Explore the possibilities
Robotic finger design simulation
Simulate robotic fingers before fabrication. Compare skins, touch sensors and controllers across manipulation tasks.
Explore the simulation ↗Research Hub
Simulation results
across physical systems.
Compare experiments, inspect recorded results and open the original methods and artifacts.
Explore the capabilities below.
Vary the conditions
Large-scale cyberattack simulation
Simulate cyber disruption and physical failures at a fictional energy company. Compare how defensive responses affect electricity delivery and business operations.
Explore the simulation ↗Create observations
Reactor fault diagnosis simulation
Create hidden reactor faults with similar readings, then test whether an intervention, a longer wait or another measurement reveals the cause.
Explore the simulation ↗Vary the conditions
Reactor diagnosis stress-test simulation
Stress-test reactor diagnosis across concurrent faults, uncertain plant conditions and different physical models.
Explore the simulation ↗Create observations
Different underground models fit the same synthetic magnetic survey
Smooth and compact inversions fit the signal but place magnetic susceptibility differently, leaving the underground structure unresolved.
Explore the simulation ↗Vary the conditions
Controlled ocean setups separate added energy from warming
Comparing carbon-dioxide changes with fixed and warming oceans separates the energy and temperature responses within an idealized model.
Explore the simulation ↗Vary the conditions
Steeper fire scenarios show flames closer to the fuel
Saved temperature and flow fields reveal a change in flame shape, while fuel-bed and ignition changes prevent a slope-only conclusion.
Explore the simulation ↗Carry the computation
Model weather leaves a reusable record for further analysis
Saved time series and jet diagnostics preserve the atmospheric evolution for checks beyond the finished film.
Explore the simulation ↗Explore the possibilities
Variable stage loading lowers the modelled mass proxy
Allowing stage loading to vary produces a lower minimum mass proxy within the reduced model.
Explore the simulation ↗Carry the computation
The early impact pressure is close to an analytical reference
The early-impact model produces a peak shock pressure close to the analytical reference, while the final crater remains outside the calculation.
Explore the simulation ↗Understand the physics
Water changes the proton-transfer pathway
Small molecular models reveal a water-assisted proton-transfer route with a lower calculated energy cost.
Explore the simulation ↗Understand the physics
Surface texture could explain the measured wetting angle
The texture model approaches the disclosed contact angle, but the wetting state remains unobserved.
Explore the simulation ↗Understand the physics
Vibration reduces contacts, but the control is still settling
Gas-free particle runs show fewer contacts under vibration; a stable control is needed to interpret the difference.
Explore the simulation ↗Understand the physics
A similar peak updraft leaves other storm properties unresolved
The simpler model produces a similar peak updraft, while other storm properties disagree substantially with the detailed model.
Explore the simulation ↗Understand the physics
Saved diagnostics trace a growing atmospheric disturbance
Saved pressure, ascent, cloud-water and rainfall diagnostics connect disturbance growth to the evolving model weather.
Explore the simulation ↗Create observations
A close gravity fit leaves depth unresolved
Recovered mass is close to the synthetic target, but a near-target data fit does not recover the bed-depth geometry.
Explore the simulation ↗Understand the physics
Particle-growth fits weaken on unfamiliar publications
Errors increase on held-out publications, exposing a transfer gap that local experiments need to test.
Explore the simulation ↗Carry the computation · Engineering note
A minimal computing workflow reaches saved output
The reported test completes the configured build, submission and time-stepping path and records restart output.
Explore the simulation ↗Carry the computation · Engineering note
A scene script produces a saved test movie
The Manim setup produces a playable movie and retains its source, establishing the rendering path.
Explore the simulation ↗Carry the computation · Engineering note
A fair compute choice still needs a matched benchmark
The research note identifies workload constraints; it provides no measured runtime or cost comparison.
Explore the simulation ↗Carry the computation · Unresolved
Mass imbalance prevents an accepted eductor result
The later run retains substantial mass imbalance and unacceptable residuals, leaving the flow calculation unresolved.
Explore the simulation ↗Original artifacts
Original figures, films and data.
Download the original figures, films, data and selected code. These files retain their scientific detail and limitations.
