Alt-BEAM Archive
Message #02840
To: Steven Bolt sbolt@xs4all.nl
From: Bob Shannon bshannon@tiac.net
Date: Fri, 30 Apr 1999 22:50:36 -0700
Subject: [alt-beam] Re: Arena-bots, was Contest!!!!
Steven Bolt wrote:
>
> On Fri, 30 Apr 1999, Bob Shannon wrote:
>
> > As already discussed, new body designs can compete against the
> > standard or modifed kits so that the body will evolve as well as
> > the brain.
>
> Won't that need at least a mail-in type of competition?
Not at all.
If you develop an improvement to the standard chassis, or innovate some
totally new and better chassis, the relative performance between the new
design and the standard chassis can be measured by competing against a
known standard.
People can read of these modifications and new designs, and adapt and
innovate
to their hearts content, and have a way to directly measure the
progress.
> You made quite a point of your `common body' approach:
>
> Date: Tue, 13 Apr 1999 19:28:08 -0700
> From: Bob Shannon
> CC: beam@corp.sgi.com
> Subject: Re: CPU again? (was Re: Beam genome)
> ---8<---
> The common body eliminates a lot of variables while permiting a
> direct evaluation of different sets of behaviors. This would
> greatly accelerate behavioral development and robot evolution.
> ---8<---
>
> Date: Wed, 28 Apr 1999 19:34:45 -0700
> From: Bob Shannon
> CC: beam@corp.sgi.com
> Subject: Re: Contest!!!!
> ---8<---
> Are you sure you dont want to just swap code instead? Its much
> faster, cheaper and easier!
> ---8<---
>
> When only code travels, the affair boils down to a programming
> contest. In which case, you might as well use a simulator. And that
> would indeed be a useless exercise.
Thats totally absurd.
Simulations are far from a replacement for actual testing and real world
performance. Just ask any test pilot (or robotics professional)!
I think we are getting way off topic, and this has already been
discussed
here once.
------------------------------------------------------------------------
eGroup home: http://www.eGroups.com/group/alt-beam
http://www.eGroups.com
- Simplifying group communications
Home