Problems with Dearnby route

ATLS works OK in TRS19 although I had to rename all the bits on my Ffestiniog Welsh Highland route, which means it probably needs an update for TRS19 just to fix the ATLS.
Just a thought, you are using the latest v2 version of Dearnby? if not that may explain the issues with the red bugs, I can clear that superscript screen here and no red bug.
Also wondering if turning traffic off might affect the DMT errors.

Using the latest version of Dearnby.

I tried deleting the file that HPL noted. Then I did a Database Repair. Still getting pauses around the 28 minute mark. I'm trying an extended database repair now.

When I start the route, that superscript box pops up and I don't really know what to do with it, so I click on clear script and minimize it. I am still getting the red bugs, and for a while I was just clearing all the session logs, but then they became too long, too frequent, and too many, so I stopped trying. Not too long after that, I was beset by pauses.

I noticed that many of the crossings were still working, but snapping open at the last minute.

Not sure what else I can do with this. Still open to suggestions. Hopefully PGuy has some input when he gets a chance to look at this.

We are currently working (slowly but surely) on the next version of Dearnby & District. We'll certainly check all the level crossings and if necessary take out any types that are not working in TRS19. Thanks to everyone that has been looking at this issue.

Neville_hill,
Dearnby & District is absolutely a classic. I'm not sure how you can make it better, but I'll eagerly look forward to V3.

Now if I could only figure out how to make V2 stop pausing...

Heinrich505

Yes, I'm using V2 of the Trainspotter session (kuids in post 1). I don't get any more superscript red bugs after the start - although looking at the superscript messages they do seem to be all associated with BR guv luggage vans so it might be possible to get rid of those vans. I do get loads of red bugs after about 10 minutes, all from DMT/TRC errors.

Perhaps converting to ATLS is a bit too ambitious - I'm not sure that the DMT scenery assets could easily be done in ATLS.

Anyway, it's very good to hear that Neville Hill is working on the next version, so I'm very much looking forward to that. Meanwhile, I'll look at where the level crossings actually are, and see whether disabling some of the narrow guage trains will mean that the level crossings are never used - at least that way I can get on with playing with the route. I'm interested in using Enhanced Interlocking Towers, though I shall probably need to go back to v22 of the named objects library for that....

Peter

Version 3 is in my possession at the moment; I'm putting in many signals throughout some new sections, including extensive single lines. Sessions will be using junction locking & unlocking to restrict single line access to one consist at a time. Testing has got it all working apart from one mysterious wee glitch.....

I notice that the level crossings are not usually working (not opening & closing when they should) so that traffic drives through trains. Obviously they are all ghosts as nothing crashes. Sometimes a crossing decides it will work. Who knows why? I've tried alternative crossings here and there (mostly very old assets from several Trainz versions ago) and they do work as they should. I'll ask Neville if he finds them acceptable quality-wise. It'll only take 3 years and 7 months to replace them all. (Kidding)!

My TRS19 Build is 105100 (the original Platinum Build). Apart from the level crossings, I seem to have no problems with freezes, red bugs or anything much. I get crashes to the desktop but very rarely .... seemingly at random. Often I'll run the same session in the same way and it'll go for hours without a problem.

TRS19 is far too variegated i' the Build numbers; and possibly so complicated now that bugs arise spontaneously (and also disappear in the same fashion) because the complexity has become chaotic (in the mathematical sense). The program leaps from one strange attractor to another, the rascal.

Still, Version 3 will appear eventually. But that Neville! - he already has a Version 4. with miles and miles of new track. Can Trainz cope with over one million junctions and signals in a single route?

Lataxe, oscillating between Surveyor and the woodworking shed.

Some of the crossings in particular one of the Narrow Gauge ones are being kept open by a junction in close proximity, just needs changing then it works properly on the main NG line. Cant remember which one but one of the DMT crossings is actually working properly, think it was an angled one, which makes me think the problem may be fixable.

neville_hill said:

We are currently working (slowly but surely) on the next version of Dearnby & District. We'll certainly check all the level crossings and if necessary take out any types that are not working in TRS19. Thanks to everyone that has been looking at this issue.

Click to expand...

I'd hold on and give Pierre a chance to sort out the problem with version 22 of Named Objects Library. I'm sure these issues are all library / script related, which is also not doing a lot for frame rates!

Just to help avoid going down any rabbit holes, I would point out that many of the issues referenced so far (red bugs, crossings not always behaving) have always been present, for me at least, in both TANE SP3 and SP4, and for both versions of the excellent D&D route / sessions. So I don't think it is necessarily anything specific to TRS19.

