Unexplainable signal behavior

I'm nearly finished testing the latest (and greatest) activity yet,
but I have run into a puzzling issue.The activity is on MP3.1. Player starts at Kalispell, and eventually ends the run at Shelby. Near the end of the activity, as the player train reaches a signal between marker 1069 and 1068 I encounter a red stop signal which seems bound and determined to stay that way. The odd thing is that all opposing AI trains which started at Shelby have long since passed by me, and AI trains which were ahead of me at some point, were deftly terminated out of sight.Now... there no longer seems to be any reason for this stop signal ahead, but there the stubborn thing is. There is a complex crossover switch pattern at this point, but no traffic left on the map but me.
I seem to be locked out of the Shelby yard. I thought about toggling the signal failed, but I'd really like to know what is provolking this. Anyone got any ideas?Perhaps someone has seen something like this before. I did use the editor to peel back the end point of the player train and reconstruct the final part of the path in question to see if this would snuff out the quirk, but it will take another 8 hours to test it.Any insights?

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: unexplainable signal behavior, yardlimit, Sep-22-04, 09:02 PM, (1) RE: unexplainable signal behavior, NewEnglandRailFan, Sep-22-04, 09:44 PM, (2) RE: unexplainable signal behavior, yardlimit, Sep-23-04, 09:10 PM, (3) RE: unexplainable signal behavior, andyknott, Sep-29-04, 07:20 AM, (4) RE: unexplainable signal behavior, yardlimit, Oct-02-04, 09:13 PM, (5) RE: unexplainable signal behavior, andyknott, Oct-04-04, 03:10 AM, (6) RE: unexplainable signal behavior, yardlimit, Oct-05-04, 10:27 PM, (7) RE: unexplainable signal behavior, yardlimit, Oct-11-04, 12:51 PM, (8) RE: unexplainable signal behavior, yardlimit, Oct-11-04, 08:15 PM, (9)

Conferences | Topics | Previous Topic | Next Topic

I may have answered my own ques. I remapped the path for the final few miles as I pointed out. Then I set the start point and time well forward and ran it. The signal issue seems to have cleared up.
Must have been a path puke. God only knows...

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

I had this happen once. Did your path end at the very end of the track? thats what the problem was for me.-Dave

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

No, not quite. I've read about how that can cause a problem. I try to keep the end point marker within boundaries the program can recognize without hiccoughs. I've learned to deal with passing paths, head to head stand-offs, and the like, but I've never encountered a stubborn red with no trains except mine active in the path. Just didn't make sense. I'll tell you though, no one will ever learn everything they need to know about this wacky editor.

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Guys,I was hoping someone would come up with the definite answer to this as I was cussing at the PC last night.Running the train on its own, it goes the full length of my double track main, but when the train follows close on the heals of an AI train, it stops at a red and won’t proceed even though all other traffic has finished. I solved (fudged) the problem by lengthening the gap between trains. This has somewhat spoiled the activity but it now works, but I am still interested if someone can answer this problem.
Andy
UK

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Oct-02-04 AT 09:26 PM (EDT)

Andy:I had to deal with a lot of this sort of thing when I re-worked "Second Section" for KHP 2.0. There is no one fomula to solve these kinds of problems. You must resolve it on a case by case basis. In "Second Section" one follows a long frieght similar to the player train some 70 miles. Occasionally, I would actually see the first train in the distance. Over that route you could follow through on a "red" at a resricted speed of 15mph. The tricky part was not so much the schedualling of the train I was following, but the careful formulation of passing paths for oncoming AI trains. These were the cause, more than anything of inexplicable stops. Much would change from one run of the activity to the next depending how the speed and efficiency of the player train was handled. The key was not allowing the lead AI train to get in conflict in a passing path with an oncoming AI train. I had to map a distinct and separate path for EACH of five trains, paying particular attention to passing points, even though perhaps several of them might have used the same path with different departure times. Only then did the darn thing begin to behave itself. I am convinced that what we don't know about that editor could fill volumes. I'll also add that I have read that the simulator algorithm uses a formula which evaluates conditions for AI trains and the player train some 3 boxes forward. It does a fair job, but it can get befuddled. Increasing your train interval can obviously diminish your waits at reds, or eliminate them altogether. Sometimes you can actually make better time following a train by dramatically reducing speed from what would normally be permitted. It sounds screwy, but it seemed to help diminish lengthy waits, thereby increasing overall average speeds.

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

I was fighting a loosing battle with the PC again last night. As you say every fix is different.
I am sure there is a relationship between “stuck at Red” and the number of signals on the route. The idea that MSTS works 3 signals ahead may add weight to this.When I get the time (never?), I intend to create a test route of a single track with one passing loop to test the various activities. I can then try adding signals to see if there is a pattern to the “stops”.Don’t we just love Microsoft.Andy UK

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LAST EDITED ON Oct-05-04 AT 10:29 PM (EDT)

Andy:I recently tested my activity, and encountered the red signal problem again. I thought it had gone away. I began to dig into the AI train paths coming out of Shelby. My instincts told me PATHS might be involved. I don't have a firm answer yet, but I did find something bizarre. I had two AI freights using a common path (same name) starting out of Shelby at two different times --but they had entirely separate starting points. How the hell I accomplished that contradiction; I have no idea. That should not be possible. In any event I reconstructed the player train end point, and also the screwed up AI freight paths. This may have been the source of the anomalie, but I won't know for a day or two. stay tuned...

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Still slaming into an incorrigable red between mm 1068 & 1069 in mp31.
In spite of any attempt at a fix. I redid the entire path. no small feat. starts in kalispell, reverses at two points, goes to Whitefish, drops off freight, goes through loop tunnel (another revers pt) and proceeds to Shelby with lots of pickups on the way.
The editor seems to digest the path cleanly, but clearly I am doing something that convinces it that there is a train ahead of me, or a train coming at me on both sides of a dual track box -at a time when there should be no trains present. --whew

Yardlimit

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Tremedous new insight!A lot of experienced editor hacks are going to laugh at this and call it ignorance, but thats ok.Here is why I was stuck at an inexplicable red, and I'll bet a lot of designers besides me dont know this. In spite of the fact that I had the end point of the player path well within the boundaries of the Shelby yard, the activity code did not recognize this, and pass this knowledge effectively on to the program code.First clue. I eventually noticed that the Shelby station was not listed in the path box along with the other stations I passed through. Etheridge was the last station recognized; the last berg before arriving at Shelby. So what was the deal with the red light?
As close as I can figure, it was the first signal beyond Etheridge that was no longer located within the recognized boundaries of the acivity. Consequently, it was not going to respond to activity conditions. It was not on the map as far as the running code was concerned. The solution. Evidently in the Shelby yard, and perhaps many other yards in many other routes, one has to place the end point rather deep into the yard before the code recognizes that this is the legitimate end point on the path. In the case of Shelby, the termination point must be well past the passenger platform which is damn near to the end of the yard. I had no clue. This might also explain why an AI train may get hung up somewhere in its path without explanation, but these things have to be examined on a case by case basis. Some people will get a horselaugh out of this, but I don't care. I'm just glad enough to finally understand what the %@$# is going on.And don't imagine the editor, for all its skills at choking on every insignificant piece of debris, will give you a clue your end point lies in "never-never" land.

Yardlimit

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | How To... | File Library ] Copyright © 2004 by Nels Anderson. All Rights Reserved.