[cwvs]

Hello All!

I have read a lot of posts concerning diesel smoke and how it is displayed in Open Rails. I decided on a whim to modify my dieselsmoke.ace file to see if I could get a better result. What I got is amazing! By opening the alpha channel in PSP7 and selecting "Brightness/Contrast" and setting -60 Brightness with 10 Contrast, I have very realistic smoke (almost clear, but darkens when increasing throttle). What I didn't expect to see was heat shimmers from the exhaust! Apparently, by using such a dark alpha, heat shimmers can be achieved. Try it yourself! Just back-up dieselsmoke.ace before modifying. Then see what you think. I would post pictures, but it is hard to capture, a video would be better, but I do not have the capability.

Bob :)

[_o_OOOO_oo-Kanawha]

Why don't you post your modified dieselsmoke.ace file here? Zip it first and add it to your message.

[cwvs]

Okay! Here it is! Give it a try and see what you think.

Bob :)

Attached File(s)

•  Dieselsmoke.zip (96.49K)
  Number of downloads: 604

[_o_OOOO_oo-Kanawha]

Thanks. I hardly see any smoke, so I think it is too light, both in v0.8 and in x1565.

The smoke emitters aren't influenced by any OR setting? It is all in the .cs source code, which is mainly particle driven.

Your file could be used to simulate heat haze over exhaust and radiators, but that would require a new .eng file and lots of tweaking

[cwvs]

Thanks for the feedback. Maybe I can try again with a not-so-dark alpha. I thought it was pretty close, and saw smoke on change of power settings. I am thrilled because the result was much better than I had expected. The heat shimmers were a totally unexpected bonus. I will try making some modifications to the file and see where we end up.

Bob :)

[_o_OOOO_oo-Kanawha]

Please remember, Bob, the current smoke/exhaust emitter and texture file vs. particle driven technique are still highly experimental and bound to change from what I have read on this board.

It is nice to tinker with things like these, but different players have different wishes and requirements, so you cannot please them all.

Was there the same difference between your dieselsmoke.ace and the one most of us use, which is from 3DTrains, when in MSTS proper?

[cwvs]

I don't know, I'll have to look at that. I do not use MSTS much anymore. And I can't remember if this is the default smoke or 3DTrains, I have used both. I will do some more experimenting. I was just excited about how it turned out that I wanted to share. Feel free to experiment yourselves. I just found the heat shimmers (for lack of a better term) an unexpected bonus!

Bob :oldstry:

[_o_OOOO_oo-Kanawha]

Walter, what dieselsmoke.ace was used in your video? Or did it come with the FT locomotives Ontario & Western units?

Diesel exhaust description.

I've read an interesting thread on this subject lately which related exhaust to the workings of Diesel locomotives load, rise and fall of r.p.m. etc. , but can't remember exactly where.

Railworks has a more sophisticated exhaust particle emitter system, but the setup and inner workings are like voodoo magic, as there is hardly any documentation. It allows smoke/exhaust to rise and fall, disperse and dissolve. But having to animate a couple of thousand particles more, however small, puts a burden on the simulation and rendering engines. It still lacks scripted speed and intensity variations and most locomotives carry a balloon of smoke which doesn't look prototypical.

[copperpen]

I am getting similar smoke behaviour with the BLW-ZT F3. At idle the smoke puffs out in a huge square-ish cloud, dissipates and puffs again.

[rdamurphy]

copperpen, on 17 April 2013 - 12:05 PM, said:

I am getting similar smoke behaviour with the BLW-ZT F3. At idle the smoke puffs out in a huge square-ish cloud, dissipates and puffs again.

On the F3, my guess is it uses a very small value for one of the parameters int the .eng file, if someone could post the .eng, I could take a quick look at it. I wouldn't recommend ANY values lower than 1.

Diesel exhaust is difficult to simulate, and I have not given up yet on experimentation with it. But there are obviously limits. I've been a truck driver my entire working life and I've spent a lot of time with diesels, including working as an engineering test driver for two diesel manufacturers. Diesels burn diesel fuel, obviously. They're not supposed to "smoke" at all. If they do, there's something wrong with them. In the case of early Detroit diesels, what was wrong with them occured in the design department. They ended up supercharging AND turbocharging the 8V92 diesels to reduce emissions.

Often when you see black smoke, it's a slow or bad turbo, "dirty" fuel (high soot content) or a clogged air filter.

The problem with the simulation currently is that we're basing the amount of smoke on the rpm of the engine, with some adjustment thrown in for throttle setting. What we need to do a better job is to reference the effort the diesel is producing. A diesel under load doesn't smoke, but puts out hotter exhaust, engine temperature varies also.

What I tried to do was come up with something that is a compromise between a Fairbanks Morse diesel which puts out copious amounts of smoke (which is odd, since the engines were originally designed for submarines) to a modern GEVO with computer controls designed to prevent them from smoking.

Looking at the video, first, let me check the .eng parameters, we may need to add a mininum amount of particle values just to prevent issues like that one. I didn't actually want to do that because, again, there are some locomotives which really shouldn't put out any exhaust at all.

Robert