A new optional signal filter that is switched on individually for every timeframe. In the standard logic, every bar whose histogram value exceeds the previous extreme of the current zone produces a signal — including the repeated breakouts that follow a pullback inside the same wave. With New Max/Min Series enabled, a timeframe signals only during the first uninterrupted series of new maximums (above zero) or new minimums (below zero). The first bar that fails to set a new extreme locks the zone, and no further signals are produced until the histogram crosses zero.

Eight independent bool switches, one per timeframe. All of them are false by default, so existing setups behave exactly as before. A timeframe with the switch off keeps the standard logic, which means series-mode and standard-mode timeframes can be combined freely in one multi-timeframe configuration.

The Signal_Mode_Settings block previously offered a choice between arrows for the current (unclosed) candle or arrows for the closed candle, where the signal of the closed candle is carried over to the next candle. It now has a third option — SIGNAL_MODE_BOTH — which runs both engines at the same time: the chart shows Current-candle arrows and Closed-candle arrows together, and the timeframe table shows two Signal columns side by side.

The timeframe table shows all eight timeframes, but in the Strategy Tester loading the history of every one of them at the very first tick can freeze the start of a test. With lazy loading the timeframes you did not tick in Use M1…W1 are not touched at start: their table cells show Wait and they are calculated later, one timeframe at a time, from the lightest (W1) to the heaviest (M1). The table itself is also created only after a short, configurable delay. On a live chart nothing changes. A new step-by-step debug log helps to find exactly where a slow or frozen start stops.

With EnableDebugLog = true every important stage is appended to DeepFaust_debug.txt in the terminal’s Common\Files folder. The file is opened and closed on each line, so even if the terminal hangs, the last line shows where it stopped. In the Strategy Tester the file is recreated at the start of every test; each line carries the local time and the milliseconds elapsed since OnInit. If the file cannot be opened, the reason is printed once in the Experts journal.

14:02:11 +0ms    | OnInit START tester=1 sym=EURUSD period=15 ShowTable=1 MaxBars=2000
14:02:11 +412ms  | OnInit: PrecalcBase BEGIN M15
14:02:12 +903ms  | OnInit: PrecalcBase END M15
14:02:12 +921ms  | OnCalculate #1 rates_total=4310 prev=0
14:02:12 +925ms  | GATE: start counting, rates_total=4310
14:02:14 +2931ms | GATE: OPEN (bars passed=7)
14:02:14 +2950ms | DEFERRED table: CreateCompactTimeframeTable BEGIN
14:02:15 +3377ms | LAZY BEGIN W1
14:02:15 +3390ms | LAZY W1: PrecalcBase BEGIN

Format example — timings and values are illustrative.

A new input that adds an adjustable delay to every tick processed by the indicator in the MetaTrader 4 Strategy Tester. Even at the lowest position of the tester’s speed slider a visual test can run too fast to follow how the table, the arrows and the multi-timeframe signals form; SlowdownTester lets you slow it down to any pace — from a single microsecond up to one hour per tick. On a live chart the setting is completely ignored.