Re: PBM design and formulae - a fresh start

From: gym@dcs.ed.ac.uk (Graham Matthews)

Date: Tue, 06 Jul 1993 11:48:17 +0000

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

In other words any game that claimed it was modelling tennis to any
sensible extent should include imprecision as one of the key factors in
the model. Thats what I have been saying all along. Now knowing the
exact formulas precludes imprecision.

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

In fact it would make a very large difference because above a certain height
some shots get hard to play because of the player's lack of muscle
strength, tendon strengths etc. The thing is discrete - below height X
(different for each player) a topspin backhand is easy to play, above
it is not.

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

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

You seem to be restricting what PBM games can be. There is no a priori
reason why a PBM version of tennis could not simulate the imprecision
in tennis? You need to make a careful distinction between real time
and game time but there is no reason why this cannot be done in
principle.

graham

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
Re: PBM design and formulae - a fresh start
markiel@callisto.pas.rochester.edu (Andrew Markiel)
Mon, 05 Jul 1993 09:28:53 +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
Re: PBM design and formulae - a fresh start
markiel@callisto.pas.rochester.edu (Andrew Markiel)
Mon, 05 Jul 1993 09:28:53 +0000