userGriglercancel

1,874 Commits over 883 Days - 0.09cph!

2 Years Ago
configured ghostsheet with default length constraints
2 Years Ago
added default length constraint to be used by all bones without their own (before this allowed for full compression, which almost always gives bad results)
2 Years Ago
updated test cloth mat
2 Years Ago
▊ ▆▅▇▇▍▋▇ ▋▆▋▌▍▍▊▅▍▉▊ ▊▄▌▋▆█▍ █▋ ▅▆ ▌▅▍ ▌ ▄▋▅▆▅▊▅▆▇█ ▌▍ █▅▆██▋▍ ▌▍▋▉▌▌██▄▅ ▉█▊▌▅▌▌ ▄▊▇ ▄█ ▌█▇▇██ ▇▉▊▋▅▅▅ ▍▆▆▇▇▋ ▅▅▄▌ ▄▆▄ ▇▆██▋▇▍▇▅▆▅ ██▍ ▅▉▉▋▇ █▋ ▆▉▉ ▊▇▊▌▇▋▊ ▇▅▇▄▄▄▅ ▍▋▆▋▋ ▄▅▌▍▋▄▊ ▉▄▋▌▇▅▊
2 Years Ago
cape cloth prefab setup
2 Years Ago
merge from main
2 Years Ago
updated test scene
2 Years Ago
resetting bone transforms on cloth when it's re-enabled to avoid spike in velocity with LODing
2 Years Ago
pickaxe chain controller LODing
2 Years Ago
min/max length constraints respected properly and slightly reduced precision in collision detection
2 Years Ago
corrected LODing on cloth controllers for ghostsheet
2 Years Ago
Re-mapped weight curves to relaxation values - also reorganised cloth material SO to make it a bit more readable
2 Years Ago
some chain material setups
2 Years Ago
fixed wrong rotation constraint axis being used for one of the limits
2 Years Ago
basic gizmos on constraints
2 Years Ago
fixed length modifier and gravity on capes
2 Years Ago
cape cloth material
2 Years Ago
ghostsheet setup, split bone controllers for better collision control
2 Years Ago
improved edit-time gizmos so you can see colliders and constraint setups
2 Years Ago
split gizmos between configuration and playmode
2 Years Ago
cloth changes - filtering raymarched collisions with a large distance from bones to improve stability with extreme rotations and the cost of some clipping in these cases - made sibling constarints a bit stricter
2 Years Ago
constraint tuning - sibling linear constraints are directonal, taking the average of the orientation between them - parent receives half the constraint easing as well - tuned some cloth materials for these changes
2 Years Ago
converting rotation constraints to radians when processed
2 Years Ago
merge from main
2 Years Ago
drop tickrate on some cloth assets
2 Years Ago
contacts are found once at the beginning of the tick and solved alongside distance constraints
2 Years Ago
Separated rotation constraint data from bone data
2 Years Ago
improved collision responses to be resolved positionally without adding velocity into the system
2 Years Ago
fixed summing of position adjustments with sibling constraints and removed their rotation contribution
2 Years Ago
reduced max tick rate on cloth
2 Years Ago
updated test scene
2 Years Ago
simplifying burst cloth in prep for separation as jiggle-only solution - position based dynamics, with hard rotation constraints - configurable relaxation factors to allow for artist control over bounciness of constraints
2 Years Ago
additional profiling, a few quick wins on performance
2 Years Ago
initial skinned oni cloth setup for ghostsheet
2 Years Ago
small perf modifications to oni cloth and initial setup for global solver updater
2 Years Ago
adding obi cloth - still need to clean out samples and other fluff
2 Years Ago
switch to velocity verlet integration
2 Years Ago
hitbox gizmos match their actual size
2 Years Ago
more burst cloth - changed integration method - reduced position and orientation to a single spring value each, collision still managed with separate spring - cleaned up unused cloth material properties
2 Years Ago
further burstcloth modifications - collisions resolved with spring forces - added dirty broadphase to collision resolving - capsules colliders reinterpreted as sets of spheres, cheaper testing outweighs increased collider count - added collections packaged for better native containers
2 Years Ago
merge from main
2 Years Ago
test scene changes
2 Years Ago
some initial balanced values for some cloth assets
2 Years Ago
improved burstcloth - replaced hard positional constraints for length and shape to force based spring system approach, aiming to give artists more control over the bounciness of joints but also allows motion to propegate better - sibling constraints are also treated as length forces - removed now-unused fields from material - other small improvements to reduce overhead
2 Years Ago
burstcloth improvements - rotational constraints respected - constraint normals are correct now, significant improvement to length constraints - some jobsystem/burst optimizations
2 Years Ago
merge from main
2 Years Ago
watersystem handles casts made from underwater properly - no false positives when slightly underwater - able to get a positive result when looking at the surface from underwater
2 Years Ago
merge from main
2 Years Ago
merge from main
2 Years Ago
added simple pid implementation for controlling orientation when steering - needs more tuning