State Power Alliances 2015

[zinthorne]

Originally Posted by Dunngeon

I agree with your first pick, although 4488 performs much better on the right feeder for obvious reasons.

I don't agree with your landfill pick for 2 reasons, the first is that 4911 doesn't produce 2 capped stacks consistently, and second is that 2471 is at least 200ms faster into the cans than 4911 was. 4911 wasn't even close to Mean Machine in terms of speed. Also, 4911 was a single can grabber, whereas 2471 was a 2 can grabber.

948 will provide the auto, but little else the alliance needs. This third bot would preferably be able to play with tipped over cans, and create 2 of their own capped stacks. 948 doesn't do tipped over bins and rarely (ever?) did 2 full stacks. 1983 would be my pick here. They consistently do 2 stacks, capped and noodled, provide decently quick can grabbing ability, and are extremely reliable.


Left: 1983
Mid: 2471
Right: 4488

Just a note 4911 had a fairly fast 2 rc grabber at champs that nearly got their alliance into finals. They were a 3rd pick due to some unlucky qualification match pairings. They only got to play one elimination match, but beat the other robot to the cans and got both.

My ideal alliance would be 4488, 3663, and 2471. I picked this alliance to maximize scoring, and consistency. 3663 and 4488 tied in PNW for the farthest playing team in worlds this year as division finalists. Both are very consistent, and both have can grabbers. 2471 has a 2 can grabber that is also fast. 4488 would put up three stacks of 6 capped and noodled. (They could start by grabbing a rc in the middle or by the right feeder station depending on the alliance being played...) 3663 would take the left station and one rc from in front of it, creating two stacks one capped and noodled and one not. 2471 would place on more can for 4488 to use and then make one stack of their own, cap it, then cap 3663 second stack. The remaining rc (if in possession) would be put on the highest stack remaining. Once 3663 is done with the second stack they can go and do a 4-6 high stack from the landfill.

[The other Gabe]

Originally Posted by zinthorne

Just a note 4911 had a fairly fast 2 rc grabber at champs that nearly got their alliance into finals. They were a 3rd pick due to some unlucky qualification match pairings. They only got to play one elimination match, but beat the other robot to the cans and got both.

My ideal alliance would be 4488, 3663, and 2471. I picked this alliance to maximize scoring, and consistency. 3663 and 4488 tied in PNW for the farthest playing team in worlds this year as division finalists. Both are very consistent, and both have can grabbers. 2471 has a 2 can grabber that is also fast. 4488 would put up three stacks of 6 capped and noodled. (They could start by grabbing a rc in the middle or by the right feeder station depending on the alliance being played...) 3663 would take the left station and one rc from in front of it, creating two stacks one capped and noodled and one not. 2471 would place on more can for 4488 to use and then make one stack of their own, cap it, then cap 3663 second stack. The remaining rc (if in possession) would be put on the highest stack remaining. Once 3663 is done with the second stack they can go and do a 4-6 high stack from the landfill.

a few problems with this, at least for me. the first is that your entire alliance has issues with downed containers (which was the main reason I chose 4911 over 2471; I've seen them handle the cans better overall than Mean Machine). While they had a great robot, the main reason CPR advanced so far into playoffs was their brilliant scouting, allowing them to create a super solid alliance in one of the strongest divisions; keep in mind that 492 technically placed higher, qualifying for Einstein (even if they never played a match on there), and that was because they were on the best team in their division. Also CPR would have neither the time nor the ability to go from the Landfill (I have gone to every event they have, and they have never done landfill). for your alliance, I feel like the Skunks would make more sense (assuming you want all 4 containers, if not, 955)

[quote=Dunngeon] "I agree with your first pick, although 4488 performs much better on the right feeder for obvious reasons."
I was visualizing the feeder stations from the step, not behind the alliance wall, sorry for the confusion
[quote=Dunngeon] "948 will provide the auto, but little else the alliance needs. This third bot would preferably be able to play with tipped over cans, and create 2 of their own capped stacks. 948 doesn't do tipped over bins and rarely (ever?) did 2 full stacks."

