Evolution of Steel Seal CSV's

When designing the difficulty levels for Steel Seal, our team used Comma Seperated Value’s or CSV’s to dictate what spawned and how often. Throughout the life of the game there were different versions of these values that dictated the spawning rates of each object. In this post, I will compare the original versions of the CSV files to the ones that shipped with the game, and explain the changes that I made to them.

Pictured below are two different versions of the CSV’s of difficulty 1. The first picture is the first version of this sheet, and the second is what was shipped with the final game. The first thing you probably notice is that the original has a lot more lines to it compared to the final. The original CSV had all the spawners on the lowest difficulty, even though some things weren't spawning at all. This was unnecessary memory usage which was removed in the final build. Another huge difference is the amount of enemies spawning, as well as what enemies were spawning. Originally, all enemies were spawning in difficulty 1, and the spawn rates were fast, which meant players were only lasting a couple of seconds per play through. After watching friends and peers play the game, they would get frustrated at how fast they were losing. We didn't want Steel Seal to be a frustrating game to play, so to improve this, I reduced the Sharks spawn time from 3-5 seconds, to 4-7 and the Orcas and Polar Bears being harder enemies, I pushed them back to a slightly higher difficulty.

Another piece of feedback I received from peers was that the amount of snacks spawning was very low. With them only spawning once every 4-7 seconds, there were large spaces in time where you wouldn't actually collect a snack. To improve this, I greatly increased the spawn rate to every 1-2 seconds. This meant there is almost always a snack on screen that the player can collect. The amount of snacks spawning was originally very low because I based it off of the player collecting every snack that spawned. This assumption didn't take into account the snacks that would be missed because of obstacles and enemies. This increase in snacks made it that a player could miss a snack, and not feel like they are missing out.

 Difficulty 1 - First Pass

Difficulty 1 - First Pass

 Difficulty 1 - Final

Difficulty 1 - Final

Unfortunately, I don't have an original version of the Difficulty 2, but I managed to recover a difficulty 3 original CSV. 

The first image below is the first pass of the third difficulty, while the second image is the final version. In this difficulty, you can see that unlike difficulty one, there are a more spawning in the final version then there is in the first pass. This is because of how  we made each play through feel different to the last. In the original sheet, each object had its own spawner, with varying times and chances of spawning, but the way it was set up, each play through was either very similar to the last, or felt like a completely different level. Because it wasn't a 100% chance that the shark and orca would spawn, you could go for 6-9 seconds without even seeing an enemy. Players said that it felt weird having the huge gap, and one even going as far as saying they thought they bugged the game out. To remedy this, I made the shark spawn 100% of the time between 3-5 seconds. I then increased the chance of the orca spawning to 50%, which doubled the chance of seeing one. The polar bear originally spawned between 20-30 seconds, which in some cases was longer then the entire difficulty, which meant he didn't spawn at all. We fixed this by making the ice spawn every 7 seconds, with an 80% chance of it having a polar bear on it. This added some needed variation to the ice caps in the game.

Another thing we recieved an overwhelming amount of feedback on was the spawn rates for powerups and snacks. Originally, it was programmed that as the difficulty rose, the amount of snacks was dropped and so was the amount of powerups. This would have been okay, but the amount of time it took for a powerup to spawn, stacked with the chance of it spawning, meant they saw nearly no play in the later levels. This was especially noticeable for the powerups. This is definately not what we wanted, so in the final version as difficulty increased the spawn time was dropped dramatically, but the chance of the spawning was kept around the same level. This meant you saw almost 3 times as many of some powerups.

 Difficulty 3 - First Pass

Difficulty 3 - First Pass

 Difficulty 3 - Final

Difficulty 3 - Final

In the early versions of the game, there were only 3 difficulty levels. Because there was only 3 difficulties, there was a dramatic change in spawn rates moving from one to the next, and because there were only 3 of them, the time the player was in a specific difficulty felt like a long time. To fix this, we added 2 more difficulty levels in. This smoothed out the difficulty curve and made the change a lot less obvious, as well as shortening the time on each difficulty, making the game feel less repetitive. 

Over the 13 weeks of trimester, the CSV's changed dramatically. This was from the help and feedback of peers and team members, as well as my skills at balancing were improving throughout the trimester.