Chief Delphi

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   General Forum (http://www.chiefdelphi.com/forums/forumdisplay.php?f=16)
-   -   Team Update 6 (http://www.chiefdelphi.com/forums/showthread.php?t=154328)

Chris is me 27-01-2017 15:27

Re: Team Update 6
 
Quote:

Originally Posted by MasterMentor (Post 1637153)
I appreciate the summary, but it would be more appreciated if you'd not incorrectly summarize the Team Update.

Both the FIRST and the AndyMark fields are both "official fields" used at Official FIRST Robotics Competition events. Otherwise, are district events "official"?

I don't understand what you're trying to say here. Both the update and the original post use "official" to refer to the FIRST supplied field. The manual once said all regionals and championship fields use the FIRST design, now it has exceptions (and as the only other legal design is the AndyMark one, it follows that those events are using the AM field). He didn't say anything about AndyMark no longer being an official field.

---

As for the update, the manual always allowed you to grab your own rope - it now explicitly allows you to DAMAGE your own rope. That is the change.

Koko Ed 27-01-2017 15:39

Re: Team Update 6
 
Quote:

Originally Posted by GeeTwo (Post 1637124)
Hooray! It also wastes the time of the thousands of people who read all of the answers as they are posted. ::rtm::

This why CD is a good training program to wean the inexperienced from making such a painful error in judgement. Being told to "search before you post" enough times should set you on the right path in the FIRST community!

rich2202 27-01-2017 16:00

Re: Team Update 6
 
Quote:

Originally Posted by Kevin Sevcik (Post 1637164)
The question is if I can place the gears for Rotor 2, THEN place the gear for Rotor 1, then start turning Rotor 2.

Gears can be placed in any order. To wit: The pre-populated gears that may be removed for Worlds.

FYI: The issue of Rotor order was also discussed in Q82
https://frc-qa.firstinspires.org/qa/82

FYI2: Once a gear is used to turn a rotor, it cannot be removed.
H10

jvriezen 27-01-2017 18:08

Re: Team Update 6
 
Quote:

Originally Posted by rich2202 (Post 1637180)

FYI2: Once a gear is used to turn a rotor, it cannot be removed.
H10

Not quite:
H10: GEARS stay installed. Once a ROTOR is started, the PILOT may not remove any GEARS used to start it.


This would indicate gear sets can be rotated, short of starting the rotor, and a non-pre-installed gear can then be removed. If you can get the same or another gear back on the gear set and continue rotating before 10 seconds elapsed, you'd still be good, I'd think, otherwise you have to start a new three rotations.

Depends upon where the sensors are, I'd think one on first and last gear in set would be sufficient. But maybe there's only a sensor on one of those and they rely on refs to see that you don't cheat the system.

s-neff 27-01-2017 18:55

Re: Team Update 6
 
Quote:

Originally Posted by OP
It's week three, and it looks like FIRST is STILL refining field designs.

Been waiting for that particular update since Kickoff. Still waiting for some missing [non-critical] dimensions. Not impressed with the quality of the field drawings this year, not surprised since it's the most complicated field I've ever seen. Sympathy for the GDC engineers working on it.

Quote:

Originally Posted by jvriezen (Post 1637236)
This would indicate gear sets can be rotated, short of starting the rotor, and a non-pre-installed gear can then be removed. If you can get the same or another gear back on the gear set and continue rotating before 10 seconds elapsed, you'd still be good, I'd think, otherwise you have to start a new three rotations....

But as soon as the PILOT starts another rotor with that gear, it's locked down... so why bother spending time turning the rotor and risking an out-of-order-rotors penalty if you go too far, if you'll still need to pull a gear off the lift to complete it later?
Quote:

to see that you don't cheat the system.
How about we just don't try to cheat the system?
Folks are overthinking the gears rules.

Siri 27-01-2017 19:25

Re: Team Update 6
 
Quote:

Originally Posted by s-neff (Post 1637246)
But as soon as the PILOT starts another rotor with that gear, it's locked down... so why bother spending time turning the rotor and risking an out-of-order-rotors penalty if you go too far, if you'll still need to pull a gear off the lift to complete it later?

I figure most of these questions relate primarily to top-tier autonomous routines. For instance, if you're expecting a 3-gear autonomous from your alliance, and two gears arrive first roughly at once, you'd probably prefer to put them in Rotor 2 and start rotating that set while gear 3 shows up. (This assumes Rotor 1's automatic start is relatively quick in comparison to Rotor 2.) The timespan you wait will almost certainly be less than 10 seconds in a 15 second total autonomous. However, if gear 3 subsequently fails (robot missing peg, etc), you'd still want to stop turning Rotor 2, remove a gear, and put it in Rotor 1 for the single autonomous bonus. This approach rests on the legality of rotating (not engaging) rotor cranks out of order and removing a gear after having rotated it (but not engaged it).

