R•E

Rocket Car Physics Lab

Original experiment

Kestrel R1

Choose your experiment.

0MPH
WHEELS DOWN
Pitch 0°Roll 0°Accel 0.00 gLateral 0.00 gLoad +1.00 g
RANDOM.EXPRESS / KESTREL R1

An original rocket car. Six bite-size experiments. Your move.

Force-based conceptual model, original vehicle. Reveal-inspired demonstrations and unvalidated experiments are labeled.

Four moves. One kickflip.

W / Drive → Space / Jump → D / Right → release.

Follow the timed cues at the right. Release the previous control at each cue. Shift is not needed.

Make yourself comfortable.

Assist applies bounded jet forces and uses reserve. It cannot guarantee a landing.

Touch controls stay below the driving view. Settings save on this device; experiment medals last for this session.

Play with the idea.

Rocket Car Physics Lab is a conceptual vehicle simulation about grip, gravity and finite thrust. Drive the original Kestrel R1 through six experiments. No account or download required.

Free Lab

Unlimited gas (U) powers all jet and attitude nozzles without draining your saved reserve. Turn it off to resume that reserve. The switch stays selected through Free Lab restarts and mode changes in this session; scored challenges always use finite gas. Hard crashes recover automatically, refill reserves and release held controls. Settings stay selected. Use Restart or R any time.

Boost launch target

The original 1,800 kg Kestrel model is tuned to approximately 0.81 s from full standstill, with no rollout, when Drive and Thrust start together, the reserve is full and default settings are used. Its two rear nozzles apply 20.2 kN each; they consume the finite 310 kN·s shared reserve. These are game assumptions, not Tesla specifications. Delaying either input changes your measured time.

The official reveal announces 0.81 seconds at about 18:30, after rejecting a 0.17 display. The available source does not establish rollout, timing equipment or independent validation. Matching the announced number does not validate this model.

Controls

WASD or arrow keys drive and brake. Space fires upward jump jets. Shift fires forward thrust. In the air, left/right commands roll jets. Release to activate the optional, bounded attitude assist. Q/E command paired side jets; X fires the front stopping nozzles. Tire brakes have no effect in free air. F toggles suction, G changes the wing. R restarts, Escape pauses. Touch buttons support several fingers.

What is real, and what is imagined?

Inverted ramp, fleet kickflip and pressure sprint are inspired by the October 15 Tesla reveal: ramp at 6:03, kickflip at 10:56, aero at 14:38 and acceleration at 15:58. See the official flip clip. These are gameplay adaptations, not exact reconstructions. Visible white plumes in the official clip inform the effect; exact nozzle hardware and pressures were not established from the footage. Our original roll system uses opposed side forces at different heights to create a torque couple; counter-firing slows the roll. White clouds represent condensation in cold exhaust, not visible nitrogen or combustion smoke.

Reporting from Tom’s Guide and Car and Driver corroborates the demonstrations. Exact choreography was not independently inspected frame by frame. The announced 0.81-second 0–60 time is a manufacturer event claim, not an independently verified measurement or this game’s target.

Tunnel inversion, precision landing and free lab are fictional what-if experiments. Suction works only close to a surface; it does not lift the car in free air. All jets share a finite 310 kN·s impulse budget; Restart refills it. An independently powered suction fan operates only with a seal. Optional attitude and landing assistance uses the same jet reserve and is visible in the HUD. It is disabled during tunnel falls; only your explicit jet inputs can interrupt freefall. Wing down improves grip but adds drag. Suction holds nearby surfaces and adds fan drag. Flatten the wing or switch suction off for faster straight runs, but give up grip and inverted adhesion. Conceptual AWD drivetrain: 1,400 N m combined motor torque, 9:1 reduction, 0.37 m rolling radius, 94% driveline efficiency, 650 kW motor power cap and 18,000 rpm limit. Low-speed wheel force is about 32 kN before the tire-grip limit; it tapers with speed. Braking blends up to 200 kW / 9 kN regeneration within the existing 24 kN total brake request, not on top of it. These are original lab assumptions, not Tesla specifications. Declared conceptual parameters: 1,800 kg, 9.81 m/s² gravity, tire friction coefficient 1.1, 14 kPa fan pressure across 2.6 m² effective sealed area, 45 mm maximum seal gap, 70 kW fan and 8 MJ fan energy. Jets: 40.4 kN rear, 24 kN front, 9 kN lateral, 42 kN lift, 3.6 kN·m roll authority. Tank feasibility, fan sealing hardware, structural loads, tire compliance and detailed aerodynamics are not validated. It is not engineering validation, vehicle advice, or an affiliated Tesla product.

