How to do simple variable running costs in NML?

[NekoMaster]

So I thought I would try something while I modify the OPenGFX Trains GRF to provide some more 32BPP vehicles modified for my own use, and I wanted to run it with Pineapple rail set.

Pineapple rail set has variable running costs where things cost more to run then when they are stopped (like loading/idling)

I was wondering how I might do a simple variable running cost set up where trains cost less when stopped/idling and cost their normal amount when moving/operating.


I'm not very good at understanding new coding or coming up with new stuff and of course the below doesnt work (NMLC doesnt know what "if" means)

Code: Select all

        if vehicle_is_stopped = 1
		than
		running_cost_factor:			90;
		else
		running_cost_factor: 			180; // $  6,748 xxx

Perhaps someone could help me with that? Or is it even possible to just do a simple if else statement to have things cost less when stopped/idling.

[Tony Pixel]

You make all right, but there are wrong syntax, this is right:

Code: Select all

Running_cost_factor = (Vehicle_is_stopped = 1) ? 90 : 180;

Or You can use switch:

Code: Select all

switch (FEAT_TRAINS, SELF, sw_vehicle_name_running_cost_factor, vehicle_is_stopped) {
1: 90;
180;
}

Read this (https://newgrf-specs.tt-wiki.net/wiki/NML:If) or this (https://newgrf-specs.tt-wiki.net/wiki/NML:Switch)

[Supercheese]

I do it by checking the vehicle's speed: if the speed = 0, then reduce running costs to 25% of what they would be when in motion.

Code: Select all

switch(FEAT_ROADVEHS, SELF, running_cost_switch, current_speed) {

    /* Running cost reduced to 1/4 if vehicle is not moving. */
    
	0: 				VEHICLE_RUNNING_COST_FACTOR / 4;
	
	return 			VEHICLE_RUNNING_COST_FACTOR;
}

Where VEHICLE_RUNNING_COST_FACTOR is of course the desired full value.

[NekoMaster]

I got it working thanks to some help from my good friend GWYD

Just make a switch above your item code like this

Code: Select all

//	RUNNING COST SWITCH
//	===================================
	switch(FEAT_TRAINS, SELF, switch_running_costs, current_speed) {
	0:			12/2; //IDLING COST TARGET:	 $	1/2 COST
	return  	12/1; //RUNNING COST TARGET:	$	1,640
	}
	

And then in your graphics section, you can add this to the top for variable running costs. By full run costs I mean the cost without any division.

Code: Select all

	graphics {
		running_cost_factor:					switch_running_costs; // USES A SWITCH FOR REAL COST
		purchase_running_cost_factor:		12; // DISPLAYS RUN COST IN PURCHASE MENU, COPY FULL RUN COST HERE

[PikkaBird]

Well done getting it working.

I don't know if we're creating a "standard" here, but like Supercheese I use 25% for the reduced cost when stopped.

[Wahazar]

BTW, what is a computer performance costs of running costs switch? I mean, how often it is checked?

[Gwyd]

Well a running cost switch works the same way as any other switch... So there's one way to find out. Use the same switches to change graphics! I shall do some coding...

[EDIT] So I did some coding. My results for the frequency of updating are inconclusive. Graphics seem to be constantly updated. Stats such as power do not. Here the graphics show 0..9 based on the speed in mph (0=1..10, 1=11..12 etc). Power is set to 9125 for speeds 0..70, then to 1 for more than that. src attached. GPLv2.

Try putting the vehicle onto a stretch of track which alternates between 2 types of electrification (you'll need any other train set installed to allow this). Watch the power in "train details"

[EDIT 2] Some very odd things trigger the power switch to be checked. NewGRF changing? Yes. Stop/Start? Yes. Pausing? No. The actual speed of the vehicle? No. Find more!

[PikkaBird]

Watch the power in "train details"

Do note that the power being recalculated and the text in the train details being updated are not the same thing. The latter is not a desync risk and therefore does not have to be prioritised.

In any case, vehicle callbacks (and vehicle NewGRF in general) have never been shown to have an appreciable performance impact.

[Wahazar]

In any case, vehicle callbacks (and vehicle NewGRF in general) have never been shown to have an appreciable performance impact.

So my I make a conclusion, that running costs callback for each train vehicle (for example costs depending on vehicle load) would not affect computing performance?

I'm trying to improve balance between passenger/mail and freight cars.

[NekoMaster]

Well done getting it working.

I don't know if we're creating a "standard" here, but like Supercheese I use 25% for the reduced cost when stopped.

Do you mean 25% off the running cost (So you have 75% of the original cost) or bringing the running cost down at idle to 25% of the cost (So if the cost was $1000/yr, it would then be $250/yr at idle)

[PikkaBird]

Do you mean 25% off the running cost (So you have 75% of the original cost) or bringing the running cost down at idle to 25% of the cost (So if the cost was $1000/yr, it would then be $250/yr at idle)

The latter.