Bill's Blog 6/15/11: "What percent of viewers read the title?"

[Jim Wilks]

Originally Posted by sgreco

The problem with the minibot was not the concept of a miniature robot performing a task, but rather the junk that they required us to use, and the strict limitations on extra parts we could use.

That, in a nutshell is exactly the crux of the minibot fiasco this year.

[Francis-134]

I know this is far from a realistic survey, but did anyone lose matches this year because of accumulated small delays in the match schedule? I know there were some events that has systemic field problems (Finger Lakes had some delays if I recall), but I did not hear of any events that had a number of robot/field issues that led to a loss of matches.

Basically, I'm trying to understand if adding a time limit to the time it takes for a robot to connect is going to solve any real problems.

Also, the question asked seems oddly worded, i.e. reduce the number of matches you have by building time into the schedule or introducing a connect timer. It's like asking someone if they would like to eat pig slop or chum for dinner; neither option is very good, so why can't we just change the choices?

[Craig Roys]

Originally Posted by Chris is me

I honestly believe that the statistics for the minibot and field connection time are backward and someone just made a mistake. I can't figure out how so many people are fine with missing a qual match because of NI problems.

I don't think anyone is suggesting having a team miss a match due to NI problems - that's a different story. It seems to me, though (and maybe I'm mistaken), that many of the connection issues are not so much due to NI as they are to not having things properly set up - e.g. not having the cRio plugged in after tethering (something we've been guilty of a couple times...oops). Teams need to be prepared when they walk out on the field rather than make everyone else wait while they try to find the problem which is often something that should have been taken care of in the pits. Now, that being said, the issue becomes how do you determine if it's a problem due to NI or due to lack of preparedness by the team as both are going to result in that flashing blue/red light at your driver station?

[Aren Siekmeier]

I also was rather surprised and a bit disappointed at the 60.8% in favor of a hard cutoff. As several people have already said, there is already a system in place that gives the FTA discretion to leave the team out if its getting out of hand; having a hard stop gets rid of this discretionary, decision-making element, and seems to me to be, quite frankly, Draconian. It's already true that if the schedule is going to run late because of connections, or if its known that connection isn't possible (because something isn't plugged in), the FTA can agree with the team to drop them from the match. And the 2-3 minutes of intro time is plenty to do this in.

Teams have put in a long (or short?) 6 weeks on these robots, and it would be a shame to cut them out of a match (or 3) because they were 10 seconds away from connecting. If connection to FMS really is a problem, it needs to be tackled from the pits, with more aggressive efforts to inspect early on and get ALL the teams out there for at least one practice match, and getting help to these teams when the issues come up at an event. And don't get me wrong, they do a great job with that already, but the solution is not penalizing the teams with problems, it's helping them.

[rachelholladay]

Honestly, I think a lot of people of my team never want to hear the word minibot in their life ever again. Every team I saw kept going back and rebuilding them between regionals or before CMP. Since the rules allowed you to hold back your minibot from ship, people kept working. I love build season just as much as the next fella, but this continuation seemed to disrupt the purpose of the 6 weeks.

After working as a volunteer at a regional (Bayou) three times and CMP twice, I would not favor a set time to connect. It isn't realistic. Every time we had to tell a team that they couldn't compete in a match because they couldn't connect, they got very angry. Although I definitely sympathized with them, some people let their behavior get our of control and lost sight of gracious professionalism. In my mind, a few minutes is worth letting people see their hard-work compete.

I agree that it would have been nice if they would have listed the reasons that people considered a second event to be a drag. I know last year, when we attended 2 regionals and CMP, a few of our mentors found it to be draining only because the amount of personal vacation time they used. We don't like to compete in only one regional. (Six weeks of work for only three days of competition, its over too soon!) The ticket to CMP is never guaranteed. (except for Hall of Fame and the Sustaining) Some members of our team shy away for going to two regionals because if we make it to CMP, it makes three events. However by only going to one regional there is the possibility that its your only event, if you don't make it CMP. Its a difficult decision we make every year, and its often decided by when our spring break falls in conjunction with the regionals. Our principal is more likely to let us go to two regionals if one is over spring break and therefore doesn't require missed school. Fortunately, he doesnt have a problem with us missing school for CMP because he realizes that its kinda sorta really important.