Angle and G telemetry

Pitch is nose elevation above the world horizon. Roll is bank relative to that horizon, positive when the right side is up; roll is undefined at vertical pitch. Accel and Lateral are kinematic COM acceleration projected forward and right in the car frame, divided by 9.81 m/s². Load is vertical specific force (acceleration minus gravity) along car-up: stationary level support is +1 g, stationary inverted support is -1 g and freefall is about 0 g. It is not tire normal load. G values use a 0.18-second smoothing filter. Contact transitions show a dash because the simplified impact solver does not resolve a meaningful peak load.

Sports car benchmark

The adjacent original sports coupe follows an independent synthetic speed curve calibrated to a 3.0-second full standing start. This is reconstructed from Car and Driver's rounded 2.8-second result plus the 0.2-second rollout they removed in their 2020 Corvette Stingray 3LT Z51 FE4 test. See their timing method. The intermediate speeds are modeled, not measured. Both cars start their clocks together at the first driving input; no rollout is subtracted. The player follows the force model and is not guaranteed to win.

Braking uses the separate MotorTrend 2020 Corvette Z51 test result of 97 feet from 60 mph, modeled as constant deceleration. It is not raw vehicle telemetry or a same-day comparison. Separate braking tests start both lanes at exactly 60 mph.

Same speed. Different brakes.

Both lanes start at exactly 60 mph. The test rig applies full tire brakes immediately. Choose whether the Kestrel also uses its front stopping jets; the benchmark has tire brakes only.

Suction is OFF. Same level road, full reserve. Benchmark: 97 ft, modeled from the published Corvette Z51 test.

EXPERIMENT COMPLETE

Beautifully done.

Original browser game · Updated October 18, 2026

Boost, flip and stick the landing.

Rocket Car Physics Lab is a free browser game about cold-gas thrust, traction and timing, inspired by reported demonstrations at Tesla’s October 15, 2026 Roadster reveal. Take the original Kestrel R1 from a desert launch lane to an inverted ramp, then try a fleet-clearing kickflip.

Random.Express created the vehicle, graphics and sound for this independent experiment. It is not affiliated with Tesla, and its physics values are gameplay assumptions rather than measured Roadster specifications.

Six ways to explore grip and gravity

How to play

Choose a mode from the game menu. The kickflip briefing waits until you press Start, and central cues tell you what comes next. WASD or arrow keys drive and steer, Space fires jump jets, and Shift fires rear thrust. In the air, left/right commands roll jets. Q/E commands side jets; X fires the front stopping jets. Tire brakes work on the ground, not in flight.

Use R to restart with a full reserve or Escape to pause. On a phone, use the named touch buttons, including multiple buttons together. Settings include sound, volume, landing assistance and reduced effects. New touch sessions start with low effects for smoother play.

Questions about the rocket-car game

What powers the car and its jumps?

The Kestrel combines a modeled electric drivetrain with finite cold-gas jets. Rear jets accelerate it, upward jets lift it, and paired roll nozzles turn it in flight. White plumes visualize cold-exhaust condensation. Restart refills the reserve; waiting does not recharge it.

How can the car drive upside down?

The game models a suction fan that adds contact force only while the car stays close to a surface. Losing the seal removes that support, so gravity can pull the car away. The tunnel challenge is a fictional extension of the idea, not a demonstrated Roadster capability.

Are the acceleration times Tesla performance measurements?

No. Your times come from this game’s original force model. The manufacturer’s announced reveal figures are separate claims. The adjacent sports coupe uses a derived 3.0-second standing-start reference: Car and Driver reported 2.8 seconds with 0.2 second of rollout removed. Its intermediate speeds are synthetic. The separate 60–0 reference uses MotorTrend’s 97-foot Corvette Z51 result, modeled as constant deceleration. These are different published tests, not a same-day comparison.

Which challenges come from the Roadster reveal?

The inverted ramp and fleet kickflip adapt reported reveal demonstrations; the acceleration lane explores the thrust idea through an independent benchmark. Upside-down tunnel driving, precision landing and free lab are fictional experiments. The choreography and hardware are simplified for play.

Is it free, and does it save progress?

Yes. Play free in a modern browser with JavaScript and WebGL; no download or account is required. Preferences and dismissed tips are saved on this device. Challenge medals last for the current session.

Sources and model notes

Reveal context: official Roadster reveal, official kickflip clip, and Car and Driver’s reveal coverage. Benchmark context: Car and Driver’s Corvette test and rollout timing method, plus MotorTrend’s Corvette Z51 test.

The game’s Sources & physics panel explains its assumed mass, forces, drivetrain and telemetry. Use the lab to explore cause and effect, rather than predict a real vehicle’s limits.