Download original files +
- chemistry-workflow.png
- chemistry-workflow.svg
- design-workflow.png
- design-workflow.svg
- fire-workflow.png
- fire-workflow.svg
- fluidization-workflow.png
- fluidization-workflow.svg
- synthesis-workflow.png
- synthesis-workflow.svg
- weather-workflow.png
- weather-workflow.svg
- wetting-workflow.png
- wetting-workflow.svg
- laplace-capability-poster.jpg
- laplace-capability.mp4
- laplace-capability.vtt
- castro-impact-workflow.svg
- co2-climate-workflow.svg
- compute-pipeline-workflow.svg
- compute-research-workflow.svg
- eductor-diagnosis-workflow.svg
- gravity-inversion-workflow.svg
- magnetic-inversion-workflow.svg
- month-of-weather-workflow.svg
- project-workflow.svg
- visualization-specialist-workflow.svg
- weather-mechanism-workflow.svg
- beta_sweep_coarse.json
- beta_sweep_fine.json
- fig1_geometry.png
- fig2_anomaly.png
- fig3_sections.png
- fig4_convergence.png
- gravity_inversion_explainer.mp4
- gravity_scene.py
- resolved_spec.json
- scorecard.json
- baroclinic_explainer.py
- baroclinic_wave_explainer.mp4
- build_globe.py
- fkessler_day1_extra_diagnostics.csv
- fkessler_day1_field_stats.csv
- fkessler_day1_summary.png
- fkessler_growth_timeseries.png
- fkessler_physics_movie.mp4
- fkessler_wave_lifecycle_days0_10.csv
- globe_geometry_check.png
- render_movie.py
- baroclinic_wave.py
- fig_diagnostics.png
- fig_wave_lifecycle.png
- fkessler_timeseries.csv
- fkessler_zonal_jet.csv
- one_month_of_weather.mp4
- dc_validation.py
- inversion_scene.py
- mag3d.json
- mag3d.py
- mag3d_summary.csv
- msh3d.json
- msh3d.py
- simpeg_dc_validation.png
- simpeg_mag3d.png
- simpeg_msh_flythrough.mp4
- simpeg_msh_real.png
- simpeg_nonuniqueness.mp4
- validation.json
- analyse_sunconv.py
- analyse_sunconv2.py
- analyse_sunconv3.py
- analyse_sunconv4.py
- extrema_clean.csv
- fig1_snapshots.png
- fig2_metrics.png
- gate_results.json
- hugoniot_decay.png
- impact_theory_baseline.json
- li2d_verification.png
- lunar_impact.mp4
- lunar_impact_1d_verification.json
- lunar_impact_1d_verification.png
- lunar_impact_cavity_timeseries.csv
- lunar_impact_decay_melt.png
- lunar_impact_metrics.csv
- lunar_impact_metrics.json
- lunar_impact_peak_pressure_profile.csv
- lunar_impact_snapshots.png
- pressure_decay_correction.json
- render_iso.py
- render_rt3d.py
- rt3d_instability.mp4
- rt3d_metrics.csv
- rt3d_snapshots.png
- rt3d_vs_2d.png
- rt_amr_grids.png
- rt_growth.png
- rt_instability.mp4
- rt_snapshots.png
- sun_ejection.mp4
- sun_ejection.png
- sun_snapshots.png
- sun_summary.csv
- sunconv_metrics.csv
- tillotson_reference.json
- verified_numerics.json
- manim_smoke_test.mp4
- smoke.py
- config_progress.png
- diagnosis_final.png
- divergence_signature.png
- eductor_case_spec.json
- tutorial_rebuild.png
- camdiag.py
- co2_climate_change.mp4
- co2_climate_change_whatsapp.mp4
- co2_climate_change_whatsapp_status.mp4
- co2_scene.py
- co2_sweep_diagnostics.csv
- ebm.py
- fig_fixedsst_forcing.png
- fig_slabocean_warming.png
- fixedsst_co2_diagnostics.csv
- fixedsst_co2_explore.png
- fixedsst_profiles.csv
- slabocean_annual_series.csv
- slabocean_co2_diagnostics.csv
- slabocean_explore.png
- slabocean_zonal_warming.csv
- era5_to_lagranto.py
- lagranto_validation.png
- driver.py
- ledger.csv
- reconciliation.png
- results.json
- cfd_vs_wellmixed.csv
- cfd_vs_wellmixed_2.png
- cross_validation.csv
- experimental_data.csv
- fig_diagnosis.png
- fig_parity.png
- fig_size_roughness.png
- fit_predictions.csv
- fit_summary.json
- roughness_bet.csv
- threshold_scan.csv
- dem_vibration_results.csv
- dem_window_check.csv
- driver.py
- fluidization_model.png
- ledger.csv
- reproduction_table.csv
- results.json
- silica_fluidization.py
- step1_tfm_sweep.png
- step2_dem_breakup.png
- tfm_velocity_sweep.csv
- avdr_screening.png
- design_space.csv
- design_space.png
- driver.py
- frozen_psi_search.csv
- key_numbers.json
- system_3d_rec.png
- fig_model.png
- fig_verification.png
- model_params.json
- wrf_supercell_diagnostics.csv
- wrf_supercell_evolution.png
- fds_configuration.png
- fds_field_diagnostics.csv
- fds_flame_structure.png
- fds_mechanism.png
- fds_ros_vs_slope.png
- fds_slope_movie.mp4
- fds_slope_movie_portrait.mp4
- fds_slope_results.csv
- fds_slope_summary.png
- cost_estimates.csv
- fes_demo.png
- fig_reanchor_T.png
- functional_error_1w_vs_2w.png
- ladder_rebuilt.png
- m2m5_offset_drift_decomposed.csv
- m2m5_summary.json
- reanchored_vs_T.csv
