectual perspective I would quite simply just intensely enjoy to see more of that type of games because they seem very enjoyable to me. Sometimes I see someone walking the streets and I think how intense it would be if we would see an intellectual drama unfold from his eyes – and I very much long to see others do the same with the characters that must inspire them.

Dinner Date can be downloaded at the Stout Games Website.