New Acceleration Setting

[Emperor Jake]

How about this for an idea - 3 acceleration settings:

Original acceleration - the one where trains slow to 20km/h at every slope
Realistic Acceleration - this should be the way that realistic acceleration was earlier
Ultra-Realistic Acceleration - THe newest one that takes into account the air resistance and tunnel resistance, and any other new things people may want optional.

There have laready been a few people (including me) who have complained of the air resistance slowing down trains...
Thanks,
Jake

[Supercheese]

Well, perhaps the middle category should be something like "Semi-Realistic" or "Quasi-Realistic" then, and the final category plain "Realistic". After all, having no air resistance isn't "realistic" in the strict sense of the term...

[Terkhen]

The air drag method is not very realistic, it's just a model. That's why, in my opinion, calling it "ultra-realistic" is not really adequate (In my personal opinion none of the methods is realistic and neither is realism the goal, but that's different discussion). Besides, the increased slowness of vehicles is not only caused by air drag. See r20303 and r21106.

[Emperor Jake]

The naming doesn't really matter, call it what you want. The point is the function of the switch, to be able to turn off the air drag and have an option to set the "realistic" acceleration model to what it was like before.

[michael blunck]

The naming doesn't really matter, call it what you want. The point is the function of the switch, to be able to turn off the air drag and have an option to set the "realistic" acceleration model to what it was like before.

Do I remember correctly, that me and Terkhen had explained it to you already in that other thread?

regards
Michael

[Terkhen]

That explains my déjà vu with this thread

As I already mentioned, the best solution would be tweaking TE/air drag to match the changes in the acceleration model (if the NewGRF set authors feel that it is needed at all). I don't feel that Yet Another Setting is required in this case, as the changes are minor. Besides, adding settings to very sensitive parts of the code (such as the acceleration code) might create a noticeable performance hit.

[michael blunck]

As I already mentioned, the best solution would be tweaking TE/air drag to match the changes in the acceleration model (if the NewGRF set authors feel that it is needed at all).

I didn´t had the time to cross-check air drag behaviour in both programs, resp. with the DB Set where some engines were originally using that feature. I´ll try to run some tests on the weekend.

Other than that, IMO it´d pointless trying to satisfy every player by "tweaking" coefficients. As already stated in that other thread, the ball´s in the yard of newgrf developers now, fixing their code.

regards
Michael