Same Path, Different Body#
Week 7 — The Anatomy of a Jump
Question#
Two characters follow the exact same recorded trajectory through the air.
Same height. Same displacement. Same airtime, frame for frame.
Can they still look like they’re jumping differently?
Yes — and nothing about the path has to change#
What’s left to change, once the path is fixed?
What the physics engine owns#
- gravity, launch speed → how high, how long
- air control → how much a held direction redirects you mid-air
- release cut, coyote time, jump buffer → forgiveness around timing
None of this is about how the body looks.
What pose owns#
- arm swing
- leg tuck
- landing compression
Computed from the trajectory, after it’s already decided. Never fed back into it.
A pose that changed the outcome wouldn’t be a pose. It would be a hidden physics parameter wearing a costume.
The comparison#
Load the identical trajectory into two bodies.
Push one body’s arm swing and landing compression high. Push the other’s low. Change nothing about the path.
One reads as loose and reactive. One reads as stiff and controlled.
The overlay of their paths is a single line.
Why this had to be true earlier, not just today#
If pose leaked into collision, every gravity comparison since week 4 would secretly have been a gravity-and-posture comparison.
The separation isn’t this week’s lesson. It’s a constraint the whole course has depended on since week 4.
Across the three presets#
| Preset | Physics feel | A pose that could match it |
|---|---|---|
| Floaty | slow fall, easy mid-air correction | loose arm swing, soft landing |
| Heavy | fast fall, committed direction | tucked limbs, hard compression |
| Precise | moderate fall, locked direction | minimal swing, controlled settle |
These pairings are a starting point, not a rule — the posture comparison tool exists so you can test your own.
Exercise#
In the Jump Lab’s posture comparison, hold one trajectory fixed and design a pose for each of the three presets. Justify each choice in one sentence, on the same fixed path.