RE: Rules etc

[Duncan Louttit <duncan@swallow.co.uk>]

Hi Everyone,

         But is the 7 second mouse better than a 10 second mouse? Why 
should we have a particular penalty time?

         Many actual touches take more than 3 seconds to carry out, is this 
a sufficient penalty in itself?

         Regards,

         Duncan

At 11:43 21/10/04, you wrote:
>Hi,
>         My opinion is that although reliability is important, we should
>not unnecessarily penalise mice that happen to crash.
>
>         The real world is not perfect. Neither are mice, or the maze
>board. A mice might happen to crash due to perhaps the maze gap which
>was bigger than anticipated. The handler restarts it and it goes again,
>and clocks a timing of say 7s.
>
>         Another mice runs perfectly but say at 12s. Would it really be
>fair to penalise the mouse that ran 7s for what might have been a stroke
>of bad luck?
>
>Just my 2 cents worth.
>
>------------------------------------------------------------------------
>Daniel Chia
>
>"Genius is one percent inspiration and ninety-nine percent
>perspiration."
>
>      - Thomas Edison
>
>E-mail: danielcjh@yahoo.com.sg
>MSN: danstryder01@yahoo.com.sg
>ICQ: 37878331
>------------------------------------------------------------------------
>
>
> > -----Original Message-----
> > From: owner-micromouse@cs.rhul.ac.uk
>[mailto:owner-micromouse@cs.rhul.ac.uk]
> > On Behalf Of Martin J.Barratt
> > Sent: 21 October 2004 18:09
> > To: micromouse@cs.rhul.ac.uk
> > Subject: RE: Rules etc
> >
> > Dear All
> > I would like to add my two-pen'th to the 'touch discussion'.
> >
> > Duncan recently introduced the idea of an infinite touch penalty
>(which
> > see) and, it seems to me - with all this talk of complicated rules,
>that
> > this is a VERY simple method of determining which mouse has actually
> > performed the best on the day.
> >
> > It seems to me that there are two sorts of touch:
> >  a. A 'proper touch' where the mouse is lifted out of the maze by the
> >     handler and, possibly, replaced at the start square. This subject
> >     seems to be covered by the rules already.
> >  b. A 'nudge' where the mouse has 'got stuck' against a wall or post
> >     and requires a limited amount of intervention by it's handler.
> >     This type doesn't seem to be covered by the current rules.
> >
> > My thoughts concerning this second type of touch are:-
> > 1. No-Nudge mice beat One-Nudge mice and One-Nudge mice beat Two-Nudge
> > mice etc. with the run-time determining the placing in each
>'category'.
> > This system is wholly unambiguous and would be simple to administer.
> >
> > 2. This is an entirely defensible position as any mouse that 'gets
> > stuck' will, left to it's own devices, NEVER finish the course. (Do we
> > not use words like autonomous and independent when describing mice?)
> >
> > 3. Derek's point concerning boredom is a little specious for two
> > reasons:-
> >  a. provided that there are a number of high-speed, but perhaps not
> >     very well behaved, entrants they will serve the purpose of
>'audience
> >     interest' (if this is regarded as important) - they just won't
>win.
> >  b. the whole point of the competition is speed and so many, if not
>all,
> >     mouse-builders will be trying to maximise this aspect. I think it
> >     unlikely that a 'boringly slow but well behaved' mouse will ever
> >     achieve first place.
> >
> > 4. Building a mouse that runs quickly in a straightish line is not
>hard
> > and we should be encouraging mouse-builders to think beyond just
>speed.
> > Ryan's non-contact mouse (Red Eye) has shown that a fairly fast mouse
> > _can_ be built and succeed in it's task without intervention - it just
> > didn't do that on the day this year.
> >
> > 5. If a mouse is fast, like Red Eye, and generally behaves well but,
>on
> > the day, requires a nudge then placing it behind slower, better
>behaved
> > entrants this year is surely a satisfactory result. A little work, or
>a
> > bit more luck, may gain it first place next year!
> >
> > 6. This scoring method _could_ be applied only to the Wall Follower
> > classe(s) as:-
> >  a. In general the rest of the world does not run WF competitions and
>so
> >     there is no need for 'compatibility'.
> >  b. It is likely that, having got stuck, any true solver is in such
> >     serious trouble that a handler's nudge is unlikely to rescue it.
> >
> > Any thoughts - Anyone?
> >
> > Regards
> > --
> > Martin J. Barratt