Re: PBM design and formulae - a fresh start

From: markiel@callisto.pas.rochester.edu (Andrew Markiel)

Date: Mon, 05 Jul 1993 09:28:53 +0000

In article <C9LJn2.6FF@dcs.ed.ac.uk> gym@dcs.ed.ac.uk (Graham Matthews) writes:
>
>To whomever it was who suggested that human tennis players do know the
>equations of tennis ball motion, maybe not in mathematical form, but
>in terms of the neural net that is their brain, I would like to offer
>the following.
>
>The key point here is whether they know the equations *exactly*. And it
>seems to me quite clear that they do not. Assuming that players know the
>equations exactly then they would play perfect tennis and hence no tennis
>match would ever have a result.

No, because (1) the players don't have time to do all the explicit
calculations (certainly reflexes are an important part of tennis).
(2) even if the player knows exactly how to hit the ball (and
I would argue that an expert tennis player has a very good idea)
that does not mean he will hit it properly: humans don't have
perfect control, and they get tired. And (3) a perfect ability
to hit the ball does not help if the ball is not within reach:
this requires some element of planning to be in the right place,
and still might not be enough if the physical ability of the
player is not enough to reach the ball (knowledge is no help here).

Of course, this is compounded by the fact that the player does not
exactly know the parameters involved, but I argue that the player
knows them closely enough (assuming he's thinking quickly) that
knowing the exact parameters is of no value: does it really make
a difference in your shot if you know the ball is 12.365 inches above
the ground, instead of "about 1 foot"? Thus it doesn't matter if
you give the exact formulas or not.

I will agree that *if* the simulation is an accurate rendition of
tennis, then there is no need to give formulas, clearly since there
are people who do not know the physics equations, but can still play
tennis well. However, I argue that this is because the exact formulas
do not give you any additional useful information in a real-time
simulation of tennis. It is only in the PB(E)M game, where a player
does have the time to do the calculations, (and also, one wonders
how you would simulate imperfect control), that the exact formulas
would give you an advantage. But in this case, the player could
carefully analyze the data and get as close an approximation to
the formula as desired. And, of course, this wouldn't be a very
good simulation of tennis (and on that point, I'll agree with you).

>--
>Graham Matthews                   Go west, go west

-Andrew
markiel@callisto.pas.rochester.edu

References

PBM design and formulae - a summary
rob@ccc.govt.nz (Rob McNeur)
Mon, 28 Jun 1993 09:53:02 +0000
Re: PBM design and formulae - a summary
gym@dcs.ed.ac.uk (Graham Matthews)
Thu, 01 Jul 1993 11:40:27 +0000
Re: PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Sat, 03 Jul 1993 16:11:25 +0000

Referenced By

PBM design and formulae - a summary
rob@ccc.govt.nz (Rob McNeur)
Mon, 28 Jun 1993 09:53:02 +0000
Re: PBM design and formulae - a summary
gym@dcs.ed.ac.uk (Graham Matthews)
Thu, 01 Jul 1993 11:40:27 +0000
Re: PBM design and formulae - a fresh start
gym@dcs.ed.ac.uk (Graham Matthews)
Sat, 03 Jul 1993 16:11:25 +0000