These are edge cases of course, but they could make a very big difference in a few teams' seasons. More teams will be working toward being those few, and the work to reach that point a huge upfront investment that benefits from understanding all these little details.

rich2202 27-01-2017 19:35

Re: Team Update 6
 
If Rotors only start if Gears are installed in Rotor Order (section 3.4.2), do "turns" out of sequence count? Does turning gears for Rotor 3 do any good if Rotor 2 has not yet been started?

EricH 27-01-2017 19:41

Re: Team Update 6
 
Quote:

Originally Posted by rich2202 (Post 1637263)
If Rotors only start if Gears are installed in Rotor Order (section 3.4.2), do "turns" out of sequence count? Does turning gears for Rotor 3 do any good if Rotor 2 has not yet been started?

Nope. Rotors have to be started in order, so any work you do to turn Rotor 3 gears is better saved for Rotor 2 gears (unless Rotor 2 is rotating).

Kevin Sevcik 27-01-2017 20:34

Re: Team Update 6
 
Quote:

Originally Posted by s-neff (Post 1637246)
But as soon as the PILOT starts another rotor with that gear, it's locked down... so why bother spending time turning the rotor and risking an out-of-order-rotors penalty if you go too far, if you'll still need to pull a gear off the lift to complete it later?

Pretty sure there's no such thing as an out of order rotor penalty.

Quote:

Originally Posted by rich2202 (Post 1637263)
If Rotors only start if Gears are installed in Rotor Order (section 3.4.2), do "turns" out of sequence count? Does turning gears for Rotor 3 do any good if Rotor 2 has not yet been started?

I think this is covered by the new text in the update. "If a GEAR set corresponding to the next sequential unengaged ROTOR..." implies that any rotor past the next sequential unengaged rotor isn't going to accumulate any rotations at all. Which only makes sense. Which further shoots down plans to prerotate a gear set, because only the next gear set will accumulate rotations. So there's no point to spinning up rotor 4 unless 3 is already going. And if 3 is going, you're not spinning 4 without a full gear set, at which point you spin it up completely.

The only reasonable case for placing gears out of order is a highly choreographed autonomous, as suggested by Siri, and by me in other threads. Possibly also disinformation relating to your total gear count as well.

s-neff 27-01-2017 21:05

Re: Team Update 6
 
Quote:

Originally Posted by Kevin Sevcik (Post 1637288)
Pretty sure there's no such thing as an out of order rotor penalty.

Accurate. No incentive != penalty, apologies for the misinformation.
There *is* a penalty H14 for starting a rotor by any method besides turning the crank, which covers most of the rest of the 'creative gear scoring'
Quote:

...The only reasonable case for placing gears out of order is a highly choreographed autonomous
Doesn't this get nope'd by the same logic? [must activate rotor 1 to start counting rotations on rotor 2...] Maybe I'm misunderstanding the proposed autonomous.

Siri 27-01-2017 22:01

Re: Team Update 6
 
Quote:

Originally Posted by Kevin Sevcik (Post 1637288)
I think this is covered by the new text in the update. "If a GEAR set corresponding to the next sequential unengaged ROTOR..." implies that any rotor past the next sequential unengaged rotor isn't going to accumulate any rotations at all. Which only makes sense. Which further shoots down plans to prerotate a gear set, because only the next gear set will accumulate rotations. So there's no point to spinning up rotor 4 unless 3 is already going. And if 3 is going, you're not spinning 4 without a full gear set, at which point you spin it up completely.

The only reasonable case for placing gears out of order is a highly choreographed autonomous, as suggested by Siri, and by me in other threads. Possibly also disinformation relating to your total gear count as well.

I'm having trouble reading the new TU as definitively saying out-of-order turns don't count (even assuming actual engagement is sequential). To re-requote it with my emphasis: When a GEAR set for ROTORS 2, 3, or 4 is complete, a CRANK, a handle located with the first GEAR in the set, can be turned which engages the corresponding ROTOR. It takes three (3) full rotations to engage the ROTOR. If a GEAR set corresponding to the next sequential unengaged ROTOR remains idle for more than ten (10) seconds, the rotation count resets to zero (0).

And Q82 (in part): "ROTORS only start if GEARS are installed in ROTOR order." In other words, you won't be able to start ROTOR 4 until all other ROTORS have been started, etc.

As evidenced by the "in other words" of Q82, this is not a particularly tightly-worded section. (Requiring gears to be installed in rotor order is not in fact the same thing as starting rotors in order.) I suspect that HQ hasn't yet realized that this is of legitimate importance for some teams. As I read it now:
1) Rotors must be started in order.
2) A crank can be turned when a gear set is complete.
3) Rotation count resets when the next unengaged rotor has been idle for more than 10 seconds.

