The International Simutrans Forum

 

Author Topic: Track circuit signal doesn't create directional reservation at specific location  (Read 735 times)

0 Members and 1 Guest are viewing this topic.

Offline Ranran jp

  • *
  • Posts: 353
  • Languages: ja
Track circuit signal doesn't create directional reservation at specific location.

Check this saved game out.
Directional reservation is not created correctly near points k and g2.
(g2 works normally when it goes to north)








I tried to solve this problem in various ways.
- build the station
- put more signals
- change the number of aspects of the signal
- change the signal to bidirectional signal which is at the passing loop

However, the only way that I could solve this was to move the signalbox to another location...
Am I something wrong? (´・ω・`)

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18425
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Thank you for your report - I think that I have managed to fix this. I should be grateful if you could re-test with the next nightly build.

Offline Ranran jp

  • *
  • Posts: 353
  • Languages: ja
Thank you for working on fixing this issue.
I tested running the multiple trains on this line with today's nightly build. As a result, the train frontal collision occurred.(´・ω・`)
I looked for the cause of this, the signal does not create a directional reservation in certain situations.
Although the issue of not making a directional reservation was solved, a directional reservation may end in the middle.


For example, as shown in the following picture, A signal intermediate between i and H does not make a direction reservation when the reservation start signal is k.
This train passes through the station i, so if another train comes from the other side at this time they will collide there.




the other examples:





Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18425
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Thank you for testing. Are you able to upload a saved game in which the collision occurs at a specific place and time?

Offline Ranran jp

  • *
  • Posts: 353
  • Languages: ja
Quote
As a result, the train frontal collision occurred.
I'm apologize. This seems there was a mistake in my signal placement.
By using the two aspects signal instead of the three aspects signal so that the red reservation does not across the passing loop, collision did not occur and it worked correctly.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18425
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
It does not seem to be correct that you need to use only two aspect signals here - I should be grateful if you could upload the saved game demonstrating the collision so that I can try to fix this.

Offline Ranran jp

  • *
  • Posts: 353
  • Languages: ja
Okey. Then, please check this saved game.
I think that this collision can be avoided if it is 2 aspect signal.

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18425
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Thank you for the report/reproduction case: I believe that I have now fixed this. I should be grateful if you could re-test with to-morrow's nightly build.

Offline Ranran jp

  • *
  • Posts: 353
  • Languages: ja
I confirmed that it has been fixed in today's nightly build(#0088b46).
Thank you for fixing this. (´・ω・`)b

However, if there is no problem even if it crosses the passing loop, it is a little concern that the directional reservation indication is interrupted in the middle of the single line section as shown in some pictures above(reply #2), but these blank section can not be reserved from the opposite direction, or the signal can cancel the reservation from the opposite direction.
Therefore the train never collides in their section. That is only display issue.

For example, it does not collision even in a state like this picture. (This is very interesting!)

Offline jamespetts gb

  • Simutrans-Extended project coordinator
  • Moderator
  • *
  • Posts: 18425
  • Cake baker
    • Bridgewater-Brunel
  • Languages: EN
Thank you for confirming. The single tile gap should not create issues as there is no way in which this can ever be reserved, so this can be regarded as fixed.