18,700
edits
XML:ONCC: Difference between revisions
clarifying <Targeting>
(clarified <AIConstants>, caught big oops with "friendly" timers) |
(clarifying <Targeting>) |
||
| Line 396: | Line 396: | ||
{| class="wikitable" width="100%" | {| class="wikitable" width="100%" | ||
!width=150px| XML tag | !width=150px| XML tag | ||
!width=120px| | !width=120px| Content type | ||
! | ! Description | ||
|- | |- | ||
| <StartleMissAngle> | | <StartleMissAngle> | ||
| float | | float | ||
| | | Always has value 0.5 for TURR and ONCC; multiplied against target's distance to get error in AI's/turret's first wild shot | ||
|- | |- | ||
| <StartleMissDistance> | | <StartleMissDistance> | ||
| float | | float | ||
| | | Always has value 25 for TURR and ONCC; max value for result of multiplication above | ||
|- | |- | ||
| <PredictAmount> | | <PredictAmount> | ||
| float | | float | ||
| | | Always has value 1 for TURR and ONCC; accuracy multiplier for prediction of target's position | ||
|- | |- | ||
| <PredictPositionDelayFrames> | | <PredictPositionDelayFrames> | ||
| int32 | | int32 | ||
| | | Always has value 0 for TURR and ONCC; max value for how many frames of lookback can be used to predict target's position | ||
|- | |- | ||
| <PredictDelayFrames> | | <PredictDelayFrames> | ||
| int32 | | int32 | ||
| | | Always has value 5 for TURR and ONCC; number of frames delay in recognizing target's current position | ||
|- | |- | ||
| <PredictVelocityFrames> | | <PredictVelocityFrames> | ||
| int32 | | int32 | ||
| | | Always has value 15 for TURR and ONCC; number of frames over which AI can average target's velocity to help predict position | ||
|- | |- | ||
| <PredictTrendFrames> | | <PredictTrendFrames> | ||
| int32 | | int32 | ||
| | | Always has value 60 for TURR and ONCC; size in frames of location buffer for target | ||
|} | |} | ||
The higher <PredictPositionDelayFrames> is, the further back the AI will look in the position buffer which contains past location information for the target, however this value is always 0 in Oni's data, meaning the current location is used. The recognition of the target's current position is delayed by <PredictDelayFrames>, always 5/60 of a second, which represents something like "reaction time when tracking target". <PredictAmount> can be used to add inaccuracy to this prediction by choosing a value below 1.0, but in Oni's data it is always 1.0. | |||
===<WeaponSkills>=== | ===<WeaponSkills>=== | ||