Portal Spawning Test 02

[Chacal]

This may be an artefact of the scenario editor itself. Try opening the scenarioproperties.xml file manually and see if you can find those entries, put them in the correct order if needed.

[OldProf]

This may be an artefact of the scenario editor itself. Try opening the scenarioproperties.xml file manually and see if you can find those entries, put them in the correct order if needed.

That kind of manipulation isn't within my abilities, but thanks for the idea, anyway. I've shelved the Castle Rock experiments while I wait for inspiration to strike, but in the meantime I've started to test portals elsewhere, specifically, Hedborough North. This route has only a few portals, all placed at the ends of a section of double mainline than runs past the Hedborough Yard. Whereas on Castle Rock I was experimenting with portals that spanned two tracks, at Hedborough each of the two mainline tracks has a separate portal at each end. I hope to upload my next experiment soon. Many thanks for your suggestions and encouragements.

Regards,

[Chacal]

Tom, please post your scenario file and I'll try it for you.

[OldProf]

Tom, please post your scenario file and I'll try it for you.

I'm hesitant about uploading a scenario that cannot be run in its present condition. Is that really okay?

[OldProf]

Fresh tests on my portal experiment scenario on the Hedborough route [note: three southbound AI trains pass the Hedborough Yard, the first two pass without stopping, but the third reverses into Hedborough Yard, makes a drop-off, and then exits the yard; all three exit through the same portal]:

Case 1. All three trains stored in portal4 (covers only one track) in order of their exit: Service 3, Service 2, Service1. After save, scenario fails to load because Service 3 cannot find drop-off track. When reopened for editing, the order of trains stored in the portal proves to be Service 2, Service 3, Service 1.

Case 2. Service 1 removed from portal4 and services 3 and 2 stored in portal4 in that order. After save, scenario fails to load as in case 1. When reopened for editing, order of trains stored in portal is Service 2, Service 3.

Case 3. Services 1 and 2 removed from portal4 and placed on southbound track in proper departure order; Service 3 left in portal4. Scenario fails to load as in cases 1 and 2.

Case 4. All three trains removed from portal4 and placed on southbound track in correct departure order. Scenario opens and runs correctly.

Current Conclusion: An AI train instructed to perform a drop-off (possibly applies to pick-up as well) cannot be stored in a portal and "spawned" from it.

[Accurate]

Yep, that is true. If you approach the portal function with a scenario and get all excited to watch it work the failure can be very confusing. You have broken the secret error sequence. Perhaps an error report to RSC will get it fixed.

[OldProf]

Accurate:

Actually, RSC advises against using portals to emit trains, so they, quite rightly, would not see this as an error. I'm one of the last of the hardheads who like to saw against the grain on this matter.

Regards,

[Chacal]

Tom, please post your scenario file and I'll try it for you.

I'm hesitant about uploading a scenario that cannot be run in its present condition. Is that really okay?

I think yes, seeing that it is a test scenario.
Anyway I'll just PM you my e-mail address.

[OldProf]

My portal test scenario for the Hedborough route has now reached version 5b and I've arrived at some interesting conclusions, which perhaps should really be called theories.

As previously mentioned (and now, in my opinion, confirmed) an AI train that emerges from a portal cannot be instructed to perform a drop-off if other AI trains are scheduled to emerge from the same portal.

Once I established this, I took the dropping-off train out of its portal and moved it to another part of the route, from which it could reverse into the Hedborough yard, make its drop-off, and return whence it came (more or less). Why reverse? Because the player engine already uses the yard's Shunt track and the AI train could not enter it, even when the player train was not present. Perhaps a scenario-based track marker would resolve this situation, but I haven't tried that. Compromise is often necessary in scenario writing! Originally, I used only a drop-off instruction followed by a few drive-to instructions to get the AI train into and out of the yard area, but after dropping its wagons on the assigned track the engine did something (I could never see what) that caused a scenario-ending AI collision. So, I added a drive-to instruction before the drop-off instruction, with the result that sometimes it crashes and sometimes it doesn't. The length of the stop involved in the drive-to instruction seems to have something to do with this. I learned to save the scenario before the drop-off and continue it from the save after the collision. The collision did not occur after the continuation, but something else did.

By that point, I had 4 AI trains loaded into each of 2 portals. When the scenario ran without the drop-off AI crashing, all AI trains came out of their portals correctly, but when I had to continue the scenario after a collision, only 2 of the 4 AI trains emerged from each portal.

I'm still testing thie latest version, but will upload it as soon as possible.

"Curiouser and curiouser", as Alice said.

Latest news: This scenario, named Hedborough Portal Test has been uploaded and should be available for download soon. Please give it a try!