[Csantucci]

Eugen could reproduce Mirek's problem (I think), so I got a test case (thank you Eugen!).
Mirek, can you pls. try release 2958 and check if your problem with the GP38-2 activity has gone?

[mirekkr]

Hi Carlo and Eugen,
I've tried hard to document the activity as detailed as possible. I hope this helps with correcting the bug.
If you like reading detective novels :-), the attachment contains something like this:
- Pictures 1 and 2 show the paths, signals and speed limits (with comments)
- The TXT file logs step by step activity run in four different stages
If you don't like reading novels, here's my hypothesis about the cause of the issue:
- Speedposts for Passenger trains are somehow incorrectly used for Freight trains: their value is NOT set for the freight, but the Route's limit is used instead.
- Something must be mysterious about the program stack of speed limits: placing a new speed limit sign changes the behaviour significantly.
Tested on X2951, with Location Linked Passing = On, AI Shunting = On (of course :-).

PS: I know that the best way would be to send the Act along with the route. But this is a little complicated: the route is originally Whitefish by René Rump, but slightly modified by me (especially to better work and look in OR). And I have no permission from the author to modificate it and send it over internet.

Regards and thanks,
Mirek

Attached File(s)

•  20153003.zip (148.44K)
  Number of downloads: 139

[mirekkr]

I see Eugen was faster than me :-)
Never mind, maybe my post helps too.

Mirek

[mirekkr]

Carlo,
did some tests with the Act (and one more) and with X2958 I cannot see that 80mph any more - so this works like swiss watch!
(But - the "mystery of 34 mph" (as seen on Eugen's picture and documented in my post #22, Stage One) is still here. How about categorize this as "nothing toxic" and leave it as is? In the end, it can be easily corrected in RE.)

And - Eugen, thanks for your cooperation!

Thanks again,
Mirek

[Csantucci]

With release 2960 I should have solved also the problem of the couples of speedposts with identical position. Mirek's test activity of post #14 seems to work correctly now.

[eugenR]

mirekkr, on 30 March 2015 - 03:17 PM, said:

Carlo,
did some tests with the Act (and one more) and with X2958 I cannot see that 80mph any more - so this works like swiss watch!
(But - the "mystery of 34 mph" (as seen on Eugen's picture and documented in my post #22, Stage One) is still here. How about categorize this as "nothing toxic" and leave it as is? In the end, it can be easily corrected in RE.)

And - Eugen, thanks for your cooperation!

Thanks again,
Mirek

Hallo Mirek
This "mystery of 34 mph" is an "old" signalspeed-limit. As long as you can see it at the Position of "Picture 1", something is not O.K.
I hope with the X2960 you don't see it again.

Cheers
Eugen

[mirekkr]

Hi Carlo and Eugen,
two news:
One excellent - my test activity from post #14 on Marias Pass route works OK! The speedposts with the same coordinates make no problems anymore.
The second - bad and good: I still can see that 34mph limit. The good side of this news is that I think I have an Act for reproducing this syndrome; it's on Marias Pass (see attachment). (Note: I did the same trick with signals speed limits, i.e. modified their speeds to 41mph instead of 40mph to easily distinguish between signal limits and speedpost limits.)
/a/ You start at Summit as Freight and end at Glacier Park. All speed limits are ok.
/b/ Now, at the very start, uncouple all cars so you become Passg. Go as far as the last but one signal before Glacier Park (it shows APPROACH_2 = 41mph ASAP). TrkMon shows "future 41" at the next speedpost (P45/F40). This is ok.
Continue you journey. As soon as you clear the speedpost, TrkMon shows 45 at the next signal (APPROACH_1 = 41mph ASAP). Why 45? If the 1st signal speed has been obeyed, why not the second one?

I apologize for being annoying. I love OR and very much appreciate your work.

Regards,
Mirek

Attached File(s)

•  20150331_Act.ZIP (3.39K)
  Number of downloads: 159

[eugenR]

mirekkr, on 31 March 2015 - 11:28 AM, said:

Hi Carlo and Eugen,
/b/ Now, at the very start, uncouple all cars so you become Passg. Go as far as the last but one signal before Glacier Park (it shows APPROACH_2 = 41mph ASAP). TrkMon shows "future 41" at the next speedpost (P45/F40). This is ok.
Continue you journey. As soon as you clear the speedpost, TrkMon shows 45 at the next signal (APPROACH_1 = 41mph ASAP). Why 45? If the 1st signal speed has been obeyed, why not the second one?

I apologize for being annoying. I love OR and very much appreciate your work.

Regards,
Mirek

Hi Mirek,
Thank you for the activity it was a big help
I think the reason is that in the Sigcfg.dat all signale has SIGNALFLAGS ( ASAP )
this make that the signal-speedlimit of 41 mps is ignored, see here:
http://www.elvastowe...__1#entry164760
this is a difference to MSTS, in MSTS I can't find any reaction on SIGNALFLAGS ( ASAP )
If you cancel this flags, all works fine.

With the combination Signal-speedlimit and SIGASP ( ASAP ) something go wrong.
There are a Signal with APPROACH_2 41mi (ASAP) and one with APPROACH_1 41 km/h (ASAP) at the End of the activity.
The combination of Signal-speedlimit and ( ASAP )is a contradiction!!! But it is original Marias-Pass
At the Picture I have passed the signal with APPROACH_2 41mi (ASAP).
The TM shows a Preview for 41 mi at the following Milepost 45mi,what is wrong because the passed Signal (with (ASAP)) should not produce any speedlimit


If I have passed this Milepost 45mi the speedlimit was going on 41mi (Signal-Speed) instead of 45mi (Speed-Plate)
(ASAP-Signals should not have any speedlimit!)


This (ASAP) flags are thought for interval-signals.
In interlocking-signals as here at the entrance of a staton before switches this signalflag is wrong,
but I know it is original Marias-Ppass

[Csantucci]

I have uploaded a fix with release x2964. First of all, thank you again to Mirek and Eugen for the very qualified help.
A small explanation about the fix:
Things clearly worked wrong with the ASAP flag for activity mode before x2964 (Timetable mode is another story).
I have removed the effects of the ASAP flag for activity mode, as also MSTS does. I did this not only because for the sake of compatibility.
In fact both the solution of neglecting the speed value if accompanied by an ASAP flag, and the solution of immediately considering the speed value even if accompanied by an ASAP flag do not correspond to reality: in reality, the speed value has to be respected, but not immediately: ASAP (it's a bit undetermined...).
However, selecting the solution of neglecting the speed value would have been acceptable for player trains, but not for AI trains, that wouldn't have reduced the speed, neither immediately nor ASAP.
Hence my decision.

A better solution would be to start the effect of the speed reduction not where the signal is, but some meters (yards...) after, e.g. after 1 Km (American friends could suggest values here). However this is more complicated to implement.

[mirekkr]

Oops! My yesterday's post has been lost... Once again then:

Carlo,
your decision & solution about ASAP flag makes sense to me. (Anyone has a possibility to edit sigcfg.dat and delete/add speedlimits to the signal aspects as desired.)
I agree to Eugen's opinion that setting speed limits to signals showing APPROACH_2 is somewhat nonsense: their meaning should be only a warning to slow down before next (mostly absolute) signal. And we cannot forget about that the engineer should be familiar with the route's signals and rules before he starts his journey.

I've tested my Marias Pass activities and they work fine since X2964.
I think we can close this thread at this point (?). You've done a great work with enhancing the Activity mode and - again - thanks for that!

Regards,
Mirek