Potential Route Setting Bug

You are here: Home > Forum > Simulations > Released > Three Bridges > Potential Route Setting Bug

Page 1 of 1

Potential Route Setting Bug 27/10/2012 at 12:42 #36961
ajax103
Avatar
1103 posts
I just thought I run a route learner using a Class 377/4 booked from Lovers Walk to Lewes via Brighton and the East Coastway in the early hours of the day so not to clash with any other movements.

Anyway, I signalled it correctly into Platform 3 as I booked it to and ensure it was on the stops (buffers) with the next passing point being booked as London Road.

However I thought I just signal the train towards Hove on purpose just to see if the driver would pick up on it and advise a wrong route was being given.

However no such call from the driver was forthcoming and the driver went merrily on their way to Hove - now I know the driver could reverse at Hove then again at Wivelsfield/Haywards Heath but it would take a lot longer I believe then the direct route via the East Coastway.

Surely the driver ought to have rang up and advise of a wrong route being given?

Log in to reply
Potential Route Setting Bug 27/10/2012 at 13:37 #36966
Stephen Fulcher
Avatar
2007 posts
Sounds like a bug to me. Was the train definitely booked out via the direct line to Preston Park as I know there are a couple that are booked to go via Hove and reverse.
Log in to reply
Potential Route Setting Bug 27/10/2012 at 14:32 #36968
ajax103
Avatar
1103 posts
The train was a extra service that I inserted into the timetable, the next passing point was booked to be London Road and not Hove or Preston Park.

I know it was definitely booked to head via London Road, Falmer etc...and then onto Lewes as I checked it before I finished inserting it into the timetable.

Log in to reply
Potential Route Setting Bug 27/10/2012 at 14:35 #36969
TimTamToe
Avatar
654 posts
" said:
Sounds like a bug to me. Was the train definitely booked out via the direct line to Preston Park as I know there are a couple that are booked to go via Hove and reverse.
From what I gathered from Ajax's post, he "made up" the train with it going through Falmer etc after Brighton and wasn't supposed to end up going through / via Preston Park (or Hove).

Would say it sounds like a bug - have you tried it with just Brighton and Lewes as passing / stopping points instead of London Road as well?

Gareth

Log in to reply
Potential Route Setting Bug 27/10/2012 at 15:10 #36972
John
Avatar
884 posts
" said:
have you tried it with just Brighton and Lewes as passing / stopping points instead of London Road as well?
I've just had a go at doing that, and the driver still took the route towards Hove without complaint - in fact, he's still going. Probably at Portsmouth by now...

He refused the route towards Preston Park, though.

Log in to reply
Potential Route Setting Bug 27/10/2012 at 15:39 #36974
postal
Avatar
5189 posts
It all depends on how much the developer has written into the sim. Any "wrong route" call has to be programmed into the sim so that a message can be generated when a train due to go to Point A is put onto a route that will not take it to Point A. If the developer has not put a specific "wrong route" message in for a particular circumstance, then the driver will keep going until the train falls off the sim.

In that sense it is a bug insofar as the developer has not covered every eventuality. However, it would take much more work so taking much more development time to have every possible wrong route written into sim. Once again the developer is caught between trying to get a sim out to SimSig users or holding it back for many months while adding in all of the nice-to-haves.

“In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe
Log in to reply
Potential Route Setting Bug 27/10/2012 at 16:15 #36975
Hooverman
Avatar
306 posts
When I 1st went into Three Bridges we were still running the Seaford to West Worthing to Seaford services and on more than one occasion they had to run via Pesten Park and reverse. We have also run trains to and from Victoria via Hove and into platforms 1 & 2 when things have gone tits up at Montpelier jcn Although I appreciate it's not quiet the same it still could come in handy.
Log in to reply
The following user said thank you: TimTamToe
Potential Route Setting Bug 29/10/2012 at 16:51 #37022
TimTamToe
Avatar
654 posts
" said:
We have also run trains to and from Victoria via Hove and into platforms 1 & 2 when things have gone tits up at Montpelier jcn Although I appreciate it's not quiet the same it still could come in handy.
As I've found during a sim with a TCF coming out of Platform 3 at Brighton causing no access into / out of the carriage road and no routes able to be set on the Up Main. Glad its happened at 9pm and not during the rush hour seeing as one platform at Hove is out as well - waiting an hour for a fitter.

Log in to reply
Potential Route Setting Bug 29/10/2012 at 17:21 #37023
GeoffM
Avatar
6274 posts
Fixed. That branch was indeed missing from the invalid locations data for that signal.

In general though, it's not always possible to code for every situation of wrong route. The Preston Park - Hove - Brighton triangle is an example of where it could get difficult, where going one way would appear to be invalid unless going somewhere else first. The options available to developers are:

  • Invalid locations for all trains

  • Invalid locations for stopping trains

  • Locations which must not be the next location of a train taking this route

  • Locations, one of which must appear in the timetable of a train taking this route

  • Locations, one of which must be the next in the timetable of a train taking this route



So, in the triangle example above, we have to use the "not next location" for Preston Park / Hove for trains leaving Brighton. Route 673A and B have Hove No.1 Up Siding as an invalid location for all trains as there is no other way of getting there - well, except for shuttling backwards and forwards through Hove - which demonstrates how it can become difficult. As a result, I generally don't code shunt signals with invalid locations if it's possible to get to the same location by other means.

SimSig Boss
Log in to reply
Potential Route Setting Bug 30/10/2012 at 14:19 #37043
clive
Avatar
2736 posts
Two notes on this:

(1) The "invalid location" logic that you've described only looks as far as the next reversal of the train. So "invalid location Hove" won't stop a train from being routed to Preston Park and then reverse to Hove.

(2) The current core code has a more generic facility, so you can write tests that you couldn't in older simulations.

Log in to reply