It's modeled physics, so it's also scale dependant. I don't know about his
explanation, but to a degree I can concede that certain things might have to
change. The same physics model doesn't always work when you change scales,
for example. A bee the size of a person would not be able to fly, so I
would have to model his wing (tinsel) strength and muscle mass to about 100
times that which the actual scale allows. Then you have to consider which
physics profile you're going to follow from there. Stop at fudging the
muscle mass or continue along that line? If you continue, you've got to
change his eating habits to supply enough energy for the increased muscle
mass, which means he'd spend his entire life eating, and nothing else. That
doesn't even account for heart rate and size needed for the delivery of the
nutrients, which also don't work when kept to scale. It can get complicated,
but in the end that bee just doesn't scale well and if you want to make it
work, what you end up with doesn't resemble a bee very much.
Appearance is another thing that changes with scale. When I was doing a lot
of train modeling in HO scale, there were places that called for, let's say,
1/2 inch stone covering. In scale, each stone is then .0625 inches. On the
layout, you couldn't even see the texture and it looked wrong, so you cheat
a little by making the stone larger. It looks perfect unless you have a
scale person standing near it and some picky friend points out that the
pea-stone is the size of the person's hand. To model a train line that was
supposed to be a few miles away on a mountain, I'd change everything to N
scale to give the impression of depth (otherwise it's hard to make 20 feet
of layout look like a few miles). Is it proper scale? No, but it sure looks
better and more realistic.
I know that computer race cars aren't bees or model trains, but modeling
physics is still modeling. A road surface at 150 affects the handling of a
car. That can be modeled, but what about the shadows cast by grandstands and
other things that cool the temperature on certain parts of the track? The
track temperature should gradually cool until the center of the shadow, and
each shadow, being a different size, would have different temperatures. Take
this and a hundred other nuances, one thing affecting another, and I doubt
that we have enough processing power to model all these changes. Does that
mean that the physics model is flawed? Certainly.
There are limitations on modeling physics; many due to technology, and some
just because things don't scale well. I think we shouldn't look for, or
expect, or even want a pure physics model. Maybe the best thing a game
developer can deliver is something that makes us FEEL like the experience is
real. If fudging the physics makes something seem more realistic than a
proper model would, I'm all for it. I like realism, but I don't want to
feel the pain when I crash in GPL.
--
Slot
Tweaks & Reviews
www.slottweak.com
