step — Compute or plot step responses of SISO transfer-function models with MATLAB-compatible step forms.

step(sys, ...) plots or returns the sampled unit-step response of a supported SISO transfer-function model. With no time argument RunMat chooses a suitable grid; a scalar sets the final time and a vector supplies exact sample times. Typed-integer time values are available in RunMat mode and must be exactly representable in binary64 before entering the floating control-system simulation.

Syntax

y = step(sys)
y = step(sys, tFinal)
y = step(sys, t)
[y,t] = step(sys)
[y,t] = step(sys, tFinal)
[y,t] = step(sys, t)

Inputs

Returns

Returned values from step depend on how many outputs the caller requests.

Errors

How step works

• Supports SISO tf objects with real coefficients.
• Returns the response y as a column vector for single-output calls.
• Returns [y, tOut] for two-output calls.
• Plots the response for no-output calls.
• No-output calls accept common multi-system plot syntax such as step(sys1, 'b', sys2, 'r--', tFinal); style tokens are accepted for compatibility.
• Chooses an automatic time vector when no time input is supplied.
• Accepts a positive finite scalar final time.
• Accepts a finite, nonnegative, nondecreasing explicit time vector.
• Discrete-time models use the stored Ts sample time.
• Native typed-integer t and tFinal are RunMat extensions. Compatibility admission and exact-representability checks happen before provider access, and wide integers that would round as double are rejected.
• Response values and returned time coordinates are double because transfer-function simulation is an explicit floating computation domain; integer time input does not imply an integer output class.

GPU memory and residency

step computes on the host from tf metadata and returns host tensors.

Examples

Returning response data for a first-order system

sys = tf(1, [1 1]);
[y, t] = step(sys);

Expected output:

y and t are column vectors.

Using a scalar final time

sys = tf(1, [1 1]);
[y, t] = step(sys, 5);

Expected output:

t(end) = 5

Using an explicit time vector

sys = tf(1, [1 1]);
t = 0:0.5:2;
y = step(sys, t);

Expected output:

y is sampled at the values in t.

Plotting the step response

sys = tf(1, [1 1]);
step(sys);

Expected output:

A step-response plot is rendered.

Plotting multiple systems

s = tf('s');
H1 = 1/(s + 1);
H2 = 2/(s + 2);
step(H1, 'b', H2, 'r--', 5);

Expected output:

Both responses are plotted over the same final time.

How RunMat validates step

step validates object class, coefficient type, properness, sample time, and time-vector inputs before simulation. Unit and integration tests cover first-order response values, scalar final time, explicit vectors, discrete sample-time behavior, multi-output ordering, multi-system statement-form plotting syntax, and VM dispatch.

• Implementation: `crates/runmat-runtime/src/builtins/control/step.rs`

See Correctness & Trust for the full methodology and coverage table.

Using step with coding agents

Open a RunMat example with live inputs, then ask the agent to explain how step changes the result.

FAQ

Related Control functions

Open-source implementation

Unlike proprietary runtimes, every RunMat function is open-source. Read exactly how step is executed, line by line, in Rust.

• View the source for step in Rust on GitHub
• Learn how the RunMat runtime works
• Found a bug? Open an issue with a minimal reproduction.

About RunMat

RunMat is an open-source runtime that executes MATLAB-syntax code blazing on any GPU. It is licensed under the Apache 2.0 license.

• RunMat automatically optimizes your math for GPU execution on Apple, Nvidia, and AMD hardware. No code changes needed. Simulations that took hours now take minutes.
• Start running code in seconds. RunMat runs in the browser, on the desktop, or from the CLI. No license server, no IT ticket.

Getting started · Benchmarks · Pricing