Tyto Sensory Labs

Designed to scale

Haptics should not need to be rebuilt every time the game changes.

Different simulators expose different data. Different hardware expresses sensation differently. The physical meaning in between does not need to start from zero every time.

MCP4SH is built as a semantic telemetry to haptics layer. It stays specialised where the source or hardware requires it and stays consistent where the physics means the same thing. The current product is delivered through SimHub. The interpretation and orchestration logic is deliberately separated from one renderer or one device class.

Preserve meaning. Adapt the carrier.

01 · The architecture

Different sims. Same physical world.

One simulator may expose a useful signal directly. Another may describe the same vehicle state through a different set of telemetry. MCP4SH does not require those sources to look identical.

The goal is to understand each source at the input, translate it into a consistent description of what the vehicle is doing, coordinate that information, then express it through the available hardware.

Specialised inputs
Game telemetry
Host telemetry
Native SDK or API feeds
Domain-specific adapters
Shared semantic core
Telemetry conditioning
Physical interpretation
State and relationship modelling
Orchestration and prioritisation
Specialised outputs
Bass transducers
Haptic pads
Pedal motors
Wheels, seats and future active devices
Normalize meaning, not information. Richer telemetry should stay richer. What becomes consistent is the meaning available downstream.

02 · Fidelity without fragmentation

Specialise at the edges. Standardise the meaning.

A simulator specific input path can remain as specialised as necessary. That does not mean every downstream effect and every output device must also become simulator specific.

Use the best source available

Native signals, related signals and graceful fallbacks can all contribute without pretending they have identical fidelity.

Preserve useful spatial detail

When wheel level or directional information exists, it can remain spatial rather than being flattened into one chassis value.

Degrade honestly

When a title exposes less, MCP4SH should communicate less precision rather than inventing detail that is not there.

Concrete software example · braking and grip state

Different sources · preserved meaning
Assetto Corsa
Le Mans Ultimate
F1 25
AC Rally
AC EVO
Automobilista 2
Braking stateShared concept, source quality remains visible
Grip stateNormalized meaning without pretending every source is equally rich
Wheel / axle contextRetained where the title exposes it cleanly
Confidence / fallbackGraceful degradation instead of invented precision
Teach MCP4SH how a simulator describes the state. Do not rebuild the whole haptic system for every simulator.

03 · Orchestration

The vehicle is one system. The haptics should behave like one too.

Braking changes load. Load changes available grip. Suspension movement affects chassis behaviour. Engine, drivetrain, road, impact and tyre information all coexist. Making every lane louder at once does not necessarily communicate more.

MCP4SH treats effects as parts of one tactile scene. Interpretation determines what the signals mean. Orchestration decides how useful information should coexist when several things happen at the same time.

Concrete driving example · one Spa lap contains a whole tactile sequence

State changes over time
La SourceHeavy braking and rotation

Brake feel, forward load, tyre state and exit traction all overlap.

Eau Rouge / RaidillonCompression and crest

Heave and chassis load become the dominant information.

PouhonSustained lateral load

Long-duration chassis and tyre state matter more than impacts.

Bus StopBraking, kerb and transition

Several useful cues arrive together and need space rather than simple stacking.

Interpret

Work out what the available telemetry says about the current physical state.

Prioritise

Keep useful transient or state information from being buried by continuous activity.

Express

Render the resulting meaning through the capabilities of the target hardware.

More output is not automatically more information.

04 · Setup and tuning

Tune the rig. Not every game.

Transducers, mounting, useful frequency range, amplifiers, placement and personal preference still matter. No software removes the physical differences between rigs.

What MCP4SH is designed to remove is unnecessary repetition. Once hardware is mapped and calibrated, changing simulator should not mean rebuilding tyre, brake, suspension, drivetrain, road and load behaviour from zero.

What should remain user specific

Hardware placement Useful frequency range Gain and comfort Personal emphasis

What the system should solve first

Telemetry conditioning Physical interpretation Effect relationships Cross title baseline behaviour

Concrete setup example · calibrate the rig, then keep the baseline

One hardware pass · many titles
01 Map hardware

Assign the physical transducers and channels once.

02 Sweep response

Find the useful tactile range of the actual rig.

03 Set balance

Choose gain, comfort and personal emphasis.

04 Keep the baseline

The hardware setup remains stable when the title changes.

ONE CALIBRATED RIG BASELINE
Assetto Corsa Le Mans Ultimate F1 AC Rally AC EVO Automobilista 2
The user should mainly tune hardware and preference. The user should not have to compensate for an unfinished interpretation layer.

05 · Latency

Fast matters. Latency is not the whole haptic system.

Lower transport latency is desirable. Discrete events such as impacts, kerbs and sudden interventions benefit directly from the fastest practical path.

Useful tactile response also depends on what was detected, when the underlying state became meaningful, how it was interpreted, what else was happening at the same time and whether the output communicates the correct thing.