I have heard from Pierre that he has run the route and session as in the first post, under the new 109641 build, with only the minor initial problems re superscript and rolling stock. He was using V23 of the Named Object Library, which he says is the same as v22 apart from a fix for station search (I've asked him to send me a cdp of v23 just to be certain.)

I have also run under a "clean" installation of 109641 (full install of Trainz from scratch, downloading only the Dearnby route and session). This used V22 of the NOL, and I had no lockups, but several hundred TRC errors. Also after 1 hour running nothing much seemed to be happening at all which was weird. I need to look further into it, but it's all getting very frustrating...

Peter

Locked up here in 109641 with v22 after 59 minutes started a couple of stutters around 35 minutes. Now thinking it may be due to something that happens time related, says in the description "takes between 30 - 45 minutes to become fully operational".

Second run started locking up at 30 minutes.

Crossing gates appear to be working however traffic is stuck.

Getting very confusing.

Clam1952,
Running beta 109037 for TRS19 with the :18 version of named objects modified to :22, my lock-ups start right around the 28 to 32 minute mark. They are quick at first, and I don't know if deleting the little red bugs as soon as they appear has any effect on the lock-ups. It seems that it doesn't have much of an effect if I try quickly deleting them or if I simply ignore them from the very start. The lock-ups then quickly increase in length and frequency to the point where I have to stop the session.

I have noticed that most of the gates seem to be working, and I think traffic is also working.

The earliest portal starts emitting trains at the 9 minute mark - another starts emitting them at the 11 minute mark. Each set of portals has a different time when they start emiting consists. Several of the portals accept some of the small commuter engine consists that start on the map, and then re-emit them after a 5 minute delay.

There have been problems with portals ever since TANE, and the time-outs caused by the portal "problems" was what wrecked the consists for the CMTM system. I'm wondering if they could be part of the problem here.

Heinrich505

Good news! Pierre has been looking at the DMT/TRC errors, and has found a script problem. He has produced a version of the 5 DMT/TRC assets involved (the ones which figure in the error messages), and this works. Running build 109641 twice, for 38 and 30 minutes, I did not have a single DMT message.

He is also investigating the lockups - he thinks the ACS and Superscript systems may be involved.

Running with 109641 earlier with my "clean" system, and version 23 of Named Objects Library, I didn't get any lockups until after 105 minutes, but I think this might be because there were several drivers missing, and so far less traffic than normal.

Peter

PWJohnson,
Yikes, this is good news indeed! Please keep us posted. There is hope on the horizon.

Heinrich505

The superscript errors bug screen find <kuid2:116296:37234:3> BR guv luggage van, open for edit and delete the debug line from the extensions Tab problem solved.

I've just run the session for 2 hours in the now released Trainz plus 109641 update today, no pauses! One or two of the consists got a bit confused near the 1.5 hour and wre effectively lost miles off thier route and going backwards and forwards trying to find a path but other wise seems OK apart from the odd crossing gates being too quick to shut and too slow to open and a couple of traffic jams. It would seem to mainly relate to the positioning of the track marks and the radius of their effect which can be changed.

Malc,

Brilliant! I tried your fix on build 109641 and was also able to run for 2 hours with no lockups.

There were a few other funnies: the 5 level crossings between Ullsmere - Casterford - Dearnby Jct all refused to change so each train that I observed had to drive through the gates... I have seen these gates work Ok in the past so no idea why they shouldn't work now. And also, after about 80 minutes, I noticed about 8 locos all together on one of the 4 lines near Dirncombe. The big station at Dearnby on Sea was very quiet towards the end of the run - perhaps beacuse of the loco pileup? There were a few red bugs; 1one at the start mrelated to Superscript (GetBogieList), the one about the Blue Pullman car F, and one about a Dance Hall Brake

So there are still some problems, but that's one great step forward.

I was running with Pierre's DMT/TRC fix so there were none of those error messages (last I heard Pierre was trying to contact scottish, who owns those assets, to see how to make this generally available). I also had V23 of the Named Object Library, and ACS and Superscript libraries enabled.

Peter

pwjohnson said:

(last I heard Pierre was trying to contact scottish, who owns those assets, to see how to make this generally available).

Peter

Click to expand...

I've pm'd Pierre with his email info as it's not in his configs and haven't seen him on the forum for some time so probably not checking pms.

