|
Re: The "fix-it window"? Jolly good show -or- is 23 cups of coffee/day really excessi
The stipulation in the new Fix-it Window rules that the coding must be complete and not touched two days after ship date is a little harsh. I don't know about most teams, but on my team the programmers get to see the robot in its final form for about 24 hours before it ships. That's not spread out over a week of work; that's one day. Much of the testing of functions can be done apart from the robot, but gain constants are only really determinable using the robot that will be running on the field at regionals or something very similar. In the past we have built a clone and run our tests on it. This way we could get our values close enough that we could adjust them from the stands/practice field and re-download for the next match. It was invaluable.
However, with the new phrase in the Fix-it Window rules we now have very little time to do anything with our code before we have to stick it in a box and put it on a shelf like our robot. I assume I understand the reason for this, that not all the new teams can, or know to, build a clone robot and so they are at a disadvantage. However, I know that if I am going to not work on the code for nearly as long because I have to put it down 2 days after ship-date I'm probably going to learn a bit less and the experience will be overall less valuable for me as a senior-veteran than it could have been. We did most of our autonomous mode development after ship-date last year, ending up with over 30 different autonomous modes, a nifty data structure, and the ability to climb the platforms to the top (assuming our motors don't die on us mid-attempt, which they did later in the season).
This year there is great potential for autonomous mode. The vision tetra challenge is going to take a little thinking, a little writing, and a whole lot of debugging. The dubugging won't happen without the period between ship-date and the first regional. We aren't going to use something in our program that we aren't confident in. It isn't hard to make a box-bot to test a camera routine on, but if we aren't allowed to change our code we can't.
That is why I take issue with the new Fix-it window. While it doubtlessly levelled the playing field on the mechanical end of things the programmers might have been judged a bit too harshly.
Let it be noted that I, and the rest of MOE, will abide by whatever decision FIRST makes on this matter and would definitely encourage others to do the same.
Last edited by PCA : 10-01-2005 at 22:44.
Reason: Rewording
|