Upcoming Games


Kings Cross 24/09/2017 16:00:00 - tjfrancis

Upcoming Events

Bugs noticed

You are here: Home > Forum > Simulations > Released > Oxford PSB > Bugs noticed

Page 1 of 1

Bugs noticed 23/07/2017 at 20:56 #96575
peterb
Avatar
380 posts
Here's a few issues I've noticed while playing Oxford in 2009 era - so can't say if they extend to other eras too.

- LC locations - I asked callers to call back when a train passes. At Islip, I got an 'oi' after an up train had apparently passed the crossing. In the down direction I had a train 'pass' Islip LC before arriving at the station. I think I also recall someone at Manor Farm calling back when the down train in question was passing Kennington loops.
- At one point, DX59 failed. Trains were initially reporting 'standing at a red signal', waiting a rough two mins, then calling back 'still waiting at unlit signal' and the correct phone call.
- Then, DX55 failed, so I had a train standing at SB2211 off sim. But the train never phoned up requesting permission to SPAD which I thought trains would do? When I told the train to SPAD and it got to DX55, it also behaved in the same way as trains at DX59 above.
- Trains also don't appear to phone up to report a ACOA which again I thought was customary?
- Also, a red signal will approach lock when the route is cancelled from it when a train has just passed it (the TC ahead of the signal is occupied). Why is this?
- I had a train reversing (manually - not using the shunt sig) out of the dock at Oxford. When it got to the end of TC TFR, before points 245, the train incorrectly reported that they were set against the train which they weren't. Plus, this also happened for a train occupying TC TCN, approaching points 277B.
- The TD didn't step from AW2 to 11 in a standard routing when other trains were qued behind off sim

Last edited: 23/07/2017 at 22:54 by peterb
Reason: None given

Log in to reply
Bugs noticed 24/07/2017 at 15:13 #96584
GB
Avatar
64 posts
Don't know if its true to real life but seeing 120 seconds approach locking platform 3 (signal 71) to either signal 75 on the UPL or signal 25 on the up main without a train in section.
Last edited: 24/07/2017 at 15:15 by GB
Reason: None given

Log in to reply
Bugs noticed 24/07/2017 at 18:28 #96588
GeoffM
Avatar
3548 posts
peterb in post 96575 said:

- Also, a red signal will approach lock when the route is cancelled from it when a train has just passed it (the TC ahead of the signal is occupied). Why is this?
Just this for the moment - this is correct. You'll notice the route drops away anyway a couple of seconds after the first TC clears. It was a recent change to correct a long standing (but unnoticeable to most but the most pedantic) issue with approach locking.

SimSig Boss. Website issues: Check here first: https://www.simsig.co.uk/Wiki/Show?page=newwebsite
Log in to reply
The following user said thank you: peterb
Bugs noticed 24/07/2017 at 18:34 #96589
Stephen Fulcher
Avatar
1341 posts
GB in post 96584 said:
Don't know if its true to real life but seeing 120 seconds approach locking platform 3 (signal 71) to either signal 75 on the UPL or signal 25 on the up main without a train in section.
This is non-comprehensive approach locking, whereby the Interlocking does not "look back" behind a signal to see if a train is approaching, but simply enforces a timeout when the route is pulled regardless of what is going on within the track circuits. It is extremely common, and in this instance is correct.

Log in to reply
The following user said thank you: BarryM
Bugs noticed 24/07/2017 at 19:17 #96592
peterb
Avatar
380 posts
GeoffM in post 96588 said:
peterb in post 96575 said:

- Also, a red signal will approach lock when the route is cancelled from it when a train has just passed it (the TC ahead of the signal is occupied). Why is this?
Just this for the moment - this is correct. You'll notice the route drops away anyway a couple of seconds after the first TC clears. It was a recent change to correct a long standing (but unnoticeable to most but the most pedantic) issue with approach locking.
Was previous practice not prototypical then?

Log in to reply
Bugs noticed 24/07/2017 at 19:30 #96593
GeoffM
Avatar
3548 posts
peterb in post 96592 said:
GeoffM in post 96588 said:
peterb in post 96575 said:

- Also, a red signal will approach lock when the route is cancelled from it when a train has just passed it (the TC ahead of the signal is occupied). Why is this?
Just this for the moment - this is correct. You'll notice the route drops away anyway a couple of seconds after the first TC clears. It was a recent change to correct a long standing (but unnoticeable to most but the most pedantic) issue with approach locking.
Was previous practice not prototypical then?
Not entirely!

SimSig Boss. Website issues: Check here first: https://www.simsig.co.uk/Wiki/Show?page=newwebsite
Log in to reply
Bugs noticed 24/07/2017 at 21:05 #96594
peterb
Avatar
380 posts
An update on the last TD-stepping issue - for some reason the approach berth had filled with the ID of the train at MM36 behind it, when the train in question passed AW2 it filled the berth on 11 with ****
Log in to reply
Bugs noticed 26/07/2017 at 21:32 #96636
Stephen Fulcher
Avatar
1341 posts
peterb in post 96575 said:
Here's a few issues I've noticed while playing Oxford in 2009 era - so can't say if they extend to other eras too.

