Huh, my previous video really took on a life of its own and was my most popular post ever.
Anyway, back to our regularly scheduled programming. I fixed the bug and this is what the state of my procedural animation is actually like. Still pretty bad, just less funny (but still a bit funny).
[#]procgen
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
The bug turned out to be that I had forgot to call a function which calculates some info (like approximate ground height) derived from the footstep positions. The animals try to lift their legs up to a bit above this approximate ground height, but it was never set and thus zero, which was above their heads.
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
I recently split the system into two parts: One plans footstep positions on the ground for each leg and another poses the legs and body accordingly, each leg always being in a transition between two steps.
The posing part is state-less meaning I can test any footstep configuration out of context.
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
This split means that I can switch out the footstep planner with another one. And this is why I could make a script that just copies footstep positions from a hand-authored animation and tests my body pose algorithm against that.
(I have another script which analyses footstep positions for animation clips.)
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
Further incremental progress. The change here is how I determine the proportion of time the feet are lifted versus standing, plus how I determine how high to lift the feet. More details in replies. This makes the torso altitude much more stable than before.
Next I want to focus on how joint angles.
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
For all the example animations I plotted the foot flight duration as a % of stride duration compared to the stride length relative to leg length. Those are the colored dots.
The white dots is the function I made to approximate this.
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place
For all the example animations I also plotted the foot lift height (relative to leg length) as a function of the horizontal distance between the foot step position on the ground and the hip (also relative to leg length).
The interesting thing is that different gaits with different step lengths take up different amount of that horizontal space. Some reach much further forward/backward than others because the strides are longer. But the parts they have in common overlap nicely.
=> More informations about this toot | More toots from runevision@mastodon.gamedev.place This content has been proxied by September (ba2dc).Proxy Information
text/gemini