"Dud" robots

[magnets]

Originally Posted by Yalib

What about a robot losing communication? just e-stop him immediately? because usually a robot who lost com will "come back to life" after a number of seconds..

Well, if its a three second communications drop (which we and all of our alliance experienced at CMP a few times last year), then wait. If the cRIO reboots, it'll be more like 20 sec, and if the radio goes down, more like a minute, which is too long. You can't e-stop an unconnected robot.

[ToddF]

Originally Posted by SoftwareBug2.0

The problem is not that your robot could malfunction and lose you the round. The problem is that you get put in the position where you have to tell your alliance partners that you think their robot's autonomous mode is worse than useless. That's a good way to make people resent you.

Pointing out to me that my shooter might malfunction and lose the match for my alliance will not cause me to resent you. However, designing your robot in such a way that it loses me a match might. You better believe that our scouts this year will be noting which teams cost their alliances matches this way.

This aspect of the game is actually one of my favorites this year. It makes designing a shooter which must be pre-loaded a HUGE risk. If your robot fails, you cost your alliance the match. So you have a couple options:
A) Start with the ball on the floor and load it before shooting. That way, if your robot starts out dead, you don't cost your partners the match.
B) Don't shoot at all.

I see this rule as part of a deliberate effort on the part of the rules committee to make powerhouse teams less dominate. This year, the risk vs reward for building a shooter is very high. If you build a shooter which must be preloaded, you risk losing matches for your team. Even if your shooter works perfectly, it is easily defended against by a robot ramming you as you are shooting. This lowers the advantage that elite teams have had in the past, bringing them down towards the effectiveness level of a team which builds a simple pusher bot.

Remember the teams last year who asked, "Why did we spend all the time and effort to build a 3 tier climber, when we can be beaten by two robots who only feeder-station load, shoot, and hang for 10 points?"

Answer: The game designers are deliberately designing the games so three rookie teams who work together can beat an elite team who plays alone.

[MichaelBick]

Originally Posted by Racer26

THIS is a humongous part of why I'm not a fan of Aerial Assist on first analysis.

Seems to me the elite tier robots should easily be able to turn 150ish points on their own (20 pts [truss toss +high goal] x 6 cycles + 25 auto). Especially so if they have two defenders clearing the road for them on this wide open field. Total number of cycles is unlikely to change much with assisting, if anything going down slightly.

Adding one assist per cycle, brings the cycle points to 30. If doing so 4 times costs you 2 solo-cycle time periods (going from 6 cycles to 4 in a match), you have no net gain.

Assists are one of the most important parts of the game, for one simple reason. The first seeding tiebreaker is based on assist points.

[Daniel_LaFleur]

Originally Posted by Racer26

THIS is a humongous part of why I'm not a fan of Aerial Assist on first analysis.

Seems to me the elite tier robots should easily be able to turn 150ish points on their own (20 pts [truss toss +high goal] x 6 cycles + 25 auto). Especially so if they have two defenders clearing the road for them on this wide open field. Total number of cycles is unlikely to change much with assisting, if anything going down slightly.

Adding one assist per cycle, brings the cycle points to 30. If doing so 4 times costs you 2 solo-cycle time periods (going from 6 cycles to 4 in a match), you have no net gain.

Do you really believe that a robot will consistently and repeatably be able to retrieve the inbound pass, drive to the truss, toss the ball over the truss, retrieve the (now bouncing) ball, drive to a scoring position, shoot and score in the high goal, and get back to retrieve another inbounds pass in less than 25 seconds?

I think you are, very much, overestimating most teams capabilities. In the end, there may be 1 or 2 teams in all of FIRST that might be able to pull this off on the competition field, but I doubt it.

[Daniel_LaFleur]

Originally Posted by SoftwareBug2.0

The problem is not that your robot could malfunction and lose you the round. The problem is that you get put in the position where you have to tell your alliance partners that you think their robot's autonomous mode is worse than useless. That's a good way to make people resent you.

Any team that lets me know that their autonomous mode may be 'worse than useless' is marked high in my book because they are honest (we can help them with the autonomous mode).

It's the ones that tell me that they can do it all, when they are barely functional that bother me.


Also:
Is anyone concerned that if a bumper falls off your robot will be disabled? And this could happen with a ball inside your machine?

[dubiousSwain]

Originally Posted by Daniel_LaFleur

Any team that lets me know that their autonomous mode may be 'worse than useless' is marked high in my book because they are honest (we can help them with the autonomous mode).

It's the ones that tell me that they can do it all, when they are barely functional that bother me.

