Joined: 18 Jun 2004
Posts: 3842
Location: Queensland near Brisbane
Swiss System
Did people like the
Swiss System
employed during the antweights?
In my opinion, it worked well. It gave us many close matches. No robot got an easy passage to the finals which a random draw can give on occasion.
Is another system better? _________________ So even the rain that falls isn't actually going to fill our dams and our river systems
Tue Oct 02, 2012 7:27 am
Valen Experienced Roboteer
Joined: 07 Jul 2004
Posts: 4436
Location: Sydney
hmmm interesting,
issues i can see
putting winning bots against winning bots could lead to taking out the big name bots relatively early in the comp.
still need to handle people being in and out of individual rounds.
The big difference with robot competitions to regular ones is people don't often die in chess lol.
HAH if your using a seeded comp, perhaps do the reverse, pit winners Vs losers like some bad 80's martial arts movie where they fight through hoards of minions before the final battle ;->
What do people think of seeded comps vs random anyway? _________________ Mechanical engineers build weapons, civil engineers build targets
Tue Oct 02, 2012 4:08 pm
seanet1310
Joined: 08 Nov 2006
Posts: 1265
Location: Adelaide
I quite liked it.
For show reasons I think we need to make it clearer how the system worked (many of the builders where not 100% sure at the time) and make sure we talk up the decider fights that we get the winners from (similar to finals) as everyone loves a good final/decider.
Even if unseeded after the first round a winner vs looser can be done easy enough.
No system is perfect. The Swiss for 19 robots gave us a winner while allowing all working robots to have maximum number of fights. The standard way (semi round robin + finals) we would have got 3 rounds + finals max Double elimination has many problems with 20 robots.
This got from memory 5 rounds with only 1 robot pulling out and that was at round 3 or 4.
Jake raises some good points.
Most systems need to manual handle in and out for individual rounds and during this event that was done where possible putting the out robot against the bye (as we had 19 not 20) but in the end someone may get points from a single round forfeit.
The 'big name bots' taking eachother out becomes more of an issue as the event goes on. The first round is random fights. Second is winners vs looses but with in this instance 10 winners and 10 looses it is almost a random opponent as we had no prevision for KO to only vs other robots who managed to do a KO.
Double elimination also quickly forces the winners to fight other winners while reducing the number of fights each round. This is a little more extreme in the focussing but based on the one event it was run this seams to be an acceptable risk with a large number of bots and limited time frame. _________________ Remember to trust me, I am an Engineer.
HAH if your using a seeded comp, perhaps do the reverse, pit winners Vs losers like some bad 80's martial arts movie where they fight through hoards of minions before the final battle ;->
sounds like a good concept _________________ <Patrician|Away> what does your robot do, sam
<bovril> it collects data about the surrounding environment, then discards it and drives into walls
Tue Oct 02, 2012 8:15 pm
Daniel Experienced Roboteer
Joined: 30 Aug 2005
Posts: 2729
Location: Gold Coast
As Sean said, we didn't have time to fit in the finals before we started tearing the arena down. We only would have got in 3 rounds before an 8 robots final, which wouldn't have been enough to to get a true top 8 robots and we would have had play offs for the bottom spots.
I think I could have put the points table on the big screen better so people could see who was winning.
The biggest problem was when I had to manually type in the next round and I had Philip leaning over my sholder telling me who he wanted to fight next. I'll fix up the automated bit next time.
Yeh swiss system was a good compromise for the day. I personally would prefer to have round robin and finals but due to the number of robots this just purely wasn't possible.
I am also not sure how well it would work with Feathers but for ants the taking out the top competitors isnt really an issue IMO.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum