Navigating the plot

Five rows of icon buttons, grouped by what each one acts on. A button that is on is highlighted.

The legend buttons are covered in the Legend tab and the X-axis choice in Time base. The line/dot switch drives the size slider in Settings, which is where the width itself is set.

Everything here is remembered across sessions except the plot cursor, which starts on each time. The plot applies the remembered choices as it is created, so they survive a reload whether or not you ever open the View panel.

Auto-follow

With auto-follow on, the X axis scrolls to keep the latest samples visible, the normal mode during live acquisition.

A zoom or pan that carries the newest sample off the window horizontally switches it off automatically. A tall spike leaving the window vertically does not: that is a property of the signal, and Y auto-scale pulls it back on its own.

Re-enable it with the auto-follow button, the Home key, or Ctrl+A.

View all data

Frames every buffered sample of every trace, which is how you see a whole session after zooming into part of it. Unlike Reset view, it leaves auto-follow alone: the look-ahead gap keeps the newest sample inside the framed window.

Pausing

Pause (⏸ or Ctrl+P) freezes the plot while data keeps streaming in the background, which is how you read a moment that just went by without losing what comes next. Press it again to resume.

The legend is an overlay on the plot naming the traces it draws, in the same tree shape as the Active panel: folder rows for the parameter, leaf rows for the channels, each with its colour. It is on by default and is switched from the Legend row of the View panel.

Selected traces are shown in bold, which is the other half of the emphasis that draws them thicker on the plot.

Working from the legend

Collapsed folders and the position you drag it to are both remembered. A press counts as a collapse only while the pointer stays within about four pixels; anything further is a drag.

The right-click menu

A legend row offers the same actions as the corresponding row in Active: Hide from plot, Only show, Remove from plot, Lock / Unlock, Copy full path and Copy value, applied to every signal beneath a folder row. It adds one item of its own, Add background / Remove background, which paints a solid card behind the legend for a plot busy enough to lose it among the traces.

Right-clicking empty space in the legend, rather than a row, reaches the plot’s own menu instead.

What it leaves out

With show hidden off, a hidden trace loses its legend entry; with it on, the entry stays and the row is dimmed. A legend with no rows to draw hides itself rather than leaving an empty card on the plot.

The cursor follows the pointer across the plot, marks where it crosses every visible trace, and labels each crossing with that trace’s Y value in the trace’s own colour. The X position is shown at the bottom of the cursor line.

That makes it the fastest way to read a set of signals at one instant: no clicking, no measurement to place.

The live cursor

A second cursor, pinned to the most recent sample of every trace rather than to the pointer, with the same value labels. It is drawn whether or not the pointer is over the plot, so the current values stay readable while you work in another panel.

It is suppressed in XY mode, where “most recent” has no position along the X axis, and switching the main cursor off hides it too.

For deltas between two points rather than values at one, use Measuring.

The X axis row of the View panel picks what the horizontal axis counts: a stopwatch for seconds since the plot started, which is the default, or a calendar for the controller’s own wall clock. In wall-clock mode ticks read 06:25:10, and the date for the visible window sits at the right-hand end of the axis where the (s) unit otherwise goes.

Everything that names a position along X follows the switch, so nothing on screen can contradict the axis: the grid lines, the cursor’s X readout, and the P1 / P2 values on a measurement card, which become full 2026-08-25 06:25:01.162 UTC stamps. ΔX stays a duration in both modes, because 2.5 s is 2.5 s either way.

While wall-clock mode is showing, a CSV export gains a matching UTC Time (ISO 8601) column.

UTC, and only UTC

The protocol carries epoch microseconds and no timezone, so the controller’s local zone is not knowable from the browser. Desk names the zone rather than guessing it, which is also how the controller’s own logs are written.

Where it does not apply

Wall-clock time needs an absolute timeline, and three situations have none: before any data has streamed, in XY mode, where X is a signal’s value rather than a time, and during a replay, whose X column is an offset from an origin the file does not record.

In each of those the axis stays in elapsed seconds. The choice is not reverted, so it takes effect by itself once live data arrives.

Absolute time is a label

Trace X values stay seconds-since-anchor whichever mode is selected; only the tick labels are shifted by the anchor’s epoch. That is what keeps deep zoom usable, and it is why switching the time base never moves or re-scales a trace.

Clear throws away buffered history, so anything not yet exported is gone. It is for starting a fresh look at a signal, not for tidying the view. Reset view to default does that without losing data.

Right-clicking a row of the legend opens the legend’s menu instead. Right-clicking empty legend space reaches this one.

Resetting the zoom

Three routes, and they differ:

Saving the plot as an image

Save as image… opens a dialog for the format (PNG, JPEG or WebP), the resolution (1×, 2× or 3× the size on screen) and whether the background is transparent. JPEG carries no alpha, so transparency is disabled for it.

There is no file-name field: the browser’s own Save dialog opens next, with a timestamped name such as plot-2026-09-03_14-22-01 already suggested, and that is where the name and folder are chosen. The image is put on the clipboard as well, as PNG whatever format was written, so it can be pasted straight into an issue or a document.

The capture includes the traces, the axes, the grid and the legend as real text. Hidden traces are left out of the legend in the image, since they are not drawn in the plot either.

More plot room

Click the ⤢ button in the Plot tab, or press Ctrl+M, to maximize the plot over the whole workspace, and again to restore the layout. Each Y axis also carries a small chevron at its top that shrinks that axis to a sliver and hands its width to the plot, which is worth knowing once three or four axes are stacked beside a narrow panel. The traces on a collapsed axis keep drawing and keep auto-scaling; only its ticks and labels go.

Panels can also be popped out or closed entirely. See Workspace.