Integration
How every figure on this site is derived
What every number on this site is measured against, and where each derivation comes from.
In one sentence Every headline figure on this site is a range, measured across real implementations and tied to a baseline architecture and a vehicle segment. This page states that baseline and the basis for each figure.
The reference baseline
Unless a page says otherwise, “conventional” means a passenger car in the compact-to-mid-size segment with the following architecture.
| Aspect | Conventional baseline | ASG |
|---|---|---|
| E/E architecture | Point-to-point plus several buses and the gateways between them | One ring carrying power and data |
| Control units | 40 to 100+ ECUs, mostly function-specific | Universal zone modules; 11 in the worked example |
| Harness | 2–3.5 km of wiring per vehicle | 200–300 m per vehicle; 217 m in the worked example |
| Timing | Per-bus, validated separately for each function | One 1 ms commit, fixed for the whole vehicle |
The figures
| Figure | Value / range | What it compares | Basis |
|---|---|---|---|
| Wiring reduction | 90% · 7–17× depending on baseline architecture and vehicle segment | 200–300 m on ASG vs a 2–3.5 km conventional harness | Measured across test vehicles, pilots and simulator builds; vehicle-specific, so given as a range |
| ECU reduction | 5.7× | Universal zone modules vs a conventional function-specific ECU set | Vehicle-dependent: over 90% on ECU-heavy platforms, less on already-lean architectures |
| Saving per vehicle | $3,000–7,000 USD | ASG bill of materials and harness-fitting labour vs the conventional architecture | Modelled from copper, ECU count and cost, and install labour; widens with vehicle complexity and build location |
| Communication efficiency | 98% | Useful information as a share of what the channel carries (a communication figure, not an energy figure) | Bench and pilot testing, plus the protocol structure: one writer per value, no duplicate traffic |
| Transfer rate | 2.7 Mbps | Effective rate for a like-for-like comparison with a CAN-class bus | Raw channel rate weighted by the useful-information share; a comparison aid, not a selling point |
Anchored below so links from elsewhere on the site land on the right row.
Wiring reduction
Stated as 90%, or 7–17× depending on baseline architecture and vehicle segment. The comparison is 200–300 m of ASG wiring against a 2–3.5 km conventional harness for a vehicle of the same size; the compact-car worked example lands at 217 m.
The figure comes from actual implementations: test vehicles, pilot projects and simulator builds. Every vehicle is different, for the obvious reasons, so it is given as a range rather than one number.
ECU reduction
Stated as 5.7×. It compares a set of universal zone modules against a function-specific ECU set, and it depends heavily on the vehicle. On a platform carrying more than a hundred ECUs the count falls by over 90%; on a leaner, already-consolidated architecture the reduction is smaller. The worked example uses 11 modules.
Modern zonal architectures
On a recent zonal design built around high-performance controllers linked by Automotive Ethernet, a small number of large control blocks rather than a wide ECU set, the wiring and ECU differences narrow. There the distinction is not cable count. ASG acts on many moving parts in the same synchronous instant; an architecture of high-performance controllers on a switched network schedules and reconciles them across the network instead. That difference, and why it matters for coordinating multiple actuators, is set out on ASG vs Automotive Ethernet.
Saving per vehicle
Stated as $3,000–7,000 USD. It is a modelled figure: the cost of the copper removed, plus the cost and number of ECUs it replaces, plus the labour to fit the harness. It is not a logistics or lifetime figure.
Each layer of that estimate adds uncertainty, and the result is strongly vehicle-dependent. A low-volume vehicle with a large ECU count, assembled by hand in a high-wage plant, sits well above the range; a cost-optimised vehicle built on an automated line sits at or below its lower end. One currency is used throughout, and the saving compounds across a platform and a model cycle.
Communication efficiency
Stated as 98%. It is the share of what the channel carries that is useful vehicle state, rather than protocol overhead, addressing or retransmission. It is a communication figure, not a statement about vehicle energy, and it is a single value, not a range.
It comes from bench and pilot testing, and from the protocol itself: ASG has one writer per value and broadcasts it once, with no duplicate traffic. CAN and Ethernet instead move a continuous stream of data from which each node picks out the few frames that concern it.
Transfer rate
Stated as 2.7 Mbps. It is the raw channel rate weighted by the useful-information share above, expressed so it can be set beside a CAN-class bus. It is a comparison aid, not a selling point: ASG is a synchronous shared-state architecture, not a message bus, so a like-for-like bitrate understates the difference, in the way that judging a series circuit against a parallel one by a single number would.
Timing figures
The 1 ms commit period and the ~10 µs local block reaction are modelled values, defined on the real-time OS and hardware pages. They do not vary with vehicle or module count over the supported range.
EV range
ASG makes the whole vehicle state available in real time, once every millisecond. That vector is what recuperation and energy-allocation algorithms need to recover more braking energy and spend it better, so a meaningful range improvement is possible on the platform.
ASG’s own modelling of this is promising but not a proven, measured result, and it would vary with the model, the drive cycle and conditions. So this is stated as a capability of the platform, not as a figure: the site carries no percentage for range.
Questions
Where do these figures come from?
Real implementations: test vehicles, pilot projects, simulator builds and a worked integration example, together with the structure of the protocol itself. Because every one depends on the vehicle, they are given as ranges rather than single guaranteed numbers. The formal write-up of each derivation, with its baseline and boundaries, is being prepared with the founders.
Is the 98% an energy-efficiency figure?
No. It is communication-channel efficiency: the share of transmitted information that is useful vehicle state rather than protocol overhead. ASG carries no duplicate traffic, so little of the channel is wasted, unlike CAN or Ethernet where a node filters the few frames it needs out of a continuous stream. It says nothing about vehicle energy or range.
Is there a percentage figure for EV range?
No. ASG makes the real-time vehicle vector available that recuperation and energy-allocation algorithms need, which makes a range improvement possible on the platform. The site states this as a capability, not a percentage: any figure would depend entirely on the model, the drive cycle and conditions, and it is not yet a proven result.
Are these numbers the same for every vehicle?
No. Every figure is vehicle-dependent. An ECU-heavy premium car and a cost-optimised high-volume car sit at opposite ends of each range; a commercial vehicle, or a modern zonal architecture already built on high-performance controllers, lands elsewhere again.