Page 1 of 1
ARS Strange Decisions 13/11/2024 at 12:52 #159152 | |
chrisdmadd
255 posts |
I've created a 2024 TT for Tyneside and getting some use out of the Blyth and Tyne line at last. The issue I'm having is with a new service but by no means at a new location for this sim. ARS is hell-bent on sending trains out of Newcastle in the wrong order at the same time every hour. I'm either missing something very simple on the TT or there's an issue with ARS for this location which I've never noticed before. A screenshot is attached but I can send the save to anyone that's interested in looking. Many thanks Post has attachments. Log in to view them. Log in to reply |
ARS Strange Decisions 13/11/2024 at 14:34 #159153 | |
postal
5269 posts |
chrisdmadd in post 159152 said:I've created a 2024 TT for Tyneside and getting some use out of the Blyth and Tyne line at last. The issue I'm having is with a new service but by no means at a new location for this sim. ARS is hell-bent on sending trains out of Newcastle in the wrong order at the same time every hour. I'm either missing something very simple on the TT or there's an issue with ARS for this location which I've never noticed before.I must be missing something obvious but could you explain a bit more (or post a saved game). The screen shot seems to show 2T15 (08:50 Newcastle - Ashington) signalled away with 1S03 (08:56 Newcastle - Edinburgh) waiting departure in P4. 2T15 is timed in front of 1S03 to Benton Junction so the ARS would seem to be working exactly as expected. Is the problem further down the line (like at Manors or Heaton) with 1S03 being given priority there? With a saved game the ARS Debug can be examined to see where the decision is being generated. “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 |
ARS Strange Decisions 13/11/2024 at 14:56 #159154 | |
chrisdmadd
255 posts |
Sorry I didn’t make it very clear really. ARS has routed 2T15 out of Newcastle but intends to hold it at Dean Street Crossovers to allow 1S03 to be routed ahead of it there. That is the bit I can’t understand. I’ll PM you for an email address and send you the save if possible. Thanks. Log in to reply |
ARS Strange Decisions 13/11/2024 at 15:00 #159155 | |
postal
5269 posts |
chrisdmadd in post 159154 said:Sorry I didn’t make it very clear really.There are lots of people more expert than I am at the ARS debug side of things. You would be better uploading a saved game here. “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 The following user said thank you: chrisdmadd |
ARS Strange Decisions 13/11/2024 at 15:13 #159156 | |
chrisdmadd
255 posts |
Save File attached.
Post has attachments. Log in to view them. Log in to reply |
ARS Strange Decisions 13/11/2024 at 15:46 #159157 | |
Hap
1048 posts |
It would seem that there may be an issue with how the sim is calculating delays. Would require a wee look into the sim data. So I will log a mantis ticket for that. However, usual disclaimers apply with ARS. It's a tool and not a signallers replacement :). Craig How to report an issue: www.SimSig.co.uk/Wiki/Show?page=usertrack:reportanissue Last edited: 13/11/2024 at 16:21 by Hap Reason: None given Log in to reply |
ARS Strange Decisions 13/11/2024 at 15:50 #159158 | |
chrisdmadd
255 posts |
Thank you Craig, at least I’m not going mad thinking I’ve missed something obvious.
Log in to reply |
ARS Strange Decisions 13/11/2024 at 15:53 #159159 | |
Hap
1048 posts |
I'll log a number on here just after I get a second opinion from a couple of other tickets that kind of surround this same issue that's already on Mantis. Either way, I'll have a number next hour or so. Craig How to report an issue: www.SimSig.co.uk/Wiki/Show?page=usertrack:reportanissue Log in to reply The following user said thank you: chrisdmadd |
ARS Strange Decisions 13/11/2024 at 16:34 #159160 | |
postal
5269 posts |
If my limited knowledge of the ARS technology is correct, the clearing points seem to be governing the decision. The ARS coding has 1S03 clear of the conflicting situation at signal B631, just before Killingworth so about one mile from the junction. 2T15 is not seen as clear by ARS until it reaches signal B635 approaching Newsham i.e. about twelve miles from the junction. So although the unit would be clear as soon as it was off the TC across Benton Jn (and any clearing point or overlap restrictions), ARS is not reading it as clear until twelve miles further down the route. This wait for clearance would delay 1S03 so ARS computes that the least overall delay would occur if 1S03 precedes 2T15 to Benton Jn. IRL the signaller would be aware of this foible, check the timings and make sure that 2T15 (and anything else due to diverge at Benton Jn) was manually signalled over that stretch. “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 Last edited: 13/11/2024 at 16:36 by postal Reason: None given Log in to reply |
ARS Strange Decisions 13/11/2024 at 17:43 #159161 | |
bill_gensheet
1431 posts |
postal in post 159160 said:2T15 is not seen as clear by ARS until it reaches signal B635 approaching Newsham i.e. about twelve miles from the junction.B635 is in the loop at Benton, not Newsham. However the ARS 'clear time' is indeed based on reaching the double track at Newsham, calculated back from the Bedlington time. Inserting a formal 'Newsham LC' time (correct or fiddled) does not help. Log in to reply |
ARS Strange Decisions 13/11/2024 at 17:43 #159162 | |
bill_gensheet
1431 posts |