Telemetry state
Semantic onset
Output submitted
Physical response
State can become useful before a later threshold event.Transport still matters between semantic onset and physical response.
Fast reaction is useful. Earlier understanding can be better.

This does not make transport latency irrelevant. MCP4SH does not claim to predict information that the simulator has not exposed. It means an evolving physical state can carry useful information before a later event threshold is crossed.

06 · Scalability

Same meaning. Different edges.

A simulator and a haptic device do not need to understand each other directly. The simulator needs an input path that can describe useful physical state. The output device needs a projection path that can express that state. The shared meaning sits between them.

Scaling problem · bespoke links grow in both directions

N inputs × M outputs
Input \ Output
Shakers
Haptic pad
Pedal motor
Future device
Assetto Corsa
Le Mans Ultimate
F1 25
Flight input
Specialised input adapters preserve source-specific richness.
Shared semantic interpretation + orchestration
Specialised renderers adapt the same meaning to device capability.
Add an input once. Add an output once. Connect through meaning.

Real integrations still have unique capabilities and constraints. The goal is to keep those differences at the edges instead of multiplying bespoke interpretation throughout the whole stack.

07 · Mixed hardware

Different devices can carry different parts of the same vehicle state.

A mixed rig does not need every device to reproduce every effect. Each output can concentrate on the information it is physically suited to express.

Concrete rig example · divide the sensory work

Different carriers · one interpreted vehicle state
Semantic cue Pedal plate Seat / chassis Backrest / pad Pedal motor
Brake state
ABS / lock
Road / impacts
Rear traction
Drivetrain
Chassis load / direction
Primary carrier Useful secondary carrier
More devices do not have to mean more duplicated logic. They can mean a better division of sensory work.

08 · Built for integration

The current SimHub plugin is an implementation. It is not the intended boundary.

Today SimHub provides the public host for telemetry access, configuration and tactile output. MCP4SH.Core already separates most haptic interpretation and stateful effect logic from the host specific plugin layer.

That boundary is deliberate. The longer term direction is to let the same semantic and orchestration logic feed different hosts, SDKs and physical outputs without rewriting the interpretation model around each target.

Concrete OEM example · a new active pedal

Focus engineering on the device
Title / telemetry adapter
Conditioning and source interpretation
Shared braking · traction · grip · load semantics
Device renderer / capability adapter
OEM hardware tuning and physical expression
Build the hardware. Do not rebuild the physics interpretation stack.

09 · Output paths already exercised

The architecture is already reaching beyond one shaker topology.

MCP4SH is primarily developed and validated through tactile hardware driven by SimHub. Real user testing has also exercised other output classes. Compatibility and first class Setup Assistant integration are not the same thing, so the status is shown plainly below.

Current output-path evidence

Compatibility is not the same as first-class setup support
Output classExampleStatusSetup Assistant
Bass transducerDayton BST-1 / TT25 classReference pathFirst-class
Haptic padNext Level Racing HF8 / HF8 ProPublic user reportVia SimHub
Haptic padSRS U-ShakeReported compatibleVia SimHub
Motor hapticsPedal motors / ShakeIt MotorsDocumented routeManual mapping
Multichannel audioCreative X4 8-channelUser validatedSupported

A device being reachable through SimHub does not automatically mean every MCP4SH Setup Assistant feature supports it directly. The distinction matters because architectural compatibility can arrive before guided setup and profile generation.

10 · Proof, not promises

Measure the experience. Do not rely on the number of effects.

The architecture should be testable. MCP4SH is preparing measurements that can show where time is spent, how consistently the system behaves and how much repeated setup is actually required.

Time to first drive Install → mapped rig → usable baseline Measurement planned
Cross title retuning Same rig · same calibration · multiple sims Measurement planned
End to end latency Telemetry → processing → output → response Instrumentation under validation
Semantic onset When useful state information becomes available Capture method under validation

Future published results should distinguish telemetry cadence, MCP4SH processing time, output path delay and physical response where the measurement setup allows it. The method should travel with the number.

11 · Domain portability

A useful abstraction should survive beyond the first domain it was built around.

MCP4SH began with automotive simulation. The v1.2 development track is testing how much of the same architecture can carry into flight. Shared physical ideas such as powerplant behaviour, body load and transients can remain reusable where they genuinely overlap. Airframe specific meaning can be added where the domain is different.

The goal is not to rename automotive effects as aircraft effects. The goal is to reuse physical meaning where it transfers and introduce new domain semantics where it does not.

Automotive semanticsTyre grip · slip · braking · traction · suspension · road
Flight semanticsAirframe load · AoA · aerodynamic state · powerplant · landing / ground interaction
Shared physical meaning
load · acceleration · vibration · transients · powerplant behaviour
New domain. Same language where the physics overlaps.

Design principle

Specialise at the edges. Standardise the meaning.

Different simulators deserve access to their unique telemetry. Different hardware deserves to use its unique capabilities. Neither requires rebuilding the whole interpretation stack between them.

That is the problem MCP4SH is designed to solve.