Subject:
|
Re: r10 Project X -- Rules update
|
Newsgroups:
|
lugnet.org.ca.rtltoronto
|
Date:
|
Tue, 29 Jan 2002 15:17:55 GMT
|
Viewed:
|
577 times
|
| |
|
|
OK folks, since I don't have a robot competing, perhaps I'll throw out
some suggestions:
Some scoring rules I've heard bandied around are:
First to finish: +20 or +50...
Block wrong: -10 or disqualified
Block mispositioned: -5
I'd observe that in one suggested system (+50 for first, -5 per
misposition, disqual for incorrect) that mispositioning doesn't matter,
since there are only 8 blocks you can't add up enough penalties to
overcome the finishing first bonus. So that system reduces to:
Whichever robot finished first without an incorrect block.
I also note that if you're 1 second slower, or 1 minute slower than the
winner, it makes no difference, since time is not accounted for except
to determine who was first.
You might want to consider resorting to a system like the biathalon
uses: A time, plus penalties for things done wrong. So your system
might be:
SCORE = #seconds_to_finish + #sloppy_blocks * penalty +
#incorrect_blocks * penalty2
Lowest score wins.
Then think about robot times. I can't see a single picker finishing
anything but the simplest pattern in less than 15 seconds; I can easily
imagine someone having a functional robot that finishes in 50 seconds or
a minute. But in the hot-competition range (say, around 20 seconds),
the margin will be only a few seconds. So if you make the sloppy-block
penalty 1 second or 2 seconds, and the misplaced block penalty 10
seconds or so, you probably have a workable system.
Of course, this means someone has to watch each robot with a stopwatch,
but many people have digital watches these days, it shouldn't be an
overwhelming problem.
Jeff E
|
|
Message has 2 Replies:
 | | Re: r10 Project X -- Rules update
|
| (...) At the risk of being wishy-washy..... I like this. It makes sense, and it address rob's concern. (right?) example: Time = time in seconds + 1 sec for not perfect block + 3 sec for "wrong" blocks. (these penalty times are based on what i figure (...) (23 years ago, 29-Jan-02, to lugnet.org.ca.rtltoronto)
| | | Re: r10 Project X -- Rules update
|
| "Jeff Elliott" <jeffe@telepres.com> wrote in message news:3C56BD23.6EF3CC...res.com... (...) This is the best idea I've seen yet. I wanted to suggest scoring via time, but because I hate math :) I couldn't bother trying to come up with something. I (...) (23 years ago, 29-Jan-02, to lugnet.org.ca.rtltoronto)
|
Message is in Reply To:
| | Re: r10 Project X -- Rules update
|
| (...) example (...) I was thinking about this. And I was having a hard time finding a number that would balance out a robot that only moves 1 or 2 blocks then claims to be done. Sure, We never really have people who show up to be goof's on purpose, (...) (23 years ago, 28-Jan-02, to lugnet.org.ca.rtltoronto)
|
30 Messages in This Thread:
 
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
