Long Term Future of AWS's

Place discussions about upcoming events here in this thread.

Moderators: BeligerAnt, petec, administrator

Post Reply
User avatar
peterwaller
Posts: 3213
Joined: Fri Feb 15, 2002 12:00 am
Location: Aylesbury Bucks
Contact:

Long Term Future of AWS's

Post by peterwaller »

I have been thinking about how successful the Antweight Class and AWS's are becoming and am getting slightly concerned that we could end up as victims of our own success.
AWS43 had 93 entries and thanks to some very slick organisation, by people not taking part in the fighting, was very efficiently run but I think we could have problems in the future.
1. Many of the old type of scout hut type venues we have used in the past are starting to get a little cramped.
2. Most organisers also want to take part in the event which inevitably slows the process down.
3. The Double elimination system does mean a lot of fights.
For instance 96 robots equates to 8 groups of 12 giving 190 fight which if we just say 2 min per fight including change overs is 6.2 hours.
If we go up to 128 robots 8 groups of 16 we get 254 fight which at just 2 min per fight including change overs is 8.4 hours.
Now I am not suggesting we panic but perhaps we should be looking at ways to cope with this in the future.
Some suggestions that come to mind are:
Reducing the number of entries for each team say from 4 to 3 if we get more than so many entries.
Looking at other ways of running the event other than double elimination that involves less fights.
Possibly having 2 day events but this would put up costs dramatically (both venue and hotel) .
Maybe having qualifying at non AWS events to produce seedings probably not very fair especially on some of the younger entries who can't attend many events.
One thing is certain that when we find such a good venue as Reading with people prepared to give up their time to run the event we should do everything we can to encourage them for the future.
Rapidrory
Posts: 1160
Joined: Sun Nov 24, 2013 9:54 pm

Re: Long Term Future of AWS's

Post by Rapidrory »

Is there any way fights can be held in parallel in separate arenas, at least for the first few rounds?

This would mean splitting the entrants into two groups and running the qualifiers for each group side by side, with the winners from each group going into a semi final or final. It would require a fair bit more organization, and perhaps a more complicated version of antlog, but it would potentially speed things up a fair bit...
Rory Mangles - Team Nuts

Robots: Nuts 2 and many more...

NanoTwo Motor Controllers: https://nutsandbots.co.uk/product/nanotwodualesc
User avatar
BeligerAnt
Posts: 1872
Joined: Wed May 15, 2002 12:00 am
Location: Brighton
Contact:

Re: Long Term Future of AWS's

Post by BeligerAnt »

Some good points Peter, but I don't think (or maybe just hope!) the picture's not quite as black as you paint it.

1. We do need a reasonably large venue, and the cost of venue hire is an issue. "Sponsorship" to cover the venue cost (effectively how Reading and Brighton Uni's are available) is one possibility. Charging an entry fee is problematic, as it's hard to predict how many entrants there will be and many people already have to pay significant travel costs.
2. True, and I don't think we can realistically avoid this. We can't expect people to run an event and not be allowed to take part in it! Having a willing team of helpers to assist with the running certainly helps and as a community I think it's something we do really well.
3. Double elimination does mean twice as many fights as single, but most other systems that guarantee every competitor more than one fight take even more fights and/or are very difficult to run with non-power-of-two entries.

I think your time estimates are somewhat pessimistic. Typically we are getting through an 80-90 robot competition in around 4 hours. Whether this would be significantly affected by the adoption of proposed arena changes is unclear to me.

Reducing the team entry limit might have a significant effect. Athough there are few entries of 4 robots I think there are a number of entries of 3 rollers. Reducing the limit to 2 rollers plus 1 walker/cluster might be a way forward. It would be nice to invoke the "rule" only when the field gets too big but that might be complicated to administer! Having said that, I think there were around 50 "teams" entered into AWS43, so an average of just under 2 robots each. Maybe reducing the limit wouldn't actually have that much effect...

Fair and (relatively) simple to administer systems that work with an arbitrary number of entries are hard to come by.
Mini-leagues feeding into a straight knockout (like football World Cup) tends to require even more fights. They also tend to get complicated with 'odd' numbers of entries.
One possibility which seems a reasonably fair compromise is to run double elimination groups then run the last 16 as a straight knock-out. This gives everyone at least 2 fights but speeds up (and simplifies!) the later stages when few people are actively involved and lots of people are merely spectating. In practice it only removes about 15 fights so less than 10% of the total.

