Interactive engineering labs

Build it.
Stress it.
Understand it.

Four browser-based systems for examining material response, configurable products, simulated flow and live telemetry. Change the inputs, watch the system react and inspect the measurements.

INPUT / STATE / RESPONSE / MEASURELAB-04
04LIVE SYSTEMS
00DATA SUBMISSIONS
Calculations stay in this browser Every control exposes a response Released 01 Aug 2026

Systems you can touch

From explanation
to direct experience.

Interactive systems make relationships visible. These compact laboratories turn graphics, configuration logic, simulation and observability into controls, states and measurements you can inspect directly.

Four live systems

Change one variable.
Read the whole system.

Each lab combines a responsive visual model with a compact control surface and an explicit measurement layer.

Local runtime

The browser is
the instrument.

Rendering and calculations run locally. Device capability labels are read from this browser and are not transmitted.

Rendering path
Canvas 2D
WebGL 2
Checking
WebGPU
Checking
Logical processors
Checking
Pixel ratio
Checking
Data sent
None

Interactive workbench

Control the state.
Inspect the response.

Choose a laboratory, adjust its controls and compare the visual state with the measured outputs. Reset restores the documented baseline.

01 / MATERIAL RESPONSE

Examine light across a surface.

A compact lighting model separates diffuse and specular response. It is an explanatory instrument, not a physically complete renderer.

COATED ALLOY / STUDIO LIGHTFRAME 001
Surface
Light

Surface reading

Balanced highlight
Diffuse share
41%
Specular share
59%
Highlight spread
32%
Peak luminance
1.10
What to watch

Roughness broadens and softens the highlight. Metalness moves more of the base colour into the specular response.

Model boundary

The demonstration uses a compact analytical approximation for legibility and speed. It omits image-based lighting, geometry occlusion and spectral effects.

Implementation signal

A production material system should be verified with the target renderer, tone mapping, assets, displays and performance budget.

Technical basis: Khronos glTF material model ↗ MDN Canvas API ↗

02 / PRODUCT CONFIGURATION

Turn options into a valid product.

A product configurator is a rules system with a visual surface. Every selection must remain explainable, compatible and stable enough to quote.

MODULAR CONTROL ARRAYMC-S2-CN
Assembly
Systems

Configured system

$24,800 Valid standard assembly
Configuration code
MC-S2-CN
Bill of materials
12 items
Functional modules
3
Option delta
+$6,800
What to watch

Changing frame capacity can constrain module count. System packs affect both component count and the stable configuration code.

Model boundary

Prices are illustrative and do not represent a commercial offer. Real configuration logic would include stock, regional, engineering and approval rules.

Implementation signal

Persist option identifiers and rule versions, not visual labels alone, so quotations and downstream manufacturing remain reproducible.

Technical basis: Khronos glTF for product delivery ↗ JSON Schema specification ↗

03 / FLOW SIMULATION

See a field through motion.

Particles trace a deterministic vector field. The visual pattern changes with field strength, turbulence, population and integration speed.

SEEDED VECTOR FIELD 24017RUNNING
Field

Simulation state

Stable flow
Active particles
640
Motion energy
1.10
Field coherence
62%
Deterministic seed
24017
What to watch

Higher turbulence reduces coherent paths. Longer trail persistence reveals stable structures but can conceal rapid state changes.

Model boundary

This is an illustrative two-dimensional field, not computational fluid dynamics and not a claim about a physical system.

Implementation signal

Deterministic seeds make visual behavior reproducible for testing, review and recordings across the same implementation.

Technical basis: WHATWG canvas specification ↗ MDN animation timing ↗

04 / SYSTEMS TELEMETRY

Read the network as one system.

Nodes, links and signals form a live service map. The same graph produces an operational summary with latency, throughput and anomaly counts.

LIVE SERVICE FABRICOBSERVING
Topology

Operational reading

Nominal fabric
Active links
Relative throughput
1,200 req/s
P95 latency
66 ms
Flagged nodes
0
Node status summary
  1. GatewayNominal
  2. ControlNominal
What to watch

More links can improve alternate paths while increasing coordination. Lower anomaly thresholds surface more nodes for investigation.

Model boundary

Metrics are generated from the local demonstration state. They are not production telemetry and no external services are queried.

Implementation signal

Operational maps should connect topology, service ownership and measurable signals without hiding the underlying traces, metrics and logs.

Technical basis: OpenTelemetry specification ↗ W3C High Resolution Time ↗

Laboratory method

Make behavior
visible and testable.

Each demonstration follows the same contract: controlled input, explicit state, observable response, readable measurement and a clear boundary around what the model does not represent.

01

Deterministic core

Identical inputs produce identical calculated state, including seeded simulation and network layouts.

02

Visible controls

Every meaningful variable appears beside the visual result with its current value and practical range.

03

Measured response

Primary and secondary readings translate visual change into a reviewable engineering signal.

04

Declared limits

Each lab states what it demonstrates and what would require a production renderer, solver or telemetry platform.

From prototype to production

Six checks before
the demonstration becomes a system.

01

Is the model representative?

Validate geometry, rules, datasets and loads against the actual operating environment.

02

Can the state be reproduced?

Keep stable identifiers, configuration versions, seeds and inputs for review and testing.

03

Does interaction remain responsive?

Profile input latency, frame time, memory and main-thread work on target devices.

04

Is the result understandable?

Pair visual change with labels and measurements that do not depend on colour alone.

05

What happens without acceleration?

Define a dependable fallback for reduced capability, reduced motion and constrained hardware.

06

How will reality close the loop?

Name the production test, telemetry signal or user outcome that will verify the model.

Interactive systems engineering

Prototype the behavior.
Engineer the real system.

ADOR.IS builds product configurators, advanced graphics, technical visualisations and operational systems around measurable performance and clear system behavior.

[email protected]