[Traindude]

Although the following doesn't pertain to steam locomotives per se, I want to say one more thing about particle emitters in general that I hope can be rectified in the future.

Currently, ORTS (and in the past, MSTS) assumes that all particle emissions are emanating from an orifice that is perfectly round. However, this is not always the case. In particular, diesel locomotive exhaust stacks/ports are often oblong or rectangular in shape. Case in point:


This is the TrainArtisan BNSF SD40-2 that came with the MSTS version 1.2 update, as seen in ORTS. As you can see, even though the physical exhaust stack on the model is clearly rectangular when viewed from overhead, the emitters still assume that the stack is round, resulting in a very thin exhaust plume.

My solution is to specify size dimensions for the emitter on 2 or more axes. So, instead of this:

Effects
	(
		DieselSpecialEffects
		(
			comment
			(
				the format of these little blocks is as follow:

				offset.x offset.y offset.z
				normal.x normal.y normal.z
				width in metres

				defining the offset from the shape pivot point
				where steam / smoke should appear, the direction
				it should go and the width of the nozzle
			)

			Exhaust1
			(
                                0.0 4.738 1.485
				0 1 0
				0.1
			)
		)
      )

...the code would be this:

Effects
	(
		DieselSpecialEffects
		(
			comment
			(
				the format of these little blocks is as follow:

				offset.x offset.y offset.z
				normal.x normal.y normal.z
				size.x size.y size.z

				defining the offset from the shape pivot point
				where steam / smoke should appear, the direction
				it should go and the size of the nozzle on each
                                axis
			)

			Exhaust1
			(
                                0.0 4.738 1.485
				0 1 0
				0.74 0 0.1
			)
		)
      )

[ATSF3751]

I guess we can't say Open Rails and other developers haven't gotten anywhere in 22 years since the MSTS days so thank you to everyone who has contributed in any way to make Open Rails what it is today and hopefully we can ALL work together to make it even better in the future! Everyone has something to bring to the table so lets keep plugging away at it to make it one of the best and most realistic train simulators out there! Just wanted to put these two screen shots up to show everyone how far we really have come!

On a different note I agree with what Traindude is saying! I think both the steam and diesel exhaust needs to be looked at in the future.

[Weter]

@Traindude.
Alas, exhaust effect's diameter currently affects smoke behavior, what is done unperfect in addition.
The speed of smoke ejection (and plume height) currently proportional to this size, according to rate.
I've no idea, how rectangular shape of origin will interact with that algorytms, suspect total mess-up.

Usually, we have 2-4 thin plumes, placed close to each other on row.

[railguy]

My biggest complaint about exhaust in OR is that it only appears to be affected by train motion, while the wind effect is just random. This gets back to OR weather. Wind is not really modelled in OR and it should be. Both smoke and precipitation should be affected by wind. MSTS, for all of its other massive imperfections, made it possible for wind to affect precipitation.

[Weter]

1.James Ross have made some improvements for wind physics
2. Was wind implemented in MSTS???

[Traindude]

Weter, on 08 March 2024 - 10:36 PM, said:

@Traindude.
Alas, exhaust effect's diameter currently affects smoke behavior, what is done unperfect in addition.
The speed of smoke ejection (and plume height) currently proportional to this size, according to rate.
I've no idea, how rectangular shape of origin will interact with that algorytms, suspect total mess-up.

Usually, we have 2-4 thin plumes, placed close to each other on row.

Well, yes, but using multiple emitters for just one stack or port is, IMHO, just "a short-term solution to a long-term issue." The more we can do in ORTS without necessarily "faking it", the better off we'll be.

railguy, on 09 March 2024 - 08:26 AM, said:

My biggest complaint about exhaust in OR is that it only appears to be affected by train motion, while the wind effect is just random. This gets back to OR weather. Wind is not really modelled in OR and it should be. Both smoke and precipitation should be affected by wind. MSTS, for all of its other massive imperfections, made it possible for wind to affect precipitation.

About 2-3 years ago, the Unstable versions did model wind and its effects on both smoke/steam as well as precipitation. However, the wind effect was deemed too excessive by some (me included) so it was not included in subsequent releases.

[Weter]

Hello.
I do agree with You, but just have checked: the smoke jet's "pressure" - i.e. speed of ejection and height - are proportional to nozzle's diameter and smoke rate parameters:
larger diameter demands higher rate for the same jet's height keeping.
So, in case, we'll get rectangular outlet, it's area (cross-section) should be calculated, in case of the current algorithm will be retained. However, that's not so complex: to calculate rectangle's area, instead curcle's.
So, if one dimension is specified - it could be considered as round/square - as it is now, but two dimensions (x,y) could give us rectangle, as You wish.
The only question then - how to stretch the texture, which is now round/square?

[James Ross]

railguy, on 09 March 2024 - 08:26 AM, said:

My biggest complaint about exhaust in OR is that it only appears to be affected by train motion, while the wind effect is just random. This gets back to OR weather. Wind is not really modelled in OR and it should be. Both smoke and precipitation should be affected by wind. MSTS, for all of its other massive imperfections, made it possible for wind to affect precipitation.

This is all going to be fixed by my PR Consolidated wind simulation #799 (clouds, particles, precipitation), but unfortunately it is not included in the Unstable Version currently.

[R H Steele]

Thank you, sir....appreciate all your work. https://www.elvastower.com/forums/public/style_emoticons/default/hi.gif

[ExRail]

James Ross, on 15 March 2024 - 11:03 AM, said:

This is all going to be fixed by my PR Consolidated wind simulation #799 (clouds, particles, precipitation), but unfortunately it is not included in the Unstable Version currently.

I was just going through your code to see how it would effect the weather type stuff I'm doing as a measure to see how far
my code would be from the next versions, and I'm not seeing any problems ahead.
I will incorporate this code in my unstable and adapt my weather types to this,
I got windspeed/direction effecting clouds in two layers, but no wind effecting snowflakes/rain/smoke
and seeing the clouds go one way while the snow the other is a wtf moment.

One big thing I'm unhappy about is the lack of sense of speed in relation to snowfall, holding still at a station and going 100 km almost feel the same,
to get that, the precipitation system must take into account the speed of the train and 1. place the particlebox further in front of the train as speed increases,
right now there is no snow particle if you go 140 km in front of you and the fall angle into the train should properly also be tied to speed so 0 km 0' 140km 45' or something.
The Map viewer has a function that knows where the train is in 30 seconds if nothing is changed and is creating a line for where precipitation boxes should be placed basically.
Maybe one box for world(external views) and one for local cab view tied to speed.


Seeing the title Diesel smoke made me think of a tool I made in 2016 but never got developed further(more then one emitter),
this is the kind of tool we should have for Open Rails/Mono to create smoke and fx

I'm on Code::blocks with Mingw-W C++ SFML +(Thor + Tgui + SfeMovie) where i made this so I'm out, but some one should :-)



If someone wants to play with it and it actually works, then changing the particle texture png change the 7 double click selection buttons on top.

Attached File(s)

•  Release.zip (6.05MB)
  Number of downloads: 28