Game Development Reference
In-Depth Information
There's Always Something More
In my recording studio example, at any one time there was a “loudest� instrument.
Most of the time, however, none of them were as loud as they
could be
. That is why
the climbing fader syndrome could happen. No matter what volume they were at,
I could always turn them up a little bit more. Eventually, however, I would have hit
the top of the range for that particular slider. I would have to then say, “It's as loud
as it will go.�
That point
would have been “the loudest.� To make a particular track
louder, the only solution would have been to turn everything
else
down.
There is a difference between the topmost value of a group and the topmost
possible
value. Making note of where this artificial edge is is essential to scaling our
possible values. Sometimes there is a limit and sometimes there is not. In football,
basketball, or baseball, for example, there is no upper limit to the score that a team
can generate. (Sure, you eventually would run into the limitations of physics and
time, but for all practical purposes, we don't consider that a problem.)
There's No Room at the Top
On the other hand, in sports like gymnastics, figure skating, and diving, there is
certainly a maximum that one can obtain. Gymnasts can't get more than a 10.0, no
matter how accomplished they are. (It used to be that figure skating maxed out at
6.0… I'm not sure what they do now. If you want to see multi-attribute utility
theory in action, check out “ISU Judging System� on Wikipedia.) To make matters
worse, there are
floors
where the gymnastics scoring starts: 8.8 or 9.2 depending on
the system. You really have to work hard to merit a score below those numbers.
(I was always curious to see what would happen if a skater just sat down on the ice
and didn't move. Is it even
possible
to get a 0 in those sports?)
Capped systems like this can generate something of a quandary. If you, as a
judge, give a contestant a 10.0 on a routine, and someone else comes along later
who performs
better
, what score can you give that one? The artificial restraints of
the scoring system have limited you to giving the second person a 10.0 as well.
When this happens, you sound a lot like my sister saying that you have just seen
“the best gymnastics routine ever�—from two different people! Certainly that is
unlikely. One of them would have been better than the other—at least
marginally
.
On the other hand, if we work with a capless system, we run the risk of encoun-
tering a phenomenon similar to the climbing fader syndrome. If there is no restric-
tion on how high a score can go, we can always justify a score by comparing it to
others. Now, the second gymnast above could score a 10.1. And the next one that
is better could score a 10.2. Unfettered by the practical logistics of physics and time,
subjective scores can get out of hand quickly. Eventually, it is entirely possible that
gymnasts of the future could be scoring 11, 20, 40, or even in the hundreds!