Clam1952,
Most Excellent!!! :clap: I tried your fix regarding the luggage van. For TRS19 beta 109037, I passed the 1 hour 17 minute mark with no lock-ups. The wife had dinner fixed by then, so I closed down. Since I have not made it past the 35 minute mark before locking up and having to close the session, I'd say that the fix worked for this version of TRS19 as well. What a cool catch! Nice detective work.

There was a holy row going on in Westacott by then, as one of the locals had gotten stuck while trying to run around his consist and was blocking several trains heading in that direction. The stationmaster was all in a tither, especially since the King George V was being held up, and that was a really bad situation. He ordered me to delete the local engine and consist too, so I did. Then traffic started moving again and things sorted out nicely.

I crashed through all the crossing gates across the rails on the way to Casterford, as they wouldn't work at all. I also drove through a gate in Badgercomb too, but it seems the others were working okay.

When I came back after dinner, I ran the route once more and this time, every crossing gate on the Casterford run was working perfectly. Go figure.

I'm still getting a lot of session log red bug errors. I'm not sure if I need to keep deleting them as they come up, or just ignore them. If I ignore them and they keep building up, will that eventually cause a problem with the session operating?

I've no clue what the session logs are blathering about. It seems that each time I run the session, there are different kuids that are coming up bugged. For instance, there are two for the BR Standard Tender BR1B-Lined Black, Late Crest. I marked the kuids as bold.

- <kuid2:648132:100748:1> : GameObject::Sniff> Target object is null (file gs.gs)
; <kuid2:648132:100748:1> : Script class: lilbsloco
; <kuid2:648132:100748:1> : MeshObject: <kuid2:648132:100748:1> "BR Standard Tender BR1B-Lined Black, Late Crest"
; <kuid2:648132:100748:1> : Script callstack
; <kuid2:648132:100748:1> : function $void@GameObject::Sniff(GameObject,string,string,bool), line -1
; <kuid2:648132:100748:1> : function $int@LilB_Loco:etermineCarPosition(), line 300
; <kuid2:648132:100748:1> : function $void@LilB_Loco::ChangeMessageHandler(Message), line 4553

- <kuid2:648132:100763:1> : GameObject::Sniff> Target object is null (file gs.gs)
; <kuid2:648132:100763:1> : Script class: lilbsloco
; <kuid2:648132:100763:1> : MeshObject: <kuid2:648132:100763:1> "BR Standard Tender BR1B-Lined Green, Early Crest"
; <kuid2:648132:100763:1> : Script callstack
; <kuid2:648132:100763:1> : function $void@GameObject::Sniff(GameObject,string,string,bool), line -1
; <kuid2:648132:100763:1> : function $int@LilB_Loco:etermineCarPosition(), line 300
; <kuid2:648132:100763:1> : function $void@LilB_Loco::ChangeMessageHandler(Message), line 4553

For the first run, after your amazing fix, these were the only two red bug script log errors. I thought maybe the fix for this was to include the kuid in the list of kuids at the bottom of the config.sys files for each kuid, as they weren't showing there. But, the second time I ran the session, after dinner, the 100748:1 error started showing up again.

I'm seeing different red bugs marking different kuids for each time I run the session, but I guess this is because the consists and engines are random every time. I did see a lot of script log errors for the DMT UK Self Build Restricted Traffic Zone 12 TRC as well as the Blue Pullman the second time around.

As noted above, can I simply ignore these script errors or should I be deleting the little red bug every time it pops up?

Once more, thanks so much for finding the fix on the lock-ups. That is so cool - now I can run the session without worrying about lock-ups. You are THE MAN!

Heinrich505

The tenders are not creating red bugs here. According to the configs they are using <kuid2:103021:100525:9> LilB\'s Script Library v1.4

The builtin version in TRS19 SP2 is <kuid2:103021:100525:21> LilB's Script Library v1.4.5.6 which is installed here.
and there is an obsolete builtin installed <kuid2:103021:100525:20> LilB's Script Library v1.4.5.6 also installed so its not using v9 in my install

I'd check and see just which version you have, 20 and 21 may have arrived with DLC, 9 to 19 are on the DLS.

Check the builtin isn't showing locally modified or open for edit, if it is revert to original.

If 9 is installed or any version other than the builtins try deleting them?

Clam1952,
I just checked and I'm showing <kuid2:103021:100525:25> LilB's Script Library v1.4.5.6 as modified. It is not open for editing.

Version :20 shows available, obsolete, and :21 shows modified, obsolete. :22, :23, and :24 are showing as unknown asset.

Unsure as to how to proceed on this. If I just ignore the red bugs, does the session log keep getting so long as to affect performance of the session?

Heinrich505