User:Oblio/Stopwatch

So, anyway I have a stopwatch now. Also, I am obsessive compulsive (undiagnosed of course).

Fixed Goal Trapping
There are many times that I have wondered about the efficacy of different minor variations on trapping, this section will give me the opportunity to have hard data to back up my thoughts. Many of the tests below don't really have real usefulness except to show some baseline numbers. Usually a trapper knows how many traps they want to lay down, they do it, then they pull. Other times, pullers on teams give people X seconds to lay down as many traps as they can. Hopefully, I'll have data for all of them eventually.

Below is the build I used to test. I will occasionally refer to casting order by number. That will refer to the below bar. Raw data is also linked below. My meager testing is 5 samples per test, which I am comfortable with as long as variance of samples is low.


 * Raw Data

Test A: Lay 4 traps

 * Methodology : Since we are casting 1 of each trap, methodology doesn't matter.
 * Mean : 10.74
 * Std Deviation : 0.105

Test B: Lay 4 traps under Serpent's Quickness

 * Methodology : Since SQ only effects the recharge of skills, I enter the stance immediately following the cast of the first trap.
 * Mean : 10.82
 * Std Deviation : 0.088
 * Notes: It should be unsurprising that Test B == Test A. The differences point towards my fallability in timing by hand.

Test C: Lay 4 traps under Trapper's Speed

 * Methodology : I precast TS and give it a second before starting trapping (to avoid possible aftercast delay which I think should not exist). Time begins on the activation of the first trap.
 * Mean : 8.82
 * Std Deviation : 0.095

Test D: Lay 8 traps

 * Methodology: 6543...5436. Casting Dust Trap other than first means this takes about 8 seconds longer.
 * Mean : 36.00
 * Std Deviation : 0.102

Test E: Lay 8 traps under Serpent's Quickness

 * Methodology: 6543...5463
 * Mean : 29.13
 * Std Deviation : 0.200

Test F: Lay 8 traps under Trapper's Speed

 * Methodology: 6543...5463
 * Mean : 28.04
 * Std Deviation : 0.131

Test G: Lay 12 traps

 * Methodology: 6543...5436...5436
 * Mean : 68.266
 * Std Deviation : 0.371
 * Note: In many ways, this is the first useful statistic. Laying 3 sets of traps will spike kill most mid to end game mobs (assuming you can get clean pulls).

Test H: Lay 12 traps under Serpent's Quickness

 * Methodology: 6543...5463...5436
 * Mean : 48.368
 * Std Deviation : 0.208

Test I: Lay 12 traps under Trapper's Speed

 * Methodology: 6543...5463...5436
 * Mean : 50.416
 * Std Deviation : 0.999 (Boy, I sucked up that batch... almost a second in deviation)

Test J: Max traps to lay under Level 11 QZ

 * Methodology : 54s EW is dropped, then a 47s QZ. As soon as QZ is dropped, trapping begins. The goal is to count how many traps can be laid before QZ dies.
 * Mean :
 * Std Deviation :
 * Notes : Using QZ death as a timer to begin a pull is standard in 3 and 4 man trapping groups. In 2-man spirit trapping groups (as opposed to echo-trappers) QZ + some count is often done. A lvl 11 QZ (16 WS) is a great indicator of the halfway point of trap expiration.

Test K: Max traps to lay under Level 11 QZ w/ SQ

 * Methodology : 54s EW is dropped, then a 47s QZ. As soon as QZ is dropped, trapping begins. The goal is to count how many traps can be laid before QZ dies. SQ WILL OR WILL NOT BE UP THE WHOLE TIME??
 * Mean :
 * Std Deviation :

Test L: Max traps to lay under Level 11 QZ w/ TS

 * Methodology : 54s EW is dropped, then a 47s QZ. As soon as QZ is dropped, trapping begins. The goal is to count how many traps can be laid before QZ dies. TS is maintained for all trap casting and recharging.
 * Mean :
 * Std Deviation :