RSI

RSI in TRScene is not only something you look at. The value drawn in the panel is the value the scenario engine compares against when it decides whether to enter a trade — the same number, from the same code.

Two things follow from that, and they are what this page is about: the number agrees with the RSI on your exchange's chart, and you can point a node at it.

Find it on the chart

RSI lives in its own panel below the candles, not on top of them. The row of tabs under the main chart switches which indicator that panel shows.

In Balanced the panel is deliberately short — enough to read the shape and the current value badge. Switch to Chart when you want room to inspect the curve itself.

The line does not start at the left edge of the chart. RSI needs period + 1 candles before it produces a value, so with the default period of 14 the first 14 bars stay empty.

Settings

Inputs

The RSI-based MA is a moving average of the RSI line, drawn in the same panel. It smooths the oscillator so a crossover between the two is easier to see than a crossover of RSI alone.

Style

Each band has its own switch, so you can keep a level on the chart without the fill, or drop the levels entirely and read the raw line.

Why the number matches your exchange

RSI has more than one accepted smoothing. TRScene pins it to Wilder's original — an RMA with alpha = 1 / period — because that is what exchange charts and TradingView use. The indicator library's own default is a faster EMA, which drifts away from those charts, so it is overridden explicitly rather than accepted.

The calculation runs in Rust, and the chart and the scenario runner call the same function. There is no second implementation for display, which is why the panel and the engine cannot disagree.

Use it in a scenario

A comparison indicator reaches a scenario through a condition row: target A, an operator, and target B. RSI is one target type among several, and the row behaves the same wherever it appears.

Open the condition editor

The properties panel shows a condition as one compact row. Click that row and the editor opens, with both targets laid out side by side and the operator between them. This is where the condition is actually built.

1. Set Target A's type to RSI. It sits under the Oscillator group.
2. Pick the Timeframe from the strip and set the RSI Period.
3. Choose the operator in the middle column. It restates itself in words — A crosses above B.
4. Set Target B (below).
5. Apply writes the condition back to the node.

What RSI can be compared against

Open Target B's type list and you get exactly three choices — the editor only offers what the comparison supports.

RSI MA hides its RSI period. The field you see is the MA period; the RSI underneath is fixed at 14. Comparing an RSI(7) target A against RSI MA therefore measures it against the moving average of RSI(14), not of your own series.

Operators

Four, and no more:

> and == are absent on purpose. An oscillator can pass through an exact value between two candle updates, so a condition that depends on landing on it — or on being strictly past it at the moment of evaluation — can miss a move it should have caught.

The distinction that matters when you wire a node is state versus event. A state operator holds true for as long as the comparison holds, so the branch it guards keeps passing. An event operator is true only on the update where the crossing happened, and false again on the next one.

Where the list lives

Two nodes expose the condition row, under different headings.

Under each row the panel restates the condition in plain language — RSI 14 (5m) breaks above 30. Read that line before you run anything; it is the fastest way to catch a condition that says something other than what you meant.

The Timing block on the Start node applies to the group, not to RSI: Moment requires every condition to be true on the same candle update, while Window allows them to become true within a period of each other.

The branching example reads RSI 14 (1h) remains above 70 — a state operator on a 1-hour series, chosen independently of the 5-minute target in the Start node above.

Chart settings are not condition settings

The two live apart. Changing one does not change the other.

• Timeframes are independent. A condition carries its own timeframe, 5m by default. You can watch a 1-minute chart while a scenario runs on 1-hour RSI.
• Identical series cannot be compared. Point RSI at another RSI with the same timeframe and period and the operator list comes back empty — the comparison would be a tautology. Change the timeframe or the period and it is allowed.
• A condition needs both a period and a timeframe to pass validation. A scenario missing either will not start.

Reference

Related

• Indicators
• Start Node
• Condition Node