[Weter]

Ah, yes. This is a separate wide question, forming a task for future development. I've read about that, but no actual results yet.

[Paul B]

Traindude, on 04 February 2023 - 06:31 PM, said:

While I enjoy the addition of the CylinderSteamExhaustxFX added with the CTN NYC Atlantic, I have one comment about it. On my machine at least, the effect seems to disappear whenever I accelerate past 8-9 mph. Is this intentional?

Hi;

It seems like this problem still exists. I got a few 'blasts' when speeds were very, very low. It seemed each successive 'blast' had a shorter duration as speed increased until such a point that the duration was so short the steam was not rendered/visible.

It seems like the cylinder cocks are a good starting point for the long wished for blast effect. And it almost works.

I wonder if there is any merit in having distinctive blasts up to a certain point then replacing it with a continuous particle stream - the point at which the human eye could not distinguish individual blasts?

Just some thoughts - thanks for the work on implementing this.

-Paul

[ATSF3751]

Traindude, on 05 February 2023 - 06:24 PM, said:

Currently, most articulated locomotives in MSTS/ORTS treat the front and rear halves of the locomotive as separate locomotives each with its own separate .eng file. This is why I am promoting the move away from these multi-section models and going for single-piece models...but that's a discussion for another thread!

I have to agree with this when it comes to articulated locomotives. It does not make any sense in Open Rails to have a single locomotive split into 2 Eng files. For MSTS it did make sense because it did not have the capability to have articulated locomotives as a single unit but I believe Open Rails can handle it.

Traindude, on 05 February 2023 - 06:24 PM, said:

The way I see it, have the AI locos start from a stop with the cylinder cocks open, and then have the cylinder cocks close when the loco accelerates past 5 MPH, or 30 seconds have passed (whichever is longer), or until the locomotive comes to a stop again.

My thought on this is when having multiple steam locomotives in a consist why not have the cylinder cocks be controlled by the Player? It looks really stupid to have your lead locomotive have the cylinder cocks open while starting and the locomotive following to not have anything. Also when having multiple steam locomotives in one consist you should be able to control both locomotives separately in some ways. This would include cylinder cocks, blow downs, whistle, along with a few other features.

That is just my take on it. Feel free to chime in if you would like.

Brandon

[Jonatan]

ATSF3751, on 27 February 2023 - 02:10 PM, said:

My thought on this is when having multiple steam locomotives in a consist why not have the cylinder cocks be controlled by the Player?
Brandon

I believe in MSTS they did? Like with sand and other controls, why shouldn't cylinder cocks on AI locos also be controlled by the player?


On steam exhaust the few puffs we currently get are indeed not worth the effort to implement into an .eng file, ORs smoke FX drowns it out at any speed or throttle setting and as said elsewhere it just seems to stop. At some speeds the way OR generates smoke is convincingly similar to exhaust puffs anyway!
I understand it's a new feature so it still needs alot of tweaking to get the code right.

I don't know how MSTS did it but yes, smoke and steam FX was better in some regards and in a steady stream regardless of speed. How Peter can replicate that I don't know but I hope he figures it out for the future. Smoke and steam is such a complex and dynamic mechanism that afaik it's almost impossible to simulate it perfectly...

I remember even seeing the smoke darkening briefly with each shovelfull of coal, if that was intentional or not by KUJU I don't know but it was very realistic for what it was.

[Laci1959]

Hello.

The following video was published in a Facebook group dealing with steam locomotives.

My link

Based on the sound of knocks, the locomotive pulls. Since the reverser is set maybe a third of the way, and the regulator is slightly open, there is almost nothing above the chimney. From time to time it is clearly visible when the heater burns a little. The fact that the German war locomotive is very powerful also belongs to the truth. Those couple of cars are not a significant burden for him.
Video source Facebook.

Sincerely, Laci 1959

[darwins]

This is a good illustration of what we should ultimately aim for. Put some coal on and smoke rapidly becomes black then gradually becomes clear again. Then put some more coal on. (Only if the fire is less hot or too much coal is put on, then the smoke will remain black for longer.)

[Jonatan]

I concur. But Peter have stated elsewhere the AI firing mechanics need a complete rebuild, at current we have a mechanical stoker for fireman and the darkening smoke mechanic wouldn't change much from what we already experience.

Again, there's a gazillion things and variables to make steam and smoke realistic. For now I think Peter should focus on getting the exhaust puffs right.

[Traindude]

Jonatan, on 02 May 2023 - 10:59 AM, said:

I concur. But Peter have stated elsewhere the AI firing mechanics need a complete rebuild, at current we have a mechanical stoker for fireman and the darkening smoke mechanic wouldn't change much from what we already experience.

Again, there's a gazillion things and variables to make steam and smoke realistic. For now I think Peter should focus on getting the exhaust puffs right.

In case anyone is interested, I am doing an analysis of the LMS' "Little and Often" training film and developing suggestions for how to improve the firing algorithms.

I also want to repeat that while I like Peter's new CylinderExhaustFX emitters, keep in mind that they are dependednt on the ORTSWheelCrankAngleDifference parameter and therefore they're only applicable to conventional "Rod" engines.

In the case of geared locomotives, the cylinder exhaust should be tied to the crankshaft rotation instead of the wheel rotation. For example, a Shay has a 3-cylinder steam engine driving the crankshaft and therefore would make 6 puffs in a single revolution of the shaft. However, since there is a 4:1 or 6:1 gear reduction between the shaft and the wheels (meaning it takes 4 to 6 rotations of the shaft to complete one rotation of the wheels), this would result in anywhere between 24 and 36 puffs per wheel rotation. This is why geared locomotives sound like they're going super-fast even though, in reality, they're only moving at little more than walking speed.

[steamer_ctn]

Traindude, on 02 May 2023 - 12:32 PM, said:

In case anyone is interested, I am doing an analysis of the LMS' "Little and Often" training film and developing suggestions for how to improve the firing algorithms.

The challenge with this approach is that it assumes the fire is behaving the same as demonstrated on the film.

This is another layer of complexity for OR, how should the fire combustion behavior be modeled? Thus we have two models that need to interact with each other.

[Traindude]

steamer_ctn, on 04 May 2023 - 12:11 AM, said:

The challenge with this approach is that it assumes the fire is behaving the same as demonstrated on the film.

This is another layer of complexity for OR, how should the fire combustion behavior be modeled? Thus we have two models that need to interact with each other.

Understood. Well, it's the thought that counts. You still appreciate having this kind of reference material though, right?

Back on topic: Do you agree that in the case of geared locomotives, the CylinderSteamExhaustFX should be tied to the crankshaft rotation and not the wheels? So would ORTSCrankshaftCrankAngleDifference ( x ) be a new parameter in those cases?