Bill's Blog, 5/11/11 "Wait, Did he say pool"

[rsisk]

Originally Posted by Ryan Caldwell

I think its interesting he said Winnow the pool and not Narrow down
win·now (wn)
1.
a. To separate the chaff from (grain) by means of a current of air.
b. To rid of undesirable parts.
2. To blow (chaff) off or away.
3. To blow away; scatter.
4. To blow on; fan: a breeze winnowing the tall grass.

Aim high with a industrial fan behind the goal?

Guys, you're treating this like a Dave Laverly game hint!

[Alan Anderson]

Originally Posted by rsisk

Guys, you're treating this like a Dave Laverly game hint!

We have to. We not going to get any more GDC hints from Dave Lavery.

[Jon Stratis]

If they go to longer matches, it's really going to be a stretch to have robots competing as actively as they are now on the same batteries... maybe they'll be giving us some new, higher capacity batteries as well?

I can't see them doing anything with ice, as people have mentioned on here... It would simply be too expensive to create/maintain a mini hockey rink in non-hockey venues. Likewise, they'd have a hell of a time convincing some venues (like basketball stadiums) to allow them to have large vats of water on the field for a water game - not to mention difficulties from filling up the pool between matches as water spills out!

I'd say it's much more likely they'll have a game object that we have to throw or shoot (maybe something like a wiffle ball?) through air currents (air currents... water currents... references to water?). How cool would it be to have 100 wiffle balls spilling across the field, with 6 foot high target goals surrounded by rotating fans so the currents are constantly changing?

[Carol]

Or Bill may just be having fun with us, to give us something to do in the long months until kickoff......


Last year the GDC committee started (I think) in June or July, so they are not starting all that much earlier this year. And the fact that everyone was already together in one place probably made it convenient to to then.

[pandamonium]

Winnowing associates with wind currents.

What object logically associates with wind currents?

Kites?

no I am not saying we will have to build a robot that can fly kites but maybe perhaps some kind of air goals...

[MagiChau]

What about an enclosed cage with fans blowing balloons around. The robot would have to move around game pieces to cover fans to redirect game pieces into a scoring box. Don't know how possible this is but it sounds fun.

[Robby Unruh]

I doubt the GDC even knows what kind of hint Bill has "just gave out" because they just started meeting, what, a week ago?

[JesseK]

If we had an incentive to leave autonomous early this year, we might not have seen the 2 or 3 tube autonomous attempts this year.

I like the idea of an autonomous-mode extension for completion of a task, somewhat: keep the extension & bonuses reasonable, then take the equivalent amount of time away from teleop so match cycles are predictable. The idea would extend the life of interesting autonomous routines, and let the teams without auto-modes play teleop sooner. The downside is that it'd grow the disparity between well-resourced (mentors, sponsors, etc) and low-resourced teams.

[Vermeulen]

Hopefully this won't turn into a flying minibot challenge.

[Billfred]

Originally Posted by Jared341

Two ideas I like:

1. An autonomous mode of variable length. For example: if your robot achieves one task in the first 15 seconds, autonomous mode gets extended by 15 more seconds. Or, take a page out of 2001's playbook, and let teams hit an "E-Stop" to manually transition from auto to teleop, giving them incremental bonuses the longer they stay in auto mode (would only be realistic with rules/a field that prevents hostile contact during auto mode). It is important that we don't unilaterally install a 30+ second autonomous period, simply because nobody wants to look at stationary robots for that long.

2. Slightly smaller fields (say, 20x40) and 2 vs. 2. Design so that each regional/division can host two fields that alternate play. Less down time, more matches for each team, easier to see what is going on/scout (3 vs. 3 can get hectic), and more flexibility in fitting the fields into a venue.

Originally Posted by JesseK

If we had an incentive to leave autonomous early this year, we might not have seen the 2 or 3 tube autonomous attempts this year.

I like the idea of an autonomous-mode extension for completion of a task, somewhat: keep the extension & bonuses reasonable, then take the equivalent amount of time away from teleop so match cycles are predictable. The idea would extend the life of interesting autonomous routines, and let the teams without auto-modes play teleop sooner. The downside is that it'd grow the disparity between well-resourced (mentors, sponsors, etc) and low-resourced teams.

I like the idea of fancier autonomous routines, but I think one thing could encourage teams to go for it even in the context of the current setup: remove the fear. We wound up ditching our autonomous routines because we didn't want to interfere with our partners--to say nothing of my fear that our coders somehow caused the robot to take off backwards across the center line (as one of our opponents did at Peachtree). If you saw us start "forward" at Peachtree only to spin around immediately at the start of teleop, it was because I was downright paranoid.

My thought? Park a momentary switch on a grip at each player station. Your driver lets go? Your robot stops. Everyone lets go? Start teleop. Don't reward staying in autonomous, or you'll see teams just holding onto the buttons to rack up the score while their robots stay stationary.

As for smaller fields and 2v2, I'm game--the current-generation field border definitely supports it, and that field size (don't remember the length) was about right for a scaled-down version of Aim High (open field, 28x38 bases with no expansion, bumpers optional). With a little extra field width and/or smaller robots, it could probably work. And it could work too--if you were able to run two matches in six or seven minutes, it could still be a net gain of robot-matches-per-minute while allowing smaller events to slow it down or run one field.

[Jaxom]

Originally Posted by Taylor

I found the opposite to be true. At early regionals/districts, there were too many no-shows, boxes-on-wheels, and incomplete programs/robots to be terribly entertaining; when the really good robots were winnowed out, the games became more exciting.
I do agree with your notion of the tubes becoming less important as the minibots became more prevalent, though.

I don't know...because in week one, there were fewer teams that had minibots. If you had one, it was very important.

Having said that...I've wanted a longer autonomous for years. A variable length auto is a cool idea, if it has the right incentives. And as long as the auto has a piece that has some points (even if a small number) for driving a pretty straight-forward pattern (say, straight ahead) there should be very few teams that can't get *something* working.

[Bill_B]

Originally Posted by Taylor

I found the opposite to be true. At early regionals/districts, there were too many no-shows, boxes-on-wheels, and incomplete programs/robots to be terribly entertaining; when the really good robots were winnowed out, the games became more exciting.
I do agree with your notion of the tubes becoming less important as the minibots became more prevalent, though.

I was thinking that a slight scoring modification would increase the lower row scoring excitement. Make logo scoring up and down diagonals count. Like tictactoe. An additional bonus for filling all nine pegs with tubes in logo sequences. Off season?

[Jim Giacchi]

Originally Posted by Billfred

My thought? Park a momentary switch on a grip at each player station. Your driver lets go? Your robot stops. Everyone lets go? Start teleop. Don't reward staying in autonomous, or you'll see teams just holding onto the buttons to rack up the score while their robots stay stationary.

I absolutely love this idea. I've always thought the E-Stop rule was a little ridiculous, if teams had a method of stopping their robots if disaster stuck, but still be able to do teleop, I think a lot more teams would try autonomous. I never understood why you got penalized for trying to do the safe thing by disabling your robot.