Settings

This is the longest page in the pack. Read it in two passes if you need to. The first section teaches the shape of one slot — once you know that shape, the other nine slots are the same dialog repeated. The later sections cover the global controls, the master smoothing pass, and the per-slot power-user knobs that only matter for certain moving-average choices.

A note before you start: most of the inputs on this dialog are the same inputs you would expect on any RSI tool. The CTX trim is wider not because RSI got harder, but because you now own the choices that a single-timeframe RSI hides from you. Treat each knob as a decision you are making on purpose.

How the dialog is organized

The TradingView settings dialog for this indicator separates inputs into a few groups:

• General controls for the oscillator levels (overbought and oversold).

• Display controls for whether the blended pair plots and how thick the blended lines are.

• Master Smoothing controls that apply one extra smoothing pass to the blended pair after the slots are combined, including the master-specific tuning fields for MA families that use them.

• RSI 01 through RSI 10 — the per-slot configuration. Every one of these ten sections has the same shape.

• RSI 01 Power User through RSI 10 Power User — the per-slot power-user knobs. These only matter for some moving-average types; they sit out of the way when they are not relevant.

The trim ships with three slots enabled (slots 01, 02, 03) at 5m, 15m, and 60m, equal weights of roughly 33.3, the on-bar-close switch on for every slot, master smoothing off, overbought at 70, oversold at 30. The remaining slots are present but disabled.

The shape of one slot

This is the section to actually read. Slots 01 through 10 each carry the same set of inputs. Once you understand them on slot 01, you do not need to read them again for slots 02 through 10.

Enable RSI NN

Boolean. Defaults: on for slots 01–03, off for slots 04–10.

When this is off, the slot does nothing. It does not compute, does not plot, does not contribute to the blend, and cannot fire alerts. Disabling a slot is the cleanest way to remove it from the picture entirely.

When this is on, the slot computes whether or not it plots and whether or not it has weight in the blend. That distinction matters because it lets you keep a slot quietly running for alerts only — see Hide RSI NN Plot and Blended Weight below.

Hide RSI NN Plot

Boolean, default off.

When on, the slot still computes and still participates in the blend (if its weight is non-zero), but its line does not draw. Useful when the pane is starting to feel crowded and you want to keep a slot in the math without keeping it in the visual.

This is not the same as setting weight to zero. A hidden slot with weight 33 still steers the blend. A weight-zero slot that is not hidden still draws.

Source

Series, default close.

The price series that feeds the slot's RSI calculation. If you set an optional ticker for this slot, the source is read from that other ticker rather than from the chart's symbol.

For most uses, leaving this on close is correct. Switch to hl2, hlc3, or ohlc4 if you want a slightly different read on intrabar bias and you have already done that work on a single-timeframe RSI elsewhere.

TimeFrame

String. Defaults: "5" for slot 01, "15" for slot 02, "60" for slot 03, empty for slots 04–10.

This is the higher timeframe the slot reads its RSI on. Empty means "use whatever the chart is on." Otherwise, set it to a value greater than or equal to the chart timeframe.

The constraint matters: a slot timeframe lower than the chart timeframe is rejected. The script raises a runtime error and names the offending slot in the message so you do not have to hunt. The reason is honest — TradingView cannot serve a higher-resolution series down into a lower-resolution chart bar without inventing values, and the script declines to invent.

If you want to study a slot at a finer resolution than your chart, lower your chart timeframe. If you want to study at a coarser resolution, raise the slot's timeframe.

See MTF and Repainting for what the higher-timeframe value actually means at any given moment in time.

RSI Length

Integer, minimum 1, default 14.

Standard RSI lookback. The classical setting is 14 and the trim ships with that. Shorter lengths produce a more reactive RSI; longer lengths produce a smoother one. There is no "best" — what works depends on the timeframe of the slot and the kind of move you are trying to read.

RSI Length and TimeFrame are two knobs that often get adjusted in place of each other. They are not equivalent. TimeFrame controls which bars the slot sees at all; RSI Length controls how many of those bars enter the momentum calculation. If you set slot 01 to a 60m timeframe and shorten its RSI Length to 7, you are reading a fast RSI on a slow series — the slot remains blind to everything that happens between 60m closes. If you want a faster read on intraday action, a lower-timeframe slot is the right lever, not a shorter length on a higher-timeframe slot.

Tradeoff to be honest about: shortening RSI length on a higher-timeframe slot does not give you back the missing information from the lower timeframes. It makes the slot more sensitive to the bars it does see. If you want lower-timeframe responsiveness, add a lower-timeframe slot.

RSI Smoothing

Integer, minimum 1, default 3.

Length used for the moving average applied to raw RSI. The slot's "RSI line" is this smoothed value, not the raw RSI.

A length of 1 effectively means "no smoothing" for the simplest MA types, which lets you see raw RSI behavior. Longer lengths produce a calmer line at the cost of latency.

RSI Type

Enum, default SMA.