I totally agree with this

[dubiousSwain]

Originally Posted by Daniel_LaFleur

Do you really believe that a robot will consistently and repeatably be able to retrieve the inbound pass, drive to the truss, toss the ball over the truss, retrieve the (now bouncing) ball, drive to a scoring position, shoot and score in the high goal, and get back to retrieve another inbounds pass in less than 25 seconds?

I think you are, very much, overestimating most teams capabilities. In the end, there may be 1 or 2 teams in all of FIRST that might be able to pull this off on the competition field, but I doubt it.

If i told you last year that someone could 3rd level climb in 6.5 seconds, would you believe me?

[Racer26]

Originally Posted by Daniel_LaFleur

Do you really believe that a robot will consistently and repeatably be able to retrieve the inbound pass, drive to the truss, toss the ball over the truss, retrieve the (now bouncing) ball, drive to a scoring position, shoot and score in the high goal, and get back to retrieve another inbounds pass in less than 25 seconds?

I think you are, very much, overestimating most teams capabilities. In the end, there may be 1 or 2 teams in all of FIRST that might be able to pull this off on the competition field, but I doubt it.

Looking at several videos of 1114 in 2008, they were able to hurdle 6-7 times in teleop fairly consistently, on a crowded overdrive field. I see no reason an elite level robot could not do a similar number of cycles in 2014.

[ninjosh97]

Originally Posted by Racer26

Looking at several videos of 1114 in 2008, they were able to hurdle 6-7 times in teleop fairly consistently, on a crowded overdrive field. I see no reason an elite level robot could not do a similar number of cycles in 2014.

Also, check out this video. http://www.youtube.com/watch?v=JElxKBOT958
This team was able to hurdle AND catch the ball, without it ever touching the floor.

I realize something like this wouldn't get you the catch points, but would still get you the truss points.

I could see some teams doing something like this.

[pfreivald]

Originally Posted by Racer26

Looking at several videos of 1114 in 2008, they were able to hurdle 6-7 times in teleop fairly consistently, on a crowded overdrive field. I see no reason an elite level robot could not do a similar number of cycles in 2014.

Defense was pretty much not allowed in Overdrive, and after you hurdled the ball it was right in front of you again--you didn't have to have it inbounded at the other end of the field.

I'm thinking the super-elite teams will probably be able to do 4-5 cycles unassisted with no defense on them. We'll see if that estimate changes as the build season progresses.

[yash101]

If che control system fails, the team with the failure would lose for no reason, especially because in most cases the failures in the cRIO are natural and happens once-in-a-while.

So you may ask why I am bringing this up. This happened to us last year. As soon as the teleop period started, the cRIO randomly died. Thankfully, we had an extra cRIO sitting in out pit so recovery didn't take too long.

However, what if this happened and the robot was loaded upon the damage? That would be shear bad luck for the entire alliance!

I actually don't remember if we were even allowed to rematch after we fixed the problem so they really have to tweak that rule.

Overall, in my opinion, I think this is the most poorly written game in the FIRST history! There isn't even an endgame to give teams more of a challenge. Also, the ball is roughly the size of the robot. It also needs to be picked up. Because of the size and what needs to be done with the ball, it isn't as simple as 2008 where you could knock off the ball and then kick it. I think one of the only possibilities is to kick the ball!

[Racer26]

Originally Posted by pfreivald

Defense was pretty much not allowed in Overdrive, and after you hurdled the ball it was right in front of you again--you didn't have to have it inbounded at the other end of the field.

I'm thinking the super-elite teams will probably be able to do 4-5 cycles unassisted with no defense on them. We'll see if that estimate changes as the build season progresses.

After you truss toss it, it will be right in front of you, just like overdrive, and like overdrive, you have to return to the other end of the field to score again. There's an extra ball acquisition per cycle, but it doesn't seem unreasonable that 6 will be possible by the elite. Remember, 2008 was 6 years ago. All of those elite teams have learned a lot in the intervening time about how to build better robots too.

[Racer26]

Originally Posted by yash101

Overall, in my opinion, I think this is the most poorly written game in the FIRST history! There isn't even an endgame to give teams more of a challenge. Also, the ball is roughly the size of the robot. It also needs to be picked up. Because of the size and what needs to be done with the ball, it isn't as simple as 2008 where you could knock off the ball and then kick it. I think one of the only possibilities is to kick the ball!

While I might be inclined to agree that Aerial Assist is one of the weakest games the GDC has come up with in my 11 years in FRC, there is definitely no problem picking up this ball. Go watch some videos from 2008, when trackballs were twice the diameter of this ball, and robots only marginally larger. Picking up a 2014 game ball is not that big a deal.