If you want the trains to run under ARS, you can lie about the times on the branch, so the ARS calculation comes out in the correct way Bedlington 09:07 worked, looked to have undesirable effects on the crossing point for 2T12, as 2T12 came through first Log in to reply |
ARS Strange Decisions 13/11/2024 at 18:08 #159163 | |
Hap
1048 posts |
Thanks gents It would be best if we can look into the sim data first before we "foil" anything. As I said in my previous post, an issue has been spotted and needs to be looked at and compared to first before going forward. Finding shortcuts to alleviate it doesn't help getting round to fixing it. Thanks Craig How to report an issue: www.SimSig.co.uk/Wiki/Show?page=usertrack:reportanissue Log in to reply The following users said thank you: eps125, flabberdacks |
ARS Strange Decisions 13/11/2024 at 22:41 #159166 | |
chrisdmadd
255 posts |
Here is another example that seems even more strange but maybe is because of the same issue. Save attached. In the same save the issue occurs again with 3D01 and 5N07 at 05:40 at the same location. Post has attachments. Log in to view them. Last edited: 13/11/2024 at 22:45 by chrisdmadd Reason: None given Log in to reply |
ARS Strange Decisions 13/11/2024 at 23:15 #159167 | |
bill_gensheet
1431 posts |
chrisdmadd in post 159166 said:Here is another example that seems even more strange but maybe is because of the same issue.That one is down to the ARS logic using the departure time of 1Y05 as the 'be clear' for the conflict 5Y05 vs 3Y80 - so does the priority calculation on that basis and makes 5Y05 wait as the 'least delay'. Seen before in #40339 (exacerbated by other issues now fixed) and suggested this is a real bug in real ARS. Log in to reply The following user said thank you: chrisdmadd |
ARS Strange Decisions 13/11/2024 at 23:44 #159169 | |
Hap
1048 posts |
chrisdmadd in post 159166 said:Here is another example that seems even more strange but maybe is because of the same issue.In real simple terms, the headcode is in the wrong berth. Moving 3Y80 into 529 berth resolves it. Effectively, being at the London end it can still get to high level bridge to where it will conflict and that's why it's giving priority. After much collaboration regarding the original post, the conclussion is that the TT that's being attempted is not compatible with the sim era. Many changes have taken place on that line. The 2T is showing as conflicting at Benton and that's why 1S is being given priority. Again, that's just putting it simple. What we have to remember is that ARS is not perfect and this is one of the circumstances where in real life ARS is also not infallible. There's other real life situations where this happens, such as Man Vic for some splits to name an example. Thanks Craig How to report an issue: www.SimSig.co.uk/Wiki/Show?page=usertrack:reportanissue Last edited: 13/11/2024 at 23:44 by Hap Reason: None given Log in to reply The following user said thank you: eps125 |
ARS Strange Decisions 14/11/2024 at 00:06 #159170 | |
bill_gensheet
1431 posts |
Hap in post 159169 said:Oddly does not work like that for me. That was though moving the TD while paused, or adding before removing. The conflict is still "R532FM/R529C" If you interpose while running, and cancel 3Y80 first, then there is a brief window for 5Y05 to route with no 3Y80 TD in the sim to conflict with it. Log in to reply |
ARS Strange Decisions 14/11/2024 at 22:45 #159185 | |
chrisdmadd
255 posts |
Hap in post 159169 said:chrisdmadd in post 159166 said:Ah ok I did wonder if it would be related to the era. Thats fine I will ensure its all mentioned in the TT notes then. Only lazy people rely on ARS anyway right?!Here is another example that seems even more strange but maybe is because of the same issue.In real simple terms, the headcode is in the wrong berth. Moving 3Y80 into 529 berth resolves it. Effectively, being at the London end it can still get to high level bridge to where it will conflict and that's why it's giving priority. Thnaks Log in to reply |
ARS Strange Decisions 15/11/2024 at 02:33 #159190 | |
GeoffM
6380 posts |
This issue is actually multi-faceted: - Benton to Newsham is a couple of very short followed by a couple of very long block sections. This is known to throw ARS predictions in real life too. - Previous timetables had barely any trains going that way and may have been timed "better" so the issue wasn't spotted - The sim's track lengths/speeds/location positions on the branch may need checking (see also above). I wondered if adding Newsham LC into the schedule would help but I'm not convinced it's in the right place from a cursory glance so perhaps not. I note that trains going that way on this coming Monday have a lot more timing points. The real thing probably had the same issue and, amongst other reasons, adding timing points likely mitigated that. While we don't add new eras to existing sims usually, because it's a lot of work for rarely any recompense, I might consider adding Benton East Jn in which should alleviate the immediate issue. Not necessarily a bug per se, but simply "not enough information to make an accurate calculation". SimSig Boss Log in to reply The following users said thank you: mldaureol, chrisdmadd |
ARS Strange Decisions 15/11/2024 at 21:27 #159201 | |
chrisdmadd
255 posts |
Excellent thanks Geoff. I'll await your decision on that one. The timetable is quite exciting for a sim that's a lot of fun but with quite limited user created timetables. Thanks for the updates, I'm continuing to finalise the rest of the timetable for a release in December. (hopefully) Log in to reply |