The moving-average style used for that smoothing step. Choices come from the shared Axiom MA library — SMA, EMA, ALMA, KAMA, FRAMA, Jurik, Laguerre, VAMA, and the common variants the library exposes. Some types ignore extra tuning; others read from the per-slot power-user knobs further down the dialog.

Practical guidance:

• SMA and EMA are familiar and easy to reason about. Stay here while you are learning the pane.

• ALMA, Jurik, and Laguerre give you smoother lines that lag less under typical conditions, at the cost of having more tuning knobs and being harder to mentally model when something looks wrong.

• KAMA and FRAMA adapt their effective length to the data, which can help in regime changes but can also change behavior between sessions in ways that are not obvious from the dialog.

• VAMA uses the library's volatility-adjusted approximation; its Vol Length field is a lookback for that adjustment, not a direct read of volume. If the adjustment behaves oddly on a symbol, treat it as a smoothing-choice issue and verify it on that feed before trusting it.

If you choose a type that uses power-user knobs, the relevant knobs in the slot's power-user section start to matter. The other power-user knobs become inert.

Signal Length

Integer, minimum 1, default 3.

Length used for the moving average applied to the already-smoothed slot RSI, producing the slot's signal value. The per-slot signal is not drawn as a separate line; the slot color compares against it.

Default of 3 produces a fast signal. Longer signals produce a calmer line that flips less often and lags more.

Signal Type

Enum, default SMA.

Independent of RSI Type. You can run, say, an ALMA-smoothed RSI line against an SMA signal, or any other combination. They tune through their own power-user knobs.

There is no rule that the two MAs should match. Mixing them deliberately is part of what makes this trim a workbench.

Line Width

Integer, minimum 1, default 2.

Visual thickness of the slot plot. Affects readability only; does not affect the math.

Blended Weight

Float, minimum 0. Defaults: 33.3 for slots 01–03, 0.0 for slots 04–10.

This is one of the most consequential inputs in the whole dialog. It sets how much the slot steers the blended pair.

A weight of zero means the slot does not contribute to the blend at all. The slot still draws (unless hidden), still computes, and still fires its own alerts — but the blended RSI and blended signal are computed as if this slot did not exist. This is the right setting for a "context only" slot you want to watch but not let influence the composite.

A weight greater than zero adds the slot to the weighted average. Weights are normalized internally — what matters is the ratio of weights across the contributing slots, not the absolute number. Two slots at weight 50 produce the same blend as the same two slots at weight 1.

What you should not do: leave all ten slots on at equal weight and then read the blend as if it were a "consensus across ten timeframes." On a single symbol, your slots are heavily autocorrelated. Equal weighting just averages overlapping windows of the same price path. See Limitations and Trust Boundaries for the longer version of this caution.

Optional Ticker

Symbol, default empty.

When empty, the slot reads its source on the charted symbol. When set to another symbol, the slot reads its source on that other symbol instead.

This is the cross-asset switch. It enables genuinely useful workflows — an index slot under a single-name equity, a perpetual slot alongside a spot pair, a sector ETF under a sector member — and it carries a category-error risk if you forget you turned it on.

If any slot has an optional ticker set, the blended pair is no longer "an RSI of the chart symbol." It is a composite that includes outside markets. Whether that composite is meaningful depends on your weighting and on the correlation between the chart symbol and the foreign series.

See Workflows for the cross-asset workflow and the warning that goes with it. See MTF and Repainting for what happens when the foreign symbol's session is closed.

The shape of one slot's power-user section

Each slot also has a "Power User" section in the dialog. Most of these knobs are inert until you choose a moving-average type that reads them. Knowing which knobs go with which type saves you from tuning numbers that are not affecting anything.

On Bar Close?

Boolean, default on.

This is the per-slot repaint switch. It is shared between the slot's RSI line and its calculated signal value — they cannot disagree on this question, by design. The sharing is not a convenience choice; the RSI and its signal are meaningful only in relation to each other, and a slot where one side is confirmed while the other is live would produce a color state that did not describe a real moment. The script refuses to construct that kind of slot.

When on, the slot returns the previous confirmed higher-timeframe bar's values for both RSI and signal. The visible slot line and its color state are stable until that timeframe's next bar closes. The tradeoff is latency.

When off, the slot returns the live higher-timeframe bar's values. The plot can move during the higher-timeframe bar's formation. The tradeoff is the very repainting behavior most readers came to this trim to control.

MTF and Repainting covers what each side of this switch actually does and how to verify it on your own chart. The summary, here on the settings page: this is a tradeoff, not a fix. There is no setting that makes the live higher-timeframe value stop moving until its bar closes — only a setting that hides that movement by reporting the previous bar instead.

MA-specific knobs

The remaining knobs in the per-slot power-user section pair with specific MA types. They split into two halves — one half for the RSI smoothing MA, one half for the signal MA. The RSI half is prefixed RSI; the signal half is prefixed Signal. The two halves tune independently because the two MAs are independent.

• ALMA Floor Offset?, ALMA Offset, ALMA Sigma — used when the corresponding MA type is ALMA. Other types ignore them.

