[thowengines]

As the team works towards a stable V1.0 I thought I should let you know of this problem which has popped up lately. I can't be more specific about when it occurred. I reckoned it was wiser to let you know as soon as possible rather than to go back testing all the earlier builds, but it is in X3000 and was in an earlier version but I can't recall its number. When uncoupling vehicles from a train, using the F9 display and mouse click on the coupler I get the NullReferenceException detailed in the attached screen dump. I have tested it on a variety of consists and a variety of vehicles at a variety of locations, all with the same effect.

Attached thumbnail(s)

• 

[thowengines]

My posting has obviously not attracted much interest! Nevertheless, it is real, and so I have pursued it further. The problem only arises in Timetable mode, and first appeared in X2989. This was the build in which Rob made various changes to the Timetable mode. I suspect my problem crept in with those changes. I hope this helps somebody pinpoint the issue. It is still there in the latest builds. I am not sure which build is the base for the stable release version, but if it is after 2988, I fear it too will contain the bug.

[roeter]

The couple/uncouple logic is due to be completely revised and rewritten as part of the ongoing development of the timetable mode. As a result, the present method where the crash occurs will no longer be used in timetable mode and thus the crash will be resolved by default (no doubt at first being replaced by others, I fear). However, that will not happen before the release of version 1.0, so indeed this crash will be carried over into version 1.0. At the moment, uncoupling or coupling in timetable mode is not supported - coupling is in fact completely disabled.
It must be realised that version 1.0 is particularly aimed at compatibility with MSTS, which ofcourse does not apply to timetables. The timetable mode is very much still a work in progress, development is continuing (at a low pace due to lack of time) and will continue after release of V1.0. No commits are made at the time, though, due to the commit rules prior to V1.0.
It is neither possible nor practicle to issue a 'stable' release for timetable mode at this moment in time. Too much is still missing, while part of what is there is still being evaluated and improved, sometimes because of errors, sometimes because of what happens is not quite what was intended. In such completely new developments like the timetable mode, the borderline between correcting errors and new development is rather blurred, and regularly certain problems can only be properly dealt with by new development like adding additional commands or parameters. As a result, code changes are also a mixture of error correction and new development, and are therefor outside the scope of allowed commits prior to V1.0. Because of this, all timetable mode changes will be held back until after V1.0 has been released.

Regards,
Rob Roeterdink

[thowengines]

Thanks Rob for your detailed response and explanation. Timetable operation is my preferred part of OR and I really appreciate the work you have put into it. I understand the position you are in, and until any developments do occur, I am quite happy to continue enjoying what we do have.