ACOA right before passing the signal goes undetected

Three Bridges T323 propagates, T322 is immediate on T326 and T322.
Wembley 125 and 149 propagate, 116 and 344 are immediate

bill_gensheet in post 136735 said:

Three Bridges T323 propagates, T322 is immediate on T326 and T322.
Wembley 125 and 149 propagate, 116 and 344 are immediate

Don't forget some of these signals have combined overlaps. The train can still be in the track section but past the signal.

GeoffM in post 136749 said:

bill_gensheet in post 136735 said:

Three Bridges T323 propagates, T322 is immediate on T326 and T322.
Wembley 125 and 149 propagate, 116 and 344 are immediate

Don't forget some of these signals have combined overlaps. The train can still be in the track section but past the signal.

For the avoidance of doubt, these comments refer solely to checking out the propagation of aspect changes, I did not test Three Bridges or Wembley Main for not calling an ACOA.

If it is as little as ~1 second, is that not near the cycle time for SimSig to recalculate things ?
[...]
Will try some experiments looking around the 20m Peter has suggested.

bill_gensheet in post 136751 said:

GeoffM in post 136749 said:

bill_gensheet in post 136735 said:

Three Bridges T323 propagates, T322 is immediate on T326 and T322.
Wembley 125 and 149 propagate, 116 and 344 are immediate

Don't forget some of these signals have combined overlaps. The train can still be in the track section but past the signal.

For the avoidance of doubt, these comments refer solely to checking out the propagation of aspect changes, I did not test Three Bridges or Wembley Main for not calling an ACOA.

Now I am very confused. You say X happens on 3B, then say you didn't actually test 3B.

???

[quote=bill_gensheet;post=136706]If it is as little as ~1 second, is that not near the
The logic involved in an ACOA happening or not happening involves various different bits of code that aren't all run at the same time. So anything within one second of the train reaching the signal is not a bug, it's just a fact of life. Think of it as the driver looking at one of the instruments at just the wrong moment.

If the ACOA happens two seconds before the train reaches the signal, it should be reported. If it happens two seconds after the train reaches the signal, it should not be reported. If either of those are violated, that's a bug. Anything that happens in that 4 second window is going to be marginal and I'm not going to spend weeks trying to adjust the code to deal with it without breaking any of 50 other features.

The initial post was that either the train should ACOA at SIG1 (say for seeing G to Y) OR at SIG2 for being R having had a G at SIG1. Either is OK, but neither still seems wrong to me.