High Precision Implementation #35
Conversation
This reverts commit ee40051.
Remaining required changes:
|
src/main/java/ninjabrainbot/io/preferences/NinjabrainBotPreferences.java
Outdated
Show resolved
Hide resolved
src/main/java/ninjabrainbot/gui/panels/settings/CalibrationPanel.java
Outdated
Show resolved
Hide resolved
| alpha += preferences.crosshairCorrection.get(); | ||
|
|
||
| // Determined experimentally, exact cause unknown | ||
| alpha -= 0.00079 * Math.sin((alpha + 45) * Math.PI/180.0); |
There was a problem hiding this comment.
I dont know if im comfortable adding this if we know so little about it. Ill think about it. What was your method of gathering the 400 data points?
There was a problem hiding this comment.
Spawn into a random world, throw and measure an eye, then move ~20 blocks to the side and measure another eye (2 throws to guarantee 100% confidence). I repeated this 200 times, after each world printing the angle and error for the 2 throws using a modified ninbot.
You can also check for a particular position by changing your yaw with commands until pixels change, then find how much the yaw interval for each pixel is shifted to the left/right. I've used this to confirm that the effect seems to also be present without tall res and for different distances.
I expect that we'll eventually get a better understanding of why this happens, and then we can probably optimize the adjustment. But I think the main thing is that the current adjustment is much better than no adjustment.
Changes: