Answers below are expanded in full rather than collapsed, so that the text can be read, searched and printed. Each entry states the setting that resolves the question and, where relevant, the companion page that documents it in depth.
Which aggregation family should a footprint chart use?
Tick counts are the usual answer when density must be comparable across instruments, because the number of price rows per bar stays roughly stable. Clock intervals remain useful when timestamps must line up with a session calendar. Volume and range bars sit between the two: they adapt to activity but produce bars whose time span varies. The choice is recorded on the footprint chart configuration page.
Why do two panels show different cluster numbers for the same bar?
Because the panels disagree about one of three parameters: the aggregation family, the cluster mode, or the session template. Feed tick depth is a fourth, rarer cause. Compare the bar rule first, then the cluster mode, then the template; in that order the difference is almost always found before any indicator is examined.
How is the value area percentage chosen?
Seventy per cent is the conventional default and a reasonable starting point. Changing it changes the edges of the area, not the point of control, and a five-point change can move an edge by several rows on a compact profile. Keep the value constant across a comparative study, and state it whenever a chart is shared. See cluster chart and TPO.
What does the diagonal imbalance ratio mean?
It is a multiplier comparing a row with its diagonal neighbour. A ratio of three means the row must exceed the neighbour threefold before a marker is drawn. Because it is a ratio rather than an absolute count, the same value transfers between instruments of different typical print size. Details are on the footprint chart configuration page.
Why does the depth ladder look different from a colleague's?
Almost always because of the aggregation step and the size reference. A step of one tick shows raw levels; a step of four sums them. A local size reference recolours the ladder when the view is scrolled, while a rolling reference stays stable. Compare the step first, then the reference window, then the confirm window used for pull and stack detection.
Is a large resting size on the ladder a commitment?
No. Resting size can be withdrawn, and the ladder reports only what the feed publishes at the moment of the update. Use the confirm window parameter to require that a change persists across several updates before it is annotated. The reasoning behind the parameter is set out on the Smart DOM depth settings page.
Why do the same instance inputs behave differently in two panels?
Because an aggregation override was set on one of them. An override lets an indicator compute on a different bar rule than the panel draws, which is legitimate but invisible in a screenshot. Clear the override to make the instance describe the bars actually on screen. See the indicator module reference.
How much recorded history is enough for replay?
Enough to cover the review cycle plus a margin. Trades are far cheaper to store than depth snapshots, so enable book recording only for sessions that will genuinely be studied through the ladder. Storage arithmetic is discussed on the replay and data feeds page.
Why does a replayed session not match the notes taken live?
The session template used during replay differs from the one used live, or the recorded tick depth is coarser than the live feed. Set the template first and compare one distinctive bar before drawing any conclusion about the rest of the session.
Which theme settings actually improve readability?
Axis label density and cluster text contrast. Reducing the number of axis labels releases the space that footprint numbers need, and raising text contrast against the darkest size band is the single change most readers notice. Palette swaps are cosmetic by comparison. See themes and rendering.
How many panels can one machine carry?
Count the visible bars times the rows per bar across footprint panels, add an allowance per indicator instance, and reserve a third of memory for the tick store and the operating system. A four-panel footprint desk is a substantially heavier load than four clock charts. The procedure is on the performance and hardware notes page.
Do these pages describe a platform for sale?
No. This is independent technical documentation of configuration behaviour, published for readers who already work with these interfaces. There is no software offered here, no service offered here, and nothing on this site collects personal data beyond ordinary server logs. See the disclaimer and the privacy policy.