Page 1 of 1
New decisions feature 16/11/2017 at 23:00 #102961 | |
clive
![]() 2766 posts |
The 4.6.5 loader release, when it is made, will include a new feature for decisions. It will now be possible to make a choice based on whether a particular scenario is in use or a particular option has been selected. To do this, when setting up a choice in a timetable, use a decision name of "[NSC!xxxxx]" (without the quotes) and a choice name of "Y" or "N" (again without the quotes). The xxxxx is replaced by a code that will be specified by the simulation author. A choice of Y will be chosen if the scenario or option has been selected, and a choice of N will be used if it has not. Until 4.6.5 is released, you can test this by creating decisions in the decisions tab using these names. 4.6.5 and onwards will not allow you to use these decision names in the decisions tab and the analyser will complain about any already there. As I said, the list of codes will be provided by the simulation developer. I have updated the Peterborough manual to give the codes; for example [NSC!NBIDI] tests whether the Spalding bi-feature (down trains to Spalding can run on the Up Slow) is available. Enjoy! Log in to reply The following users said thank you: postal, VInce, 58050, BarryM |
New decisions feature 17/11/2017 at 00:29 #102963 | |
postal
![]() 5216 posts |
Tried it on Peterborough. Most of it works properly but when I use [NSC!NOLD] it is always generating a N decision whether the original or Spalding bi-di layout is selected.
“In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Last edited: 17/11/2017 at 00:30 by postal Reason: None given Log in to reply |
New decisions feature 17/11/2017 at 07:15 #102970 | |
clive
![]() 2766 posts |
Um, where did you get a 4.6.5 loader from?
Log in to reply |
New decisions feature 17/11/2017 at 09:27 #102973 | |
postal
![]() 5216 posts |
clive in post 102970 said:Um, where did you get a 4.6.5 loader from?I didn't get v4.6.5 from anywhere, I'm using v4.6.4. As per the third paragraph of your OP at 23:00 yesterday. I have created my own decisions using the specified format in a test TT for v4.6.4 as v4.6.5 should throw an error if such a manually entered decision was found. Decisions manually created in v4.6.4 are all appearing to work correctly (allowing/inhibiting entry of train depending on scenario/closures etc) except for the NOLD key word which always returns N decision. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply |
New decisions feature 17/11/2017 at 13:17 #102977 | |
clive
![]() 2766 posts |
postal in post 102973 said:Okay. 4.6.4 doesn't understand any of this stuff, so you need to force the decision to match the choice you made using the appropriate button. It will be treating this like any other decision, so you need to check the details of what you've done (have you set weights? have you misspelled something?). What I meant was that you can test your timetables, not the new feature, using the existing mechanisms. Log in to reply |
New decisions feature 17/11/2017 at 13:21 #102978 | |
clive
![]() 2766 posts |
I've managed to get a third thing in: as well as scenarios and features you can also test on whether chains are present or not. The decision name for these is "[EXT!xxxxx]". However, testing shows that, even if you start the simulation paused, seeding is done before the user has an opportunity to set up a chain. So at present you can't use this to control seeding. But you can use it for everything else. If I can see how to fix that (Mantis 18467), I will. Last edited: 17/11/2017 at 13:24 by clive Reason: None given Log in to reply |
New decisions feature 17/11/2017 at 15:09 #102986 | |
postal
![]() 5216 posts |
clive in post 102977 said:4.6.4 doesn't understand any of this stuff, so you need to force the decision to match the choice you made using the appropriate button. It will be treating this like any other decision, so you need to check the details of what you've done (have you set weights? have you misspelled something?). What I meant was that you can test your timetables, not the new feature, using the existing mechanisms.Very strange. I cut and pasted last night so don't think I had any finger trouble and was consistently getting N when the decision [NSC!NOLD] was called no matter what layout was selected. However, I am now running test TT and [NSC!NOLD] is responding correctly so I can confirm that choices from Scenarios, Closures in force, Scale of problems and Layout all seem to be working as expected. I haven't tried the Other choices. By and large looks an excellent innovation so thanks for the work. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply |
New decisions feature 17/11/2017 at 18:22 #102989 | |
clive
![]() 2766 posts |
postal in post 102986 said:I still don't understand. 4.6.4 doesn't understand any of this stuff, so you've just set up normal decisions. Unless you're going in and forcibly selecting Y or N to match your startup options, it will just select at random when a train first needs it. The "testing" bit was for you to test your timetables would be error free ready for 4.6.5. There is no way for you to confirm that choices from (say) Scenarios are working correctly because 4.6.4 doesn't know what that means. If you change "NOLD" to "FRED" throughout your timetable, it will work on 4.6.4 (but not when 4.6.5 comes out). postal in post 102986 said: Thanks. Log in to reply |
New decisions feature 17/11/2017 at 18:55 #102990 | |
postal
![]() 5216 posts |
clive in post 102989 said:I still don't understand. 4.6.4 doesn't understand any of this stuff, so you've just set up normal decisions. Unless you're going in and forcibly selecting Y or N to match your startup options, it will just select at random when a train first needs it.Misunderstood what you meant by testing. Apologies for being on the wrong wavelength. Ran half a dozen tests last night using different codes from the Peterborough manual and 5 of them appeared to produce the correct Y or N choice. Might be a good time to take up roulette. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply |
New decisions feature 17/11/2017 at 22:35 #103002 | |
clive
![]() 2766 posts |
One more thing, done on the flight home, that will be in 4.6.5: if you have an ordinary decision with choices Mon, Tue, Wed, Thu, Fri, Sat, Sun, say, then when you use it in a timetable or activity you can now put things like "[any] Wed | Fri | Sat" or "[not] Tue | Thu". The former will trigger if any of the named choices is the actual choice made; the latter if none of them are.
