A new bug has arisen related to convoy spacing, presumably as a result of the fix to the previous bug where convoys would not adopt their proper waiting time slot when another convoy on the same line is reversing at a station.
See the following save:
https://dl.dropboxusercontent.com/u/61716/convoyspacing.sve (https://dl.dropboxusercontent.com/u/61716/convoyspacing.sve)
Convoy (1270), on the 'London STP-Sheffield (Slow)' line, is approaching Derby from the south east. When it arrives in the reversing siding after Derby, its waiting time is set at 1 hour 55 mins, despite the fact that there is no other convoy on that line waiting in the siding. That is: it fails to occupy the next spacing slot.
This doesn't seem to happen all the time -- note that it doesn't happen with the 'Sheffield (fast)' convoys at the same siding -- so perhaps the cause here is that the last 'Sheffield (slow)' convoy had only left the siding a few minutes before the arrival of (1270). But since the previous convoy is well clear of the siding before (1270) arrives, this should not trigger a bump into the next spacing slot: (1270) should be due to wait 55 minutes, not 1 hour 55 minutes.
Thank you for this report. Can you double check whether this can be reproduced with 11.4? I can't immediately see how my fix in 11.5 could cause this, and it would be helpful in narrowing down the diagnosis.
I've just tested the save in 11.4, and the bug seems not to arise -- the train's waiting time on arrival at the siding is set at 55 minutes, the correct value.
Here is what appears to be a whole sequence of cases of the same bug:
http://simutrans-germany.com/files/upload/Thetford_189000001_1891-11b.sve (http://simutrans-germany.com/files/upload/Thetford_189000001_1891-11b.sve)
The tram line 'Maltby Outer Circuit' has a timing stop at 'Maltby Cricket Ground', with a spacing of just under 6 minutes. Each newly-arrived tram should wait for at most about 3 minutes before leaving. But since upgrading to 11.5, they all wait for 8-9 minutes instead, that is, for the following timetable slot instead of the one they should be using.
Best wishes,
Matthew
Matthew - thank you for the upload, and Carl and Matthew thank you both for the report. I think that I have now found and fixed this on the 11.x branch.
Thanks as ever James. Thank you also for the fixes you made on the reversing topic -- those all sound like very positive changes.