Take Control

Configure variable jump height and air control, investigate coyote time and jump buffering, and explain the tradeoffs you landed on.

Due: 2 April 2027 · Weight: 30%

The brief#

Design a jump that’s forgiving without being formless.

Weeks 4 and 5 gave you two independent levers: how much control you have while airborne, and whether height responds to how long you hold. This brief asks you to set both deliberately, on top of the two forgiveness windows — coyote time (how long after leaving a platform’s edge a jump still counts as if you were standing on it) and jump buffer (how early a press before landing still fires the moment you touch down) — and then prove your choices with trials rather than just asserting them.

There’s no single correct setting. A speedrun-precise jump and a beginner-forgiving jump can both be defensible; what’s being assessed is whether you can show what you built and say what it cost.

What you submit#

• A configuration you can hand over. The Jump Lab’s Copy configuration control turns your current settings into a link; paste that link (or the exported values it encodes) into your report so a marker can reload your exact setup.
• Trials, not descriptions of trials. At least three runs demonstrating variable height (a short press versus a full hold) and air control (a mid-air correction), each with its measured height, displacement and airtime.
• Two forgiveness trials. One showing a jump pressed slightly after leaving a ledge still launching (coyote time), and one showing a jump pressed slightly before landing still firing on touchdown (jump buffer). State the window you set for each, in milliseconds.
• A short report (roughly 300–500 words) on what you prioritised — responsiveness, forgiveness, or predictability — and what that choice cost you elsewhere.

The visual controls in the Jump Lab are a complete route to every part of this brief. If you want to explore the optional starter-code path instead (or as well), say so in your report — it earns no automatic bonus on top of an equivalent visual-controls submission, since the brief is about the tradeoff you can demonstrate, not the tool you used to reach it.