Unexpectedly Long Daemon.g Execution Time

[curieos]

I have a daemon.g file meant to turn off the heaters after various idle intervals. I noticed the timer wasn't incrementing at the rate I expected, so I added a calculation to determine the elapsed time after the loop. I'm consistently getting elapsed time values in the 9-10 second range. I don't think this is expected?

Running RRF 3.4.6 on a 6XD in SBC Mode with a Pi 4.

The daemon file requires two global variables to be declared, bed_heater_timeout and tool_heater_timeout.

var daemon_sleep = 2

if !exists(global.heater_timer)
	global heater_timer = 0

while true
	if (state.status == "idle" || state.status == "simulating")
		var initial_time = state.upTime
		var heaters_active = false
		while iterations < #heat.bedHeaters
			if heat.bedHeaters[iterations] !=-1
				if heat.heaters[iterations].state == "active" || heat.heaters[iterations].state == "standby"
					set var.heaters_active = true
					if global.heater_timer >= global.bed_heater_timeout
						M140 S-273.1
		
		while iterations < #tools
			var current_tool = iterations
			while true
				if iterations >= #tools[var.current_tool].heaters
					break
				
				if heat.heaters[tools[var.current_tool].heaters[iterations]].state == "active" || heat.heaters[tools[var.current_tool].heaters[iterations]].state == "standby"
					set var.heaters_active = true
					if global.heater_timer >= global.tool_heater_timeout
						M568 P{var.current_tool} A0
				
		if var.heaters_active
			var elapsed_time = state.upTime - var.initial_time
			echo "Elapsed time: " ^ var.elapsed_time
			if var.elapsed_time < 0
				set var.elapsed_time = 0
			
			set global.heater_timer = global.heater_timer + var.daemon_sleep + var.elapsed_time
		else
			set global.heater_timer = 0
	else
		set global.heater_timer = 0
	
	G4 S{var.daemon_sleep}

edit: The machine has 1 bed heater and 4 tools. It's an IDEX printer, so T0 and T1 use heaters 1 and 2, respectively. T2 and T3 are duplicator and mirror mode, and they both use heaters 1 and 2.

I tested the same daemon.g file on a single tool, single bed heater machine running in standalone mode on a Duet 2, and the elapsed time is consistently less than 1 second.

[droftarts]

@curieos Sorry for the delay in replying. One for @chrishamm to have a look at, I think.

Ian

[chrishamm]

@curieos Conditional G-code is known to be relatively slow in software v3.4. I suggest you upgrade to v3.5 ASAP, because that should be significantly faster.