• KAMA/FRAMA Fast, KAMA/FRAMA Slow — used when the corresponding MA type is KAMA or FRAMA. Other types ignore them.

• Jurik Phase, Jurik Power — used when the corresponding MA type is Jurik. Other types ignore them.

• Laguerre Alpha — used when the corresponding MA type is Laguerre. Other types ignore them.

• VAMA Vol Length — used when the corresponding MA type is VAMA. Other types ignore them.

The pack does not publish library-side parameter detail beyond the names above. For practical tuning: start with the library defaults the dialog ships with, change one knob at a time, and watch the slot line for the effect. If a knob does not visibly change the line, the type you have selected is not reading it.

General controls

These apply to the whole indicator, not to any one slot.

Overbought Level

Float, default 70. Step 0.1.

Draws the upper dashed gray reference line. Read by the visual guide only — there is no alert condition tied to it. If you push the master-smoothing length up, your blended pair may stop reaching this line; that is information about your master smoothing, not about the symbol.

The 70/30 pair is a convention, not a law. The convention exists because RSI(14) on daily equity data historically crosses those levels on roughly the days most readers would call "stretched," which made 70/30 a useful shared reference in print. The levels do not mean "buy" or "sell" on a cross, they do not calibrate themselves to the instrument you are looking at, and they certainly do not map cleanly onto smoothed composites. If you are trading crypto on 5m, 70/30 will feel different than it does on a daily equity chart. Treat the guides as landmarks for where the pane sits, and adjust the thresholds deliberately when the instrument or the smoothing is telling you the defaults no longer mean what they used to.

Oversold Level

Float, default 30. Step 0.1.

Mirror of the overbought guide. Same caveats.

Plot Blended RSI/Signal

Boolean, default on.

Turns the blended pair's plotting on or off. The blend is still computed when this is off — it just is not drawn. Useful when you only want to read the slots and the blend would be a distraction.

Blended Line Width

Integer, minimum 1, default 3.

Visual thickness for the blended pair. Affects readability only.

Master Smoothing controls

Master smoothing is one extra smoothing pass applied to the blended pair after the slots are combined. It affects responsiveness and calm, and it can change what the overbought and oversold guides actually mean for you.

Enable Master Smoothing

Boolean, default off.

When off, the blended pair is the weighted average of the per-slot RSI values and signal values, clamped into 0..100. When on, that weighted average is then smoothed once more using the master MA.

Recommendation for the first sessions: leave this off. Build an expectation of how the raw blend behaves before you start smoothing on top of it. That expectation is what lets you notice when the smoothed blend is hiding something.

Master MA Type

Enum, default EMA.

The MA style used for the post-blend smoothing pass. Same library surface as the per-slot RSI and signal MAs.

Why the default is EMA rather than SMA: EMA is responsive enough to follow the blend without excessive lag, and it is widely understood. If you want a calmer master smoothing, choose ALMA or Laguerre and tune through the master power-user section.

Master Length

Integer, minimum 1, default 3.

Length of the master smoothing pass.

The honest tradeoff here is regime change. A long master length produces a calm blended pair you can read at a glance. The same long length can hold the blended pair on one side of the midline for several bars after the underlying slots have already flipped — the smoothing is doing exactly what you asked it to do, but what it is doing is hiding the moment that just happened.

If you push master smoothing past a length where the blend rarely reaches the overbought or oversold guides anymore, treat that as a sign that your guides are now lying to you about the smoothed series. Either widen the guides, lower the smoothing length, or read the slot plots as your real source for those moments.

Master tuning knobs

The master smoothing group includes its own ALMA, KAMA/FRAMA, Jurik, Laguerre, and VAMA tuning fields. They follow the same matching rule as the slot power-user knobs: a field matters only when the selected master MA type reads it.

Misuse to surface before the fact

Five settings interactions cause the most surprise in practice. Each is preventable.

• All ten slots on at equal weight on a single symbol. The blend is not a consensus across timeframes. It is an autocorrelated average. Either weight by intent or limit how many slots actually steer the composite.

• On Bar Close? off for every slot. This chases responsiveness and removes the very behavior the trim is designed to make explicit. If you want responsiveness, make the choice slot by slot; do not flip the whole rack.

• Optional ticker set on a slot, weighting unchanged. A high-weight cross-asset slot quietly turns the blend into a composite of two markets. If you load a foreign symbol, decide its weight on purpose.

• All weights zero. The blend goes NA. The blended pair stops plotting and alerts on the blend cannot fire. This is expected; if you do not want it, leave at least one weight non-zero.

• Long master smoothing during a regime change. The blend will keep reading the prior regime for several bars after the slots have flipped. Either lower the length or switch to reading the slot plots when you suspect a transition.

Where to go next

• For what the configured slots actually look like on the pane and how to read them, go to Visuals and Logic.

• For the on-bar-close switch, the cross-session wrinkle, and how to verify the switch on your chart, go to MTF and Repainting.

• For the settings interactions that turn into trust failures under pressure, go to Limitations and Trust Boundaries.

• For step-by-step setups for specific reading routines, go to Workflows.