[PAR_WIG1350]

Originally Posted by Billfred

I'm okay with there being a time limit on it, provided that time limit is reasonable for the system. (The question phrased it as "fixed", not "short", and posed it against the alternative of having fewer matches in a period of time.)

False dichotomy, I believe, is the term we are looking for here.

[MagiChau]

Originally Posted by rachelholladay

Honestly, I think a lot of people of my team never want to hear the word minibot in their life ever again. Every team I saw kept going back and rebuilding them between regionals or before CMP. Since the rules allowed you to hold back your minibot from ship, people kept working. I love build season just as much as the next fella, but this continuation seemed to disrupt the purpose of the 6 weeks.

The evolution between competition of teams' robots has not started with the minibots. I would say its unfair to point out the fact minibots went under revisions after the 6 weeks when teams have been doing that to their main robot for years.

I am probably just echoing opinions but if the next time minibots are used if they are integrated into the game properly it should turn out okay. With the GDC pushing forward with game design and testing on a quicker schedule I hope this will be enough if minibots are brought back next year.

[Chris Hibner]

Originally Posted by Francis-134

Basically, I'm trying to understand if adding a time limit to the time it takes for a robot to connect is going to solve any real problems.

Also, the question asked seems oddly worded, i.e. reduce the number of matches you have by building time into the schedule or introducing a connect timer. It's like asking someone if they would like to eat pig slop or chum for dinner; neither option is very good, so why can't we just change the choices?

It's not that teams lost matches or they want to reduce the number of matches. It's the other way around - FIRST wants to add matches.

The problem is that match schedules are being generated with an 8 minute cycle time in anticipation of these connection problems. If they dropped the cycle time to 6 minutes, there could be 25% more matches.

I believe the match schedules used to have a 5 minute cycle time a long time ago.

[Jim Giacchi]

Originally Posted by Chris Hibner

I believe the match schedules used to have a 5 minute cycle time a long time ago.

I believe that was when the robots would connect within 5 - 15 seconds.... oh for the days when we had the little countdown window and the little lights that told us everything was ok.
<rant>
Seriously though I have never been a fan of the new control system. Sometimes simpler is better and if I had a choice, I would switch back to the IFI controller in a heartbeat. Sure, I probably couldn't use any fancy image processing, but my team can never figure out how to program that anyway... and I much prefer setting two dipswitches to the nightmare that is setting up the cRio. I remember being a sophomore in high school and setting up the IFI Pbasic controller, it took me literally 30 minutes and it was the first time I had ever done it. 10 years later, as a college graduate it took days to figure it out. Does that seem right to anyone?
</rant>

Although I do believe that a significant amount of the problem is with the chosen radio. I believe there are many, many options that would be better then the standard wireless network.

[Jack Jones]

60.8% of respondents prefer giving teams a fixed amount of time to connect to the field over reducing the number of rounds in order to build in time so that all teams may connect to the field

Some in this thread has called the above a false dicotomy, which is exactly what it is.

Do they know what percentage of resopndents prefer finishing a few minutes later on Friday or Saturday over bypassing robots or reducing the number of matches?

[apalrd]

Originally Posted by Jim Giacchi

Sure, I probably couldn't use any fancy image processing, but my team can never figure out how to program that anyway...

From what I've found, most vision processing we have done has seriously compromised every other machine function, because the processor just isn't powerful enough to handle vision and other things. Which would require us to dedicate a 10th of a second or so to just processing the image, which is far too long. We tried it all three years the cRio has been around, every time we've abandoned it because of resource consumption.

On a related note, the CMUcam that the IFI system commonly used worked fairly well in testing for us this year. We didn't have the weight or need for it, but it did work a lot better than the Axis cam. Why? Because the processor dosen't do any work, the camera just feeds it a few numbers over a serial port.

I think something like the IFI system with a more powerful processor would be just awesome. Just enough more to allow a minimal OS with threads, and enough math for floating-point and a few trig operations per loop, but not enough more for them to fake us into thinking we can do vision in real-time.