- LC locations - I asked callers to call back when a train passes. At Islip, I got an 'oi' after an up train had apparently passed the crossing. In the down direction I had a train 'pass' Islip LC before arriving at the station. I think I also recall someone at Manor Farm calling back when the down train in question was passing Kennington loops. It is difficult to test this, so have logged on Mantis for Geoff to check the locations,
but I cannot confirm or deny the presence of a bug at this point. Ticket 17686.

- At one point, DX59 failed. Trains were initially reporting 'standing at a red signal', waiting a rough two mins, then calling back 'still waiting at unlit signal' and the correct phone call.
- Then, DX55 failed, so I had a train standing at SB2211 off sim. But the train never phoned up requesting permission to SPAD which I thought trains would do? When I told the train to SPAD and it got to DX55, it also behaved in the same way as trains at DX59 above. Above two points are a core code trivial bug relating to the message box. When the train phones up the message is correct. Logged as ticket 17687.
- Trains also don't appear to phone up to report a ACOA which again I thought was customary? I cannot reproduce this.
- Also, a red signal will approach lock when the route is cancelled from it when a train has just passed it (the TC ahead of the signal is occupied). Why is this? Geoff has already covered this in a previous post.
- I had a train reversing (manually - not using the shunt sig) out of the dock at Oxford. When it got to the end of TC TFR, before points 245, the train incorrectly reported that they were set against the train which they weren't. Plus, this also happened for a train occupying TC TCN, approaching points 277B.Confirmed, with TFR, if you reverse P251 then the train will carry on along the Up Main in the Down Direction. The second one is actually down to 278 points, which again when reversed the train will proceed. Logged on Mantis as there are legitimate reasons to talk a train past the signals over the routes that do not work. Ticket 17688
- The TD didn't step from AW2 to 11 in a standard routing when other trains were qued behind off simI cannot reproduce this either.
See my bold comments as to each issue above. Also can you confirm what sim version you are using please?

Log in to reply
Bugs noticed 27/07/2017 at 19:10 #97659
peterb
Avatar
380 posts
Thanks Stephen, it's loader 4.6.4/sim 1.1.
Log in to reply
Bugs noticed 29/07/2017 at 13:45 #97696
peterb
Avatar
380 posts
Re Mantis 17686: this is the message log from when it just happened again at Islip LC. The caller was asked to call back after train 2L02 (2009 up) crossed.

06:01:26 Islip: Now safe to cross [when crosser called back]
06:03:33 Islip: Sorry
06:03:58 2L02 arrived at Islip on time

Log in to reply
Bugs noticed 30/07/2017 at 07:27 #97723
clive
Avatar
1544 posts
peterb in post 97696 said:
Re Mantis 17686: this is the message log from when it just happened again at Islip LC. The caller was asked to call back after train 2L02 (2009 up) crossed.

06:01:26 Islip: Now safe to cross [when crosser called back]
06:03:33 Islip: Sorry
06:03:58 2L02 arrived at Islip on time
From the next version of the loader, the caller will make it clear whether or not the train has passed.

Log in to reply
The following users said thank you: uboat, Trainfan344
Bugs noticed 30/07/2017 at 22:51 #97736
peterb
Avatar
380 posts
A couple more, still in 2009:

A train leaves Bicester town immediately after permission has been granted. Surely it would have been more typical that a train leaves >30 seconds after permission has been given? I.e. similar to TRTS occurs a minute or more before the RA is given.

Secondly I had a train (5D61) for Oxford UCS, so I requested the slot. The train was subsequently delayed so I cancelled the slot - but the shunter said that I'd never asked for permission! This was the same response before AND after the slot was granted. Save attached.

Post has attachments. Log in to view them.
Log in to reply
Bugs noticed 12/09/2017 at 13:02 #101737
NCC1701
Avatar
120 posts
I'm playing the 2016 version but this is probably relevant to all the eras.
I just had a request from Inkpens LC to allow a flock of sheep across. I've had a few trains around that area on both lines so three times I've asked for the sheep to call back "after the next train" which they have done.
1O86 is approaching from Cherwell Valley as I've called him and asked for a block on the up line. He replied along that he would apply the block behind approaching trains, which is as I would expect, however shortly after that I get a message that 1O86 has received an ACOA at one of Cherwell's signals and the line is showing "Blocked".
Attached save is just after the message. Snapshot from 5 minutes earlier also attached, but this is before Inkpens called.
[Loader version 4.6.4]

Post has attachments. Log in to view them.
Nigel, signalman Exeter West & Llangollen
Last edited: 12/09/2017 at 13:04 by NCC1701
Reason: Added loader version

Log in to reply