I don't think that (3) explicitly means rotations within those 10 seconds while that rotor isn't the next in sequence don't count. My reading of the logical flow is ambiguous across any of the following Step 5s:
Step 1 @ t=-3 Place Gear 1 in Rotor 2
Step 2 @ t=-1 Place Gear 2 in Rotor 2
Step 3 @ t=0 Turn Rotor 2 crank for 2 rotations
Step 4 @ t=3 Place Gear 3 in Rotor 1
Then:
  • Step 5 @ t=6 Turn Rotor 2 crank another rotation and engage Rotor 2
  • or Step 5 @ t=6 Turn Rotor 2 crank another rotation but cannot engage Rotor 2 because previous rotations didn't count
OR
  • Step 5 @ t=10 Turn Rotor 2 crank another rotation but cannot engage Rotor 2 because it is more than 10 seconds from its first rotation
  • or Step 5 @ t=10 Turn Rotor 2 crank another rotation and engage Rotor 2 because the prior rotations count and the clock doesn't start until Rotor 1 engages (such that Rotor 2 is next in sequence)
OR
  • Step 5 @ t=14 Never receive Gear 3, and remove Gear 2 from Rotor 2 to install in Rotor 1, engaging Rotor 1.
  • or Step 5 @ t=14 Never recieve Gear 3, and remove Gear 2 from Rotor 2 and install in Rotor 1 but don't get points for engaging Rotor 1 because you violated the paragraph about Figure 3-13 (despite HQ's interpretation of this clause in Q82).
  • or Step 5 @ t=14 Turn Rotor 2 crank another rotation and cannot engage Rotor 2 (because now either way you're outside the 10-second window)

Trying to boil this down into a Q&A, I'm thinking:
1) It is understood that GEARS cannot be removed after a ROTOR is engaged. However, can a GEAR be removed after a ROTOR is rotated but before it is engaged? Does this differ if the rotating is less than a full revolution or is accidental or strategic (in the estimation of the referees)?
2) It is understood that ROTORs must be engaged in ROTOR order. Must GEARS be also installed (placed without rotating) in ROTOR order? [A82 seems to interpret the manual p23 this way, but the manual still says "installed" rather than engaged or rotated.]
3) If GEARS need not be installed in ROTOR order, can ROTOR N be rotated (for accumulation of the required 3 rotations) before ROTOR N-1 is engaged, assuming the third rotation and engagement of ROTOR N still occurs after N-1 is engaged?
4) If ROTORs need not accumulate rotations in ROTOR order, does the (10) second idle window from TU6 start with the first rotation or when that ROTOR becomes the next sequential unengaged ROTOR?


Good point about deception regarding the number of gears you still need. My head was still stuck on autonomous.

Kevin Sevcik 27-01-2017 22:38

Re: Team Update 6
 
Quote:

Originally Posted by s-neff (Post 1637289)
Doesn't this get nope'd by the same logic? [must activate rotor 1 to start counting rotations on rotor 2...] Maybe I'm misunderstanding the proposed autonomous.

My personal 3 gear auton plan:
1. Fastest delivery to center peg, Pilot A lifts gear and hands to PB or sets near PB.
2. PB lifts and grabs gear with second fastest delivery. (You should know which this is.)
3. PB places both gears, finishing around 10-12s into match. And starts cranking for dear life.
4. At some point before (15 - 3 cranks) seconds, Pilot A has lifted last and slowest gear and slotted into Rotor 1.

The idea is that Rotor 2 is the slowest part of the process. If you can start it first and feed it the fastest two gears, you probably have the best chance at starting both Rotors.

Honestly, sans team this year I have nothing to ground my speculations, thus my musings about 3 gear autons and gear count deception potentially playing a role at very high levels.

Siri,
I'll admit that a completely literal reading of that section with new wording could mean that the 10 second timeout doesn't apply to any rotor past the next unengaged one. Which would make this possible, assuming no auton rotors:
1. Collect 5 gears + reserve.
2. Install on Rotor 4, spin thrice.
3. Remove, install on Rotor 3, spin thrice.
4. Install 2 leftover on Rotor 2, spin thrice.
5. Install next 6 gears on Rotor 4.
6. Install last, 12th gear on Rotor 1.
7. Rotors 2-4 engage instantly. Or within 3 PLC cycles.

That just seems ridiculous and would make the 10 second timeout pointless. So I'm assuming the GDC is just poor at fully defining the Rotor behavior in prose. It'd be easier for me if they just posted that segment of the FMS code, but that's probably not going to happen.


All times are GMT -5. The time now is 12:02.

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi