Tune the space between intention and motion
“Floaty” is a useful reaction, but an incomplete brief. Is the object slow to respond, slow to stop, or hard to read? Isolate a relationship you can actually tune.
In this guide
Allow 30–45 minutes for the reading, interactive experiment, and a notebook exercise. Bring a prototype or use our Night Mail courier example. No account or code required.
By the end of this lesson
- Describe a control complaint as an observable behavior.
- Compare acceleration and coasting while keeping the task fixed.
- Separate movement changes from decorative feedback.
Give a feeling a specific cause to investigate
Steve Swink’s Game Feel examines real-time control, simulated space, and the feedback that emphasizes interaction. Use those connected parts to investigate a complaint rather than applying a generic layer of visual polish.
If the courier overshoots a doorstep, several things could be happening. Acceleration may build too much speed. Release may leave too much momentum. The camera may hide the threshold. The player may not know which part of the character is tested for delivery. Each explanation suggests a different experiment.
First hold the destination and input task fixed. Change a movement parameter, then replay the same approach. Only afterward change the camera or effects. Otherwise the result may improve without showing which relationship helped.
Park the delivery cart
Advance a cart in fixed time steps and try to stop inside the delivery zone. Adjust acceleration and coasting, then replay from the same starting point. This is a movement diagram you can operate at your own pace; a live prototype is still needed to judge timing and comfort.
A delivery, without the overshoot
Park the cart between positions 340 and 390 with speed below 10. Each button advances 0.1 seconds of simulated time. Push the opposite direction to brake.
Position 60.0 · velocity 0.0 units/s · 0.0s simulated. Approach the zone and release or brake. The dots show recent positions.
What this diagram simulates
While pushing, acceleration changes velocity, capped at ±160 units/s. While coasting, velocity is multiplied by the selected retention each 0.1s. Position integrates in five 0.02s substeps. Track edges stop the cart. Changing a setting resets the attempt. This lets you inspect motion, but does not test real-time input feel.
What did you change?
The sandbox clamps the cart at the track edges and has no collisions or slopes. Its numbers are illustrative units, not recommended settings for your engine. A useful result is a prediction such as “less coasting should reduce overshoot after release,” followed by a comparison.
| Knob | Immediate effect | Tradeoff to watch |
|---|---|---|
| Acceleration | How quickly velocity changes while steering. | Fast response may make tiny corrections harder. |
| Coasting | How much velocity remains after release. | Long glides need earlier braking or more space. |
| Target width | How precise the final position must be. | A narrow zone may test precision instead of route planning. |
Let feedback explain the motion
A wheel trail can reveal momentum. A small tilt can communicate acceleration. A landing sound can identify contact with the ground. Add an effect to emphasize something the player needs to read, then see whether it still works with the effect reduced or muted.
Do not make essential state depend only on color, sound, or camera shake. Keep a stable visual cue and offer comfortable alternatives. In a delivery game, a legible target outline may preserve the interesting parking problem better than a dramatic screen flash.
A discrete diagram cannot prove a control feels good in real time. Take your chosen relationship into a playable build on the actual input device. Ask the tester to approach, stop, reverse, and recover; watch the corrections rather than only the final success rate.
Replace “make it snappier” with a comparison
- Choose one repeatable action and describe the visible complaint.
- Record the current movement settings and keep a copy of the build.
- Change one relationship: response, release, or feedback.
- Compare the same approach with both settings on the intended device.
- Record what improved and what became harder. Keep the setting that serves your intended action.
Sources & further reading
Sources for the ideas discussed here, plus further reading. The lesson’s examples and interactive exercises are original.
Put your idea to work with the Dashpop Game Engine. Describe a game, play it quickly, and shape it yourself. Start with one change you can test.