Subject:
|
Re: r10 Project X -- Rules update
|
Newsgroups:
|
lugnet.org.ca.rtltoronto
|
Date:
|
Wed, 30 Jan 2002 14:46:42 GMT
|
Viewed:
|
641 times
|
| |
|
|
K, here’s the thing—if we do head-to-head competitions, there are only 2
‘bots involved at a time. There will be one winner and one loser. Just
like the block sorting ‘bots--every ‘bot will eventually play every other
‘bot in subsequent heats.
The reason I don’t appreciate stop watches is that someone has to use them.
One person to each ‘bot and the definition of ‘done’ per person is grossly
inaccurate. Since, at the rate we’re going, all these ‘bots are going to be
finishing close to the same time (under 40 sec for anti-‘X’), there may be
some disputes if someone thinks their ‘bot finished before the timer is
stopped. I know that in my program right now, the ‘bot moves off the board
when it places the last block, so you can see the entire surface and judge
and/or set up for the next ‘bout. Will timer guy click stop when the last
block is placed, when the arm lifts off it, or when the ‘bot parks off the
board, or anytime inbetween? The less reliance on ‘exacting mearsurements’
the easier I think the competition will be. I do have the ‘bot beep when it
thinks it’s done.
Per head to head competition:
50 points for finishing first
-10 points for incorrect blocks. We know they’re incorrect because they’re
not in the square
+5 points for good block (whether on side or just in square, not meshing)
+10 points for every perfect (meshing studs) block
Hard to dispute any of these criteria
Here’s how things may unfold
‘Bot 1 finishes first +50
has 2 incorrect –20
has 4 good +20
has 2 perfect +20
for 70 points
‘Bot 2, with a full set of perfect has +80 (‘bot 2 deserves to win in this
instance, for ‘bot 1 could have done the thing in 10 seconds but did a poor
job doing it. ‘Bot 2, taking a whole minute, but achieved a *perfect * run,
is, imho, a better ‘bot.)
Here’s the thing, if ‘bot 2 has 1 incorrect block, tie score
If ‘bot 2 has 1 incorrect and one good, score down to 65, ‘bot 1—the faster
‘bot, wins.
If your ‘bot is better (and it should be), the bout might look like this:
‘Bot 1 (Runamok Design Team)
+50 finishing first
+50 (5 perfect)
+10 (2 good)
-10 (1 dropped because some kid, pointing at Chris’s ‘bot and saying, ‘Hey
Mister…’ accidentally knocked my ‘bot, thus causing the block to drop)
= 100 points
‘Bot 2
+80 (8 perfect blocks)
= 80 points
I favour speed insofar as the ‘bot still attempts to make an ‘x’ pattern.
You get a ‘bot that says it’s done at the beginning of the match, gets +50,
but all those misplaced blocks will make sure it loses. If the random
pattern has 6 or more correct blocks at the start (hopefully not!) is the
only way a non-moving ‘bot can win for sure. (+50 for finishing first, +60
for 6 perfect, -20 for incorrect = total of 90)
Over the course of the day, with subsequent matches, the faster ‘bots will
tend to win. Ben can add up all the wins at the end of the day and come up
with an overall winner
If we do the time thing, that’s okay too.
In lugnet.org.ca.rtltoronto, Chris Magno writes:
> > I guess no one is going to cheat, but if my bot just say it is done at 0 sec
> > then my score would be 24 sec at worst. So the penalty time for wrong block
> > need to increase - I suggest 6 sec.
> >
> > Also how are we going to count the overall score? Do the bot got one point
> > when it win a 1 on 1 match, or do we add up the time score for every match?
>
>
>
>
> AHHHHHHhh its just so overwhelming......
>
> :)
>
> too many questions, so little time.
>
> I'm not going to say another thing about rules and scoring until i hear
> from more people......
>
> what say you all???
>
> I ran some more numbers and a higher penalty for a wrong block makes
> sense, 6 sec, give a time of 48 seconds if we were to assume the anti-X
> and 10 seconds puts us at 80 seconds.
>
> So here is a leading question: what is a reasonable amount of time to
> do a PERFECT anti-x??? then we can add a few seconds and divide by 8
> and call that number the wrong block time.
>
>
> --
> Chris
> Pretend this says something funny
|
|
Message has 3 Replies:
 | | Re: r10 Project X -- Rules update
|
| (...) So far, me likey!!! :) (...) ix-nay on enbay!!! I am going to Virginia that weekend.... the thing that sucks...is I am off Thursday and Friday, but have to leave Thursday afternoon for the flight. Won't be home until Sunday afternoon. Trip (...) (23 years ago, 30-Jan-02, to lugnet.org.ca.rtltoronto)
| | | Re: r10 Project X -- Rules final
|
| Well, After a lot of soul searching, I have decided that A decision has to be made as to scoring so that we can all build our robots and tune as required.... and the winner is..... This game is both about speed AND accuracy. By the very nature of (...) (23 years ago, 2-Feb-02, to lugnet.org.ca.rtltoronto)
|
Message is in Reply To:
| | Re: r10 Project X -- Rules update
|
| (...) AHHHHHHhh its just so overwhelming...... :) too many questions, so little time. I'm not going to say another thing about rules and scoring until i hear from more people...... what say you all??? I ran some more numbers and a higher penalty for (...) (23 years ago, 30-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
|
|
|
|