I think qualifying and seeding are really bad ideas. They would discourage newcomers and those that are unable to attend every event.

Two day events are also bad news. It would put the cost of attending out of the reach of many. Although I suppose it would reduce the number of entries, thus obviating the need for a two day event... ;)

Reading does seem to attract a particularly large field. Whether this is due to its location (reasonably accessible from all directions) or to Will cajoling a number of students to enter I don't know. I suspect it may be a bit of both. Having said that, other events have attracted up to 80 entries so not such a huge difference, but a bit more manageable.

Running 2 arenas at once has been tried and doesn't really work. For one thing it requires 6 judges. Also people like to watch other fights and it can be distracting trying to fight whilst another battle is going on. If a judge calls "Cease!" does it apply to your fight or not?
Gary, Team BeligerAnt
Hogi
Posts: 1002
Joined: Sun Dec 30, 2012 6:47 pm
Location: basingstoke

Re: Long Term Future of AWS's

Post by Hogi »

one of the reason's aws 43 ran so well was because Will allowed loads of time for the main AWS. no non spinner warm up, knockout fleas, very quick annihilator, tag team run in spare time at the end. maybe the way forward is just to scrap the non spinner warm up altogether and at very busy events make the FWS a single elimination contest.
Daniel Jackson.

Team Hectic.

Many antweights

Super antweights: territorial.

Fleaweights: fleadom fighter, gaztons.

Featherweights: hectic (under construction)
Rapidrory
Posts: 1160
Joined: Sun Nov 24, 2013 9:54 pm

Re: Long Term Future of AWS's

Post by Rapidrory »

It depends what level of growth we're talking about; It seems that more and more people are doing antweights these days due to lower costs and easier availability of parts etc, and theirs no sign of that trend stopping just yet. How would we get on if say in a few years time we had 200 entrants, or more? It may not be a problem just yet, but a little future proofing might be worth looking at, even if it just remains theory for now...
Rory Mangles - Team Nuts

Robots: Nuts 2 and many more...

NanoTwo Motor Controllers: https://nutsandbots.co.uk/product/nanotwodualesc
Psychostorm
Posts: 79
Joined: Sun Oct 23, 2011 6:27 pm

Re: Long Term Future of AWS's

Post by Psychostorm »

Hogi wrote:one of the reason's aws 43 ran so well was because Will allowed loads of time for the main AWS. no non spinner warm up, knockout fleas, very quick annihilator, tag team run in spare time at the end. maybe the way forward is just to scrap the non spinner warm up altogether and at very busy events make the FWS a single elimination contest.
Agree with scrapping the warm up. Disagree with making the FWS single elimination as it's a formal competition and biases towards antweights.

One point that I didn't see mentioned was the changing of Rounds 1 & 2 to melee style combat. I personally disagree but It would save time, the double elimination format & be consistent with the featherweight category.

PS. Speaking of the featherweights, they use a league first round where the top 10 points scorers get through.

Points are classified 4 to a knockout, 3 to another victory, 2 to losing middle, 1 to losing outright & 0 to nonattendance. (That's right, you can get a point for being immobilized from the start. lol)

In the event the top few aren't predominant, there's a losers melee to round one where the top few qualify. As you can tell by 10, there are more losers melees.
Hogi
Posts: 1002
Joined: Sun Dec 30, 2012 6:47 pm
Location: basingstoke

Re: Long Term Future of AWS's

Post by Hogi »

there is no real call for a non spinner warm up anyway. it's purpose is for drivers to get a little bit of driving practise in before the important main competition but any robot that's going to get to the final stages will in most circumstances have to be spinner proof to a certain extent, and anyway, why should every robot except spinners get practise time? it would probably have to go eventually anyway because of the amount of people who prefer to build and battle with spinners. perhaps two arenas would help so long as we had a seperate pannel of judges for each arena.
Daniel Jackson.

Team Hectic.

Many antweights

Super antweights: territorial.

Fleaweights: fleadom fighter, gaztons.

Featherweights: hectic (under construction)
Post Reply