I've read a few threads here and there, from what I gather you're supposed to select the track marker tab, pull down advanced, set the trigger radius and apply that to a junction spline. What I'm trying to do is the opposite of the usual, extend the radius to maximum (500 meters AFAIK) so the player train approaching the facing points will lock the switch in front of him, so an AI approaching from the trailing point side can't reset it until the player has passed. Tried what I THINK you're supposed to do with no effect, the switch is still throwable until the player engine gets within a car length of the points. This is TS2010 BTW, possibly it's different for other versions.
Junction lever radius
- Thread starter sniper297
- Start date
IsambardKingdomBrunel
Joined June 2001
Slap me around the head with a wet halibut, if i am wrong.
But i thought the advanced option for radius adjustment.
Only applied to track markers, track direction markers and triggers.
Mike.
But i thought the advanced option for radius adjustment.
Only applied to track markers, track direction markers and triggers.
Mike.
Lead us not into temptation!
Wet halibut sounds like fun, but you may be right. What I'm talking about is that zone where you approach the facing points and can no longer change them when the front coupler of the engiine gets within so many meters (furlongs hectares kilograms whatever it is). Some posts I've read say it's hard coded and can't be adjusted, others say there's a way to do it, difference is probably due to differences in Trainz versions if it's possible in any of them. Workaround right now is Trigger Multiple Signals, but if it is possible to set the mainline switches to interlock when one train is within 500 meters of the facing points it would simplify the whole setup.
Wet halibut sounds like fun, but you may be right. What I'm talking about is that zone where you approach the facing points and can no longer change them when the front coupler of the engiine gets within so many meters (furlongs hectares kilograms whatever it is). Some posts I've read say it's hard coded and can't be adjusted, others say there's a way to do it, difference is probably due to differences in Trainz versions if it's possible in any of them. Workaround right now is Trigger Multiple Signals, but if it is possible to set the mainline switches to interlock when one train is within 500 meters of the facing points it would simplify the whole setup.
ApproachMedium
Youtube dude
If you're both referring to what I think you are, radius of turnouts, you can shorten the distance required to be away from the points for the turnout to switch, by using the radius tool under advanced settings. Meaning whatever that unprototypical distance is that is set default, can be shortened. This allows for the driver (you) to stop almost on top of the switch point and still throw the turnout without derailing. I do this on all the routes I operate on, it makes it easier and more realistic to switch industries and yards.
-Joe
-Joe
IsambardKingdomBrunel
Joined June 2001
Joe,
is this change made using the same radius adjustment option.
In the advanced options screen, that appears under the Track Markers Tab.
Or something completely different ?
cheers,
Mike.
is this change made using the same radius adjustment option.
In the advanced options screen, that appears under the Track Markers Tab.
Or something completely different ?
cheers,
Mike.
From what I'm reading that's the one;
It APPEARS to work, if I go back later and use "Get trigger radius" clicking on the junction (switch/turnout) spline shows the number I set. Thinking maybe it had a limit and ignored numbers over the limit I tried setting 90 instead of 500, same thing;
About the default distance before the points lock. Maybe 20 is the max and this ain't gonna work at all, have to go back to Phil's method of using spacer signals.
It APPEARS to work, if I go back later and use "Get trigger radius" clicking on the junction (switch/turnout) spline shows the number I set. Thinking maybe it had a limit and ignored numbers over the limit I tried setting 90 instead of 500, same thing;
About the default distance before the points lock. Maybe 20 is the max and this ain't gonna work at all, have to go back to Phil's method of using spacer signals.
IsambardKingdomBrunel
Joined June 2001
Oh right it is that option Jim.
I tried it earlier and did not seem to do anything.
It does change the radius for track markers, triggers etc, though.
Mike.
I tried it earlier and did not seem to do anything.
It does change the radius for track markers, triggers etc, though.
Mike.
Yeah, using triggers to set an additional signal behind the actual trailing point signal is plan B, I thought this would be simpler and more reliable if it works.
Main problem is the AI logic itself is missing a gear - goes something like this;
Signal Stop - Check ROUTE SET
Route not set, throw switch.
Signal still red, block occupied by approaching player train.
Sit there fat dumb and happy keeping the switch locked so the route is set for AI, wait for player to clear block that he can't possibly clear because I'm keeping the switch locked the wrong way.
