- Getting a Feel for Your Attributes
- Determining the Granularity for Numbers
- The Tension Trick
- Searching for the Right Numbers
- Further Steps
This chapter is from the book
This chapter is from the book
This chapter is from the book
Searching for the Right Numbers
Once you decide on the granularity of the numbers you are going to use, it’s time to start plugging in numbers. If you have already described the feel you want with the numbers and determined the number of digits and ratio you want to use, you can do a rough pass immediately.
Keep in mind when doing a first pass at data numbers that they will almost certainly not be what you end up with. This is okay and to be expected. Until a game is tested, it is impossible to know the exact effect numbers will have on the game. Don’t think of this as failure; instead realize that you can take the pressure off the first pass. If you approach the first pass knowing that the numbers will be wrong, you don’t have the stress of trying to guess right the first time. Instead, you can just get some numbers in there. Use the targeted number of digits and rough ratios for each object and just plug them in.
Let’s go back to our racing game example from the beginning of the chapter. Say that you want to make a very simple, new-audience-friendly game, so you want to stick to single-digit numbers. This is what you came up with earlier for what the speed and acceleration should be:
Sports car: Good acceleration and good top speed
Muscle car: Fastest top speed but with less acceleration than a sports car
Motorcycle: Fastest acceleration but lowest top speed
Based on this list and the fact that you want to use single-digit numbers, you might assign the numbers shown in Table 11.1. Are these numbers right? Almost certainly not. But they’re a start.
Table 11.1 Basic data table
Car |
Acceleration |
Top Speed |
|---|---|---|
Sports car |
8 |
8 |
Muscle car |
6 |
10 |
Motorcycle |
10 |
6 |
When testing numbers, it’s a good idea to go beyond reasonable, expected numbers. To find the extents of a range, you must exceed those extents during testing. You want to try making something with too much acceleration or a speed that’s too low; for example, you might experiment with your numbers as shown in Table 11.2.
Table 11.2 Experimental data
Car |
Acceleration |
Top Speed |
|---|---|---|
Sports car |
8 |
8 |
Muscle car |
1 |
15 |
Motorcycle |
200 |
10 |
These numbers are undoubtedly wrong—and, again, that’s fine and expected. You are not trying to get the numbers right at this point. Instead, you are trying to understand your game and game engine. Can the engine handle an acceleration of 200? Does this number cause the game to crash? Does collision still work? By testing unreasonable numbers, you can understand the game and engine better, which will make it more likely that you will find interesting and exciting new results.
The great news is that with game data, there is nothing you can do in testing that can’t be undone. You can use this aspect of game making to your advantage for wild and interesting tests. Once you have broken the game in interesting ways and understand the mechanical workings better, it’s time to home in on the balance you truly want.
The next step is to test and test and test—and then tune and test more and then do more tuning and testing. On this first round of testing, the goal is to get the numbers to emulate what you wrote in your original list of what you feel you want from the numbers. Does that motorcycle feel like it has great acceleration? Does the sports car feel like it has slower acceleration but can eventually top out at the highest speed? Eventually you will find the right balance with the numbers.