A hopefully short thread on how to make encounters feel better using workout patterns as a framework.
A trap a lot of people fall into is to make fights with waves and waves of baddies where each wave is slightly harder than the last. If we graph out the difficulty, it looks something like this.
It makes sense. The encounter gets harder as it goes on. The hope is cranking up the difficulty makes it more and more exciting.
But it has a few flaws.
1. The start is MUCH easier than the end. We want to kick up difficulty at the end, but we don't want to lie to players about what they are getting into, starting a bit closer to the end difficulty would allow players to understand the level of challenge they are signing up for.
2. Because the encounter is boiling over a long period of time, its hard to tell that the water is getting hotter. I can tell the difference between an easy encounter and a hard encounter. Its much harder to tell the difference between a medium encounter and a mediumer encounter.
3. The player feels weaker and weaker as time goes on. As the encounter just gets harder and harder, the player never gets to feel that powerful player fantasy they had at the start.
On the other hand, this is a pretty standard peloton workout. The Y axis represents how fast you are asked to ride, the X axis represents how far you are in the workout.
So they don't just ask riders to go faster and faster and faster until the ride is over. There are peaks and valleys in challenge, and the ride still ends at the hardest its ever been.
Lets apply this more dynamic difficulty model to an encounter to see what it might looks like.
In this model, our encounter still ends at a climax, but it has moments where you feel powerful and moments where you work hard. It does a better job of forecasting difficulty at the start, and it creates memorable moments through readable changes in difficulty.
With a more dynamic difficulty model not only is this single encounter more exciting, but its easier to change our peaks and valleys around from encounter and encounter to make the game feel more varied and rich.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
In game design, knowing what you need is a lot more powerful than knowing what you want. Here’s an example from a raid we made in Destiny.
We were working on Wrath of the Machine— a Mad Max inspired adventure. In several encounters we were going to have something new to Destiny, balls you could pick up and throw at things.
Since these balls were going to be in a bunch of different places in the raid, and we needed a device to make them appear.
Here’s a stupid thing I did in college that I hope I’ve learned from.
I was working on my minor in writing— taking a short story class.
These classes all follow the same general pattern, every session a few people turn in stories, the whole class then takes them home to read and write up feedback.
So I get a story about a teenage girl who was in love with some boy who had super natural powers. (He might have been a vampire? I’m not sure...it’s been a long time)
I hear a lot of people say, "I like BOTW, but it isn't a good Zelda game." As someone who has a pretty massive Zelda tattoo, I couldn't disagree more.
Almost every Zelda before BOTW is built on a super hard lock and key model. The games were a series of puzzles where the devs had made sure only one solution works.
There is a strong temptation for game designers to make sure the players have to do "the right thing." You spend A LOT of time making any puzzle, and you want to make sure players get to experience the cool thing. That they get their money's worth.
The number one piece of helpful feedback I hear really talented encounter designers give to new encounter designers is “Use less unit types at once.”
This piece of advice was given to me very early in my career and was very formative. Here are some reasons why.
1. You want your AI lineup to have specific strengths and weaknesses.
Most AI are good at some things, but bad at others. So, there is a temptation to use lots of unit types in order to create a lineup where you’ve covered up any potential AI shortcomings.
Don’t do this.
By creating an AI lineup with obvious shortcomings, the player now gets to play around these.
They get to be the paper to your rock. This allows players to be smart and show their power.