Game Designers in Detail: Samantha Kalman

Game Designers In Detail is a three-question interview series between a current NYU Game Center MFA student and a PRACTICE speaker.


Charles George: Music games generally fall into two distinct camps. One type of game allows for personal expression but, at the cost of goals and structure (eg: Electroplankton).  The other generally uses the rhythm aspects of music to create structure but players never participate in the creation of music (eg: Guitair Hero).  Sentris attempts to meld the well defined structure of Rhythm games, with the creative freedom of musical toys. How did you approach bringing those two disparate styles of game together?

Samantha Kalman: One of the reasons Rez was so good was I felt like I had some control over the music from the very first moments. It left me wanting more; more influence over the music, more decisions about instrument choice, more manipulation of every kind. I also found through playing it obsessively that attempting to achieve the highest scores meant restricting the variation of musical output in some very frustrating ways. I wanted to play a game where mastery over the game would afford greater control of musical variation, not less. This was basically the starting point.

In my experiences playing music with others, I learned that song structures emerge in a loose fashion. Each musician might have A, B, C, and D parts that they play in a song, but the order in which they progress (e.g. ABABCBCBDBABA) probably changes from person to person. There are also smaller variations within each part, such as A1 or A2 where a particular rhythm or chord are different. This revealed the fundamental building blocks to any given part of a song as a collection of notes. It can be fun to come up with variations and rhythms when playing within these confines. In other words, to give the player a relatively small sandbox of sounds to play with in whatever way they choose. That seems to scratch the itch – enough – for freedom of personal expression.

The other part of what makes a song feel complete is a sense of progression, or movement through a sonic landscape. A melody has to end before a chorus can begin. I saw a parallel between musical progression and level progression – players need a sense of completion to enjoy what they’re doing and reinforcing they’re doing it “right”. So I tried to define a couple of very loose requirements to satisfy any given section of a song/level:

1) Areas in which to place a pre-defined note or notes

2) Areas in which to place a pre-defined instrument

3) Areas to fill with a freeform collection of any instrument or note

These rules seem to work — they give the player a goal to work toward, to try different things, and to see how the systems of the game transform their actions. The entire point is to allow the player to actually BE a musician, not just pretend to be one. Maintaining this philosophy helped evolve the rule set to allow for a high degree of musical freedom.

CG: Games that rely on musical or artistic skills often run into issues where users without creative skills feel left out. What did you do to draw those users in?

SK: Honestly I think the current version of Sentris still suffers from this problem to a high degree. During the kickstarter I tried to use the buzzphrase “unleash your inner musician”, hoping that people would interpret that message as “everyone has a musician sitting inside of them that is dormant to some degree and this will stimulate them from slumber”. I don’t think that message really caught on. Instead what’s happened is existing music game enthusiasts have really become attached to the game. Even now when I ask random people to play test it, I regularly receive a response of “I’m not musical at all”. So there’s still some work to be done in an effort to communicate that this is a game as simple as Tetris that will enable musical expression in everyone.

Probably the best decision I’ve made to make the game accessible to more people is the rejection of a game over state. I did this for a couple of reasons. People who are interested in making music are probably highly self-conscious about it. To have a game that’s designed to help you make music suddenly tell you YOU FAIL START OVER could be extremely discouraging. I’m working under the assumption that people who are musically inclined but not confident are very likely in a pretty vulnerable state. My secret mission is to help people feel good about making music so they’ll want to do it more. Early on I made the decision to exclude any kind of direct negative feedback from the game. Now, people are still highly self-critical and sometimes they still manage to perceive negative feedback from the game if they don’t like the music they’re hearing. But the game itself never judges what they make or applies pressure to do better or go faster. It gives everyone space to go at their own pace. It turns failure into a self-imposed state. In watching people play, I’ve been surprised at how quickly and harshly they can judge themselves. I’m still trying to figure out the right shape of ramp-up/reward curve to keep the game fun and interesting to people who don’t think they’re musical. I haven’t solved this problem yet.

CG: Did you find that the abstract visuals hindered or assisted in getting players to understand the core systems, and how did that change how you tutorialized players?

SK: The abstract nature of the game has definitely been a huge barrier for many. A small percentage of players are immediately engaged and really proactive in figuring out what each of the visual elements are. Many more will begin and continue to play without fully understanding what’s happening.

The current tutorial introduces each part of the UI and controls one element at a time. I try to give the player plenty of space to learn about each new element as it’s presented. Players spend about 3-4 minutes just becoming comfortable with the UI , the three buttons, and the musical systems before they are introduced to the puzzle elements. Feedback has revealed that it’s an extremely dense tutorial and I could spread it out even longer. I’m working on a new approach that minimizes text and gives the player an entire song to just learn about one UI element at a time.

I’ve also been allowing myself to continue changing the UI in an attempt to make it more self-explanatory. I’d love players to understand the game just by looking at it, and I’m working in that direction.

There’s another thing I’m compelled to share: the abstract visual structure of the game is completely new. Players don’t have any kind of grounding in memories of past, similar systems to rely on when trying to understand the game. I’ve seen that some people’s brains just need a little extra time to process and understand what’s happening. At GDC, a handful of times I watched players come try the game and fail to understand it. They would leave, but they would come back the next day and immediately understand everything that’s going on. It’s almost as if they needed to sleep on it and let their subconscious process the first experience they’d had before they became able to read everything that’s going on. It’s fascinating from a psychological perspective. I have no idea how to use this to improve the game experience, though.