If one piece of conditional logic, "If block is not clear reset switch and wait two minutes before trying again" could be added to the AI then there would be no need for complex workarounds. Justin says the AI eventually releases the switch if you wait long enough, I haven't been able to duplicate that on any test route I've tried yet.
With my MSTS routes and Phil's Trainz routes people sometimes comment all the extra signals make it look more like the approach lights to 32 Left at O'Hare than a railroad, so the current project is trying to cut down the extra signals while still making the AI trains behave.
Main problem is the AI logic itself is missing a gear - goes something like this;
Signal Stop - Check ROUTE SET
Route not set, throw switch.
Signal still red, block occupied by approaching player train.
Sit there fat dumb and happy keeping the switch locked so the route is set for AI, wait for player to clear block that he can't possibly clear because I'm keeping the switch locked the wrong way.
If one piece of conditional logic, "If block is not clear reset switch and wait two minutes before trying again" could be added to the AI then there would be no need for complex workarounds. Justin says the AI eventually releases the switch if you wait long enough, I haven't been able to duplicate that on any test route I've tried yet.
With my MSTS routes and Phil's Trainz routes people sometimes comment all the extra signals make it look more like the approach lights to 32 Left at O'Hare than a railroad, so the current project is trying to cut down the extra signals while still making the AI trains behave.
IsambardKingdomBrunel
Joined June 2001
Use the "Release Junction" command. That should allow the player train to continue.
Mike.
Mike.
Already tried that one, either it don't work as advertised or I'm not using it right. You CAN bring up the interface and find the driver of the AI holding the switch, give him a STOP command, then he releases the switch so you can drive past, then give him the CONTINUE SCHEDULE command. That works even without using the Release Junction rule, so I don't see the point of that particular command set.
Last edited:
norfolksouthern37
what's a normal map?
same problem i see,
as i said in the other threads the AI HAS to change the junction to figure out if he can go. if you had not set the junction before getting to them then he would know you are there and not grab the junction. to the AI the closed junction looks like the end of track, but it knows there is a junction, so it changes it to read the track in front of it. in the case you present, the AI will let go of the junction once he knows he cannot proceed where the player train is. - in the post i missed i see you understand this.
by setting the junction for your path before you actually reach it, the AI thinks there is an alternative by changing that junction. this is why it locks it, but i assure you it is temporary.
oops i missed a post up there.
all that i do when testing this scenario is keep approaching the junction. it always releases before i get to it or even the signal guarding it and lets me pass. AI vs AI is no problem at all. again, an increased distance between passing sidings should take care of this problem, but on a small layout id expect a wait while the AI figures its tracks.
did you try making the path rule work?
as i said in the other threads the AI HAS to change the junction to figure out if he can go. if you had not set the junction before getting to them then he would know you are there and not grab the junction. to the AI the closed junction looks like the end of track, but it knows there is a junction, so it changes it to read the track in front of it. in the case you present, the AI will let go of the junction once he knows he cannot proceed where the player train is. - in the post i missed i see you understand this.
by setting the junction for your path before you actually reach it, the AI thinks there is an alternative by changing that junction. this is why it locks it, but i assure you it is temporary.
oops i missed a post up there.
all that i do when testing this scenario is keep approaching the junction. it always releases before i get to it or even the signal guarding it and lets me pass. AI vs AI is no problem at all. again, an increased distance between passing sidings should take care of this problem, but on a small layout id expect a wait while the AI figures its tracks.
did you try making the path rule work?
Last edited:
IsambardKingdomBrunel
Joined June 2001
So try setting the AI Priority to 2 or 3 and the player train to 1.
Mike.
Mike.
norfolksouthern37
what's a normal map?
i am getting this in 2010 on the route and session one of you guys sent me where you start at the top of the hill with the UP sd40 and box.
IKB might have an idea there, try it.
IKB might have an idea there, try it.
Tried that, tried that too
In various combinations, release junction rule, give the AI driver a release junction instruction for every trailing point he'll pass on the way to the destination, set player engine priority to 1. Near as I can tell AI trains spawning from portals default to priority 2 and there's no way to change that, either way they set and lock the next switch ahead. Which means plan B also fails, don't matter how many signals are between him and the switch as soon as he passes the previous switch he sets the next one.
