[James Ross]

gpz, on 05 March 2014 - 11:07 PM, said:

So the problem still remains, that MSTS Variable1 is about 2/3 "slower" than OpenRails' one. Probably it is 2/pi. If everyone agrees here, I commit a change today evening, adding a 2/pi multiplier to steam Variable1. And everyone can test then if it solves the issue. Is anyone against?

Having seen no data, I cannot just agree; have you set up a locomotive that uses Variable1 in a distinct manner and compared MSTS and OR? Have you tried changing things like the wheel diameter? I don't really mind you changing it but, unless you've done good research into MSTS behaviour, we're only going to be fiddling with it again in 6 months. (Such data should go in to a comment in the code.)

[gpz]

In a couple of test cases I could also reproduce this approximately 2/3 speed difference. The problem is, that the people can do the good MSTS research cannot make changes to OpenRails code, and vice versa. The point is to make possible to them to do the measurements. Otherwise we will stuck with current behaviour, which is not bad, just people will need to adjust .sms files to match this one. It would be like cutting the Gordian knot.

[James Ross]

gpz, on 06 March 2014 - 12:56 AM, said:

In a couple of test cases I could also reproduce this approximately 2/3 speed difference. The problem is, that the people can do the good MSTS research cannot make changes to OpenRails code, and vice versa. The point is to make possible to them to do the measurements. Otherwise we will stuck with current behaviour, which is not bad, just people will need to adjust .sms files to match this one.

There's no particular reason why the people doing the testing need to make changes to OR; they only need to decipher enough of how it works in MSTS and how OR is not matching that for someone else to make the change. Just go and make whatever change it is you think will help and let's hope enough people test it. :sign_thanks:

[rdamurphy]

James Ross, on 05 March 2014 - 02:41 PM, said:

Please can uyou use the correct conversion routines (e.g. MpS.ToKpH) and not hard-code random constants into the code, thanks. :sign_thanks:

Thanks for the heads up! I'm an old time VB guy, much less OO than C#. :blush2:

I'll fix it when I get home in the morning...

Robert

[rdamurphy]

You might want to check to see if the variable1 parameter is being read from the SMS file correctly. I can check that in the morning if you'd like. I'm not sure what effect that will have on your steam locomotive code however.

Robert

[gpz]

Robert, could you please look on this thread as well, please: http://www.elvastowe...ch-curve-since/
Looks like the issue was caused by the recent mods.

[ATSF3751]

I would be willing to make changes to my SMS files for it to work in Open Rails it will just take some time to adjust all of my sound sets! Since everyone is starting to go Open Rails now I might as well start making my sound sets to work with Open Rails instead of MSTS. I can monkey around with one of my sound sets and see how it turns out!

Brandon

[rdamurphy]

Here's the problem we're going to run into, gentlemen. As James pointed out to me, this should work the same in OR as it does in MSTS. And he's correct, I can't disagree.

And, as James points out it should be fixed correctly, now, so we're not revisiting this in six months again...

So, in order to do that, every sound parameter based on VariableTriggers is current set to MPS. But the MSTS files define it as kph.

To make the corrections, we would need to correct ALL of them, not just the two I've corrected - if that's the consensus, I can do that, but I don't want to do that if we're not in agreement on it. But if I understand what James is saying is that we SHOULD go ahead and make the corrections, which I will do, however, since I'm not familiar with the sound classes it may take a couple of days and a couple of versions, so if we want to do that James, we probably need to not do the usual X version today or tomorrow until it's done and tested.

Perhaps you could do the current X Version based on 2079 instead?

Comments?

Robert

[James Ross]

rdamurphy, on 06 March 2014 - 09:24 AM, said:

To make the corrections, we would need to correct ALL of them, not just the two I've corrected - if that's the consensus, I can do that, but I don't want to do that if we're not in agreement on it. But if I understand what James is saying is that we SHOULD go ahead and make the corrections, which I will do, however, since I'm not familiar with the sound classes it may take a couple of days and a couple of versions, so if we want to do that James, we probably need to not do the usual X version today or tomorrow until it's done and tested.

Perhaps you could do the current X Version based on 2079 instead?

I, personally, don't think it matters. Open Rails is a work in progress and the weekly experimental releases even more so. I will actually be away for the weekend from tomorrow morning, so if you really want to avoid the automatic build this week let me know in the next 12 hours and I'll temporarily disable it. I don't think it's necessary, but I'm happy to do it if you ask.

[rdamurphy]

No, I think we're good now. I reverted the changes for now, and I'll work through it this weekend to find all of the changes that need to be made so the conversions are done properly. Apparently, there were far more of them than I anticipated...

Robert