I was relying on 4911 to right containers in the way my team did for CPR and 255 did for 4488, eliminating that issue. also, having been the lead scout for my team, I watched every match on Curie, and seem to remember them doing 2 capped 5 stacks in tele during elims, and 1+ during Quals (since they did co-op instead)... I have some evidence here (although my team unfortunately stole all their containers, so only one stack is capped) http://www.thebluealliance.com/match/2015cur_sf1m1 these appear to be 6 stacks, too, which makes my argument for them even better Skunks are a good pick too, though

(sorry, havent figured out multi-quoting)

[logank013]

Indiana would probably be 234 or 135 and 1024 at HP. Then landfill would probably be 1720 or 71. Maybe even 1741 at landfill too. My thoughts would be 234-1024-1720

[logank013]

On a whole league stand point, the best alliance would be 254-148-1114 to me.

[FIMAlumni]

FIM was strong as ever this year, but almost all of the top teams were HP fed (1023, 548, 1918, 67, 469, 3641, 2137, 68, 217, 314, 2054, and 1025). The only member of the "elite" FIM teams taking from the landfill was 33, so they are the first selection.
33: 2-3 stacks from the landfill, ability to grab RCs from the step missed by others in auton, 20 point auto.

Now that I have my 20 point auto and a landfill miner, I need Canburglars and HP feeding for the next alliance member. This leaves 548, 1918, 67, 2137, and 217.

548: First choice due to reliability, by worlds they were grabbing two step cans and making 4 stacks of 5 from the HP station.

I am tempted to pick one of the other 4 for the final position, but assuming 548 plays up to potential, that leaves only 10 totes for the other robot from the HP. With so many FIM robots able to do this, Canburglar speed is far more important. This leaves me with 503 and 27.

503: The super fast Canburglar puts them on this alliance, but also the ability to stack the last 10 totes from the HP.

So 33, 548, and 503 would be my the ultimate Michigan Power Alliance.

[Kevin Leonard]

Originally Posted by logank013

On a whole league stand point, the best alliance would be 254-148-1114 to me.

Now thats another interesting idea- best alliance in the world.
The problem with 254-148-1114 is that a) there are faster canburglars than both 254 and 1114's, and they run out of totes far too quickly! (Good problem to have).

The best alliance in the world should probably have the fastest canburglars, so probably some combination of Group A and Group B:
A: [67, 548, 971, 973, 1678, 1918, 3310] (Good Human Player Robots with some of the fastest can grabbers in the world)
B: [33, 118, 254, 1114] (Landfill Robots capable of 3+ stacks/match and a 20 point auto)

There's also something to be said for a robot like 225 that can use step totes during teleop (for an alliance like 1114-148 that could use all of the other totes), or for cappers like 27 that can augment another robot's abilities on the right alliance.

With IRI rules, speed no longer matters, so your best alliance changes as well. I'd argue that:
254-148-225 would be the best possible alliance under these circumstances, as 148 and 254 together can probably clear out the entire landfill and feeder station together, and 225 can grab step totes and help with the landfill if necessary.

Did any really good robot this year use upside-down totes?

EDIT: I stand corrected on the subject of can grabbing. 254's were ridiculously fast (I'm told they were the same ones 973 and 971 were using).
In which case I think the best alliance in the world would be 254-1114-225, considering between 1114 and 254, they can clear out the landfill and human player station, as well as have tons of can grabbing ability and a 3-tote auto. They also get step totes from 225.

[CJ_Elliott]

IMO, I would say for MN it would be 3130 at landfill, 2512 at human player position, and 2502 at the other human player position. 3130 could get 2 bins off of the step, and with 2512's stacking autonomous that could probably beat any other alliance in Minnesota.

[kyle_hamblett]

Originally Posted by logank013

On a whole league stand point, the best alliance would be 254-148-1114 to me.

They would probably run out of totes to stack...

[jajabinx124]

Originally Posted by CJ_Elliott

IMO, I would say for MN it would be 3130 at landfill, 2512 at human player position, and 2502 at the other human player position. 3130 could get 2 bins off of the step, and with 2512's stacking autonomous that could probably beat any other alliance in Minnesota.

I was thinking of that as well. 2512 would take out their can burglars if push came to shove for cans from the STEP. 3130 has can burglars but they have to drive back to get them(Reliable but sometimes not fast enough, part of the reason they didn't use it at champs in eliminations was because it wasn't fast enough. That's why they picked 3310 with their quick can burglars instead of running their own at champs).

This alliance would be a factory to watch honestly.. the top 3 robots in MN in one alliance.. haha

[Ben Martin]

For MAR PA, 1640 + us to get the step cans, or one robot gets one set of cans and the other does a 20 point auto. The 3rd bot could be either 2607 or 1218 with a ramp to finish out the feeder station.