So plan C, variation on the fake/invisible switch idea, use real switches in the middle of each doubletrack section, or possibly near the ends. Spurs or yards coming off the doubletrack or a single crossover going opposite the direction of travel, the idea being to get one more trailing point switch between the triggered stop signal and the exit from the single track that he keeps locking. Obviously requires some planning, each passing siding ("loop") needs to be at least twice as long as the longest AI train used, and if the AI has already passed the mid switch there needs to be some way to give a stop indication at the other end. More triggers, I was hoping for less complication rather than more. Main thing is solutions that work on small routes will be too tedious to set up on big ones, which is why I'm using track direction markers for right hand running instead of all them "navigate via trackmark" things.
In various combinations, release junction rule, give the AI driver a release junction instruction for every trailing point he'll pass on the way to the destination, set player engine priority to 1. Near as I can tell AI trains spawning from portals default to priority 2 and there's no way to change that, either way they set and lock the next switch ahead. Which means plan B also fails, don't matter how many signals are between him and the switch as soon as he passes the previous switch he sets the next one.
So plan C, variation on the fake/invisible switch idea, use real switches in the middle of each doubletrack section, or possibly near the ends. Spurs or yards coming off the doubletrack or a single crossover going opposite the direction of travel, the idea being to get one more trailing point switch between the triggered stop signal and the exit from the single track that he keeps locking. Obviously requires some planning, each passing siding ("loop") needs to be at least twice as long as the longest AI train used, and if the AI has already passed the mid switch there needs to be some way to give a stop indication at the other end. More triggers, I was hoping for less complication rather than more. Main thing is solutions that work on small routes will be too tedious to set up on big ones, which is why I'm using track direction markers for right hand running instead of all them "navigate via trackmark" things.
Sorry to jump in the middle here, but I found in testing that sometimes it is the signal that makes a difference. I believe it is how the script is written (although I may be completely wrong as I have not been able to find any info on scripting or configuration files for Trainz).
I had this exact scenario happen to me with one of the routes that I made and it drove me nuts! I happened to find some signals that I liked and decided to replace all my signals and the problem was cured.
I then made a test route and was able to duplicate the issue using the "bad" set of signals.
So might I suggest that you take a look at the signal for the AI and replace it with a different type to check and see if this will cure your problem. Please let us know if it works. I am very curious to see if we may have had the same issue.
Thanks.
IsambardKingdomBrunel
Joined June 2001
If you place a track marker in session mode ( not added to route only session ).
Just before the sig that is protecting the end of the passing loop.
You can then issue a navigate to track mark command, followed by a Wait For xx.xx mins command at the track mark.
Mike.
Just before the sig that is protecting the end of the passing loop.
You can then issue a navigate to track mark command, followed by a Wait For xx.xx mins command at the track mark.
Mike.
"take a look at the signal for the AI and replace it with a different type", yeah, still trying to figure out what's what in all these, the stock USA 3 color light and searchlights are the only ones I've seen so far aside from experimental ones.
Mike, that would work except it's timing dependent - he pulls up to the marker, stops, waits two minutes then throws and locks the trailing point, if the player happens to be on the single track stretch when the 2 minute wait time is up he still gets the switch locked in his face. Helluva way to run a railroad. :hehe: What would work with that is some type of distant signal or repeater that would let the player know that the opposing AI is in that block, so he wouldn't enter the single track section at all unless he got a clear indication for the opposing double track block. Haven't seen a US version of distant or repeater signals, the ones I have seen are on the left side of the track and don't come with instructions anyway.
Mike, that would work except it's timing dependent - he pulls up to the marker, stops, waits two minutes then throws and locks the trailing point, if the player happens to be on the single track stretch when the 2 minute wait time is up he still gets the switch locked in his face. Helluva way to run a railroad. :hehe: What would work with that is some type of distant signal or repeater that would let the player know that the opposing AI is in that block, so he wouldn't enter the single track section at all unless he got a clear indication for the opposing double track block. Haven't seen a US version of distant or repeater signals, the ones I have seen are on the left side of the track and don't come with instructions anyway.
norfolksouthern37
what's a normal map?
that kinda defeats the point of the passing sidings doesn't it? it basically puts the player train at the lowest priority too.
Similar threads