For NJ, besides those previously mentioned, I think 1403 is in the conversation--they were doing 3 stacks of 5 at district events, and they were pushing sixes at MAR champs. Also, 316 was consistently doing 2 stacks of 5 + noddle and 1 uncapped stack of 6 by the end of MAR champs.

[ice.berg]

Originally Posted by jajabinx124

I was thinking of that as well. 2512 would take out their can burglars if push came to shove for cans from the STEP. 3130 has can burglars but they have to drive back to get them(Reliable but sometimes not fast enough, part of the reason they didn't use it at champs in eliminations was because it wasn't fast enough. That's why they picked 3310 with their quick can burglars instead of running their own at champs).

This alliance would be a factory to watch honestly.. the top 3 robots in MN in one alliance.. haha

OR we could give our can burglars to someone else....

[Dunngeon]

Originally Posted by Kevin Leonard

Now thats another interesting idea- best alliance in the world.
The problem with 254-148-1114 is that a) there are faster canburglars than both 254 and 1114's, and they run out of totes far too quickly! (Good problem to have).

The best alliance in the world should probably have the fastest canburglars, so probably some combination of Group A and Group B:
A: [67, 548, 971, 973, 1678, 1918, 3310] (Good Human Player Robots with some of the fastest can grabbers in the world)
B: [33, 118, 254, 1114] (Landfill Robots capable of 3+ stacks/match and a 20 point auto.

There is only one (technically 2) robot in the world faster than 254's can grabbers and that is 1678. I'm not counting the harpoons for this, since they are a totally different strategy.

Originally Posted by themccannman

Since I don't want to clutter the forums I figured I'd post these two threads under the same title. First off, I spent all of thursday, and friday recording can grabbers on newton, and all of saturday recording grabbers on other fields during elims. I recorded at 120 fps with a samsung galaxy S6, then went back and counted frames for each grabber to try to get accurate speeds. Because of the fact that 120fps recording gives me a time resolution of 8.3ms I rounded times to the closest 10 or 5 ms increments so times aren't perfectly accurate. The other thing to note is that I didn't have a way to know exact enable times for matches that we (1678) didn't play in so I don't know exactly how long grabbers took to startup. Our triggers took about 2 frames (~15ms) to fire when tested off the field so that was the only robot I had an accurate initialization time for. All times for recorded grabbers are posted.

These times are all hook-to-the-can. I'm not putting time-to-unstealable because it's hard to tell how far the can needs to tip for the other team to not
In no particular order:

1114 harpoons - 105 milliseconds (word of mouth)
1114 2 can - 350 ms
971/254/973 cheesecake - 175 ms
148/1923 cheesecake - 175 ms
1678 (4 wraps, on the field) - 175 ms
1678 (5 wraps, pits) - 158 ms
987 - 290 ms
3310 - 205 ms
5027 - 266 ms
2056 - 250 ms
3940 - 300 ms
1918 - 225 ms
1711 - 370 ms
330 - 400 ms
2046 - 350 ms
3339 - 470 ms
368 - 370 ms
1983 - 480 ms
166 - 400 ms

I accidentally deleted some videos but I think I recall a couple of times for some teams I recorded, give me a heads up if I made any mistakes:

67 - 205 ms
118 - 205 ms
548 - 205 ms
16 - 350 ms?

[IronicDeadBird]

Originally Posted by Dunngeon

There is only one (technically 2) robot in the world faster than 254's can grabbers and that is 1678. I'm not counting the harpoons for this, since they are a totally different strategy.

Anybody have this list with number of cans they grab cause unless you swipe 3 the score can be evened up.

[Michael Corsetto]

Originally Posted by Dunngeon

There is only one (technically 2) robot in the world faster than 254's can grabbers and that is 1678. I'm not counting the harpoons for this, since they are a totally different strategy.

I'm not sure ours were faster than 254's grabbers, we never went head to head. Even if the times you quoted are 100% accurate (which they're not), winning isn't about getting to the cans, its about getting the cans away from the other guys.

Maybe we'll do a can drag (literally!) race at Chezy Champs or Capital City Classic?

-Mike

[CJ_Elliott]

Originally Posted by ice.berg

OR we could give our can burglars to someone else....

we couldn't take them, we have a ramp... #teamtether