Log in to reply The following users said thank you: VInce, KymriskaDraken |
New decisions feature 18/11/2017 at 09:30 #103010 | |
VInce
![]() 579 posts |
clive in post 103002 said:One more thing, done on the flight home, that will be in 4.6.5: if you have an ordinary decision with choices Mon, Tue, Wed, Thu, Fri, Sat, Sun, say, then when you use it in a timetable or activity you can now put things like "[any] Wed | Fri | Sat" or "[not] Tue | Thu". The former will trigger if any of the named choices is the actual choice made; the latter if none of them are.Just what I've been waiting for Clive. In the 1977 Peterborough v1.01 there are around 680 rules - I reckon this will reduce that figure by about 200 or so for when I release v 1.02 Great, thanks Vince I walk around inside the questions of my day, I navigate the inner reaches of my disarray, I pass the altars where fools and thieves hold sway, I wait for night to come and lift this dread away : Jackson Browne - The Night Inside Me Log in to reply |
New decisions feature 18/11/2017 at 23:47 #103057 | |
KymriskaDraken
![]() 963 posts |
I've just tried this with a Decision called [NSC!NSNOW]. When I try to add the Decision I get a dialog saying "Decision names beginning with '[' are reserved". It then adds "[de-reserved] to the start of the name, and does not choose when the sim is launched with that scenario active. Am I doing something wrong? Log in to reply |
New decisions feature 19/11/2017 at 00:30 #103059 | |
postal
![]() 5216 posts |
KymriskaDraken in post 103057 said:I've just tried this with a Decision called [NSC!NSNOW]. When I try to add the Decision I get a dialog saying "Decision names beginning with '[' are reserved". It then adds "[de-reserved] to the start of the name, and does not choose when the sim is launched with that scenario active.If you've manually tried to create the decision then I think you are doing more work than you need to. If I've understood Clive's various postings correctly, the core code will silently "create" a background decision for each of the scenarios, closures etc. Therefore you don't need to create a decision as the core code has already done it. All you need to do is got to the train TT and enter the decision name in the format provided by Clive then the choice (Y if the train is to enter when that scenario/closure is invoked or N if it is not). Of course I'm open to correction if I've shot from the lip again. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Last edited: 19/11/2017 at 09:16 by postal Reason: None given Log in to reply |
New decisions feature 19/11/2017 at 00:37 #103060 | |
postal
![]() 5216 posts |
VInce in post 103010 said:In the 1977 Peterborough v1.01 there are around 680 rules - I reckon this will reduce that figure by about 200 or so for when I release v 1.02Managed to take out 300 rules and 5 DotW seed trains in the current piece of work. Haven't tested it yet so optimism could have overtaken practicality but looking better than good so far. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply The following user said thank you: VInce |
New decisions feature 19/11/2017 at 09:39 #103065 | |
KymriskaDraken
![]() 963 posts |
postal in post 103059 said:KymriskaDraken in post 103057 said:I tried that and the analyser complained that the decision did not exist. It worked with the Peterborough sim so I will double-check the name with the developer.I've just tried this with a Decision called [NSC!NSNOW]. When I try to add the Decision I get a dialog saying "Decision names beginning with '[' are reserved". It then adds "[de-reserved] to the start of the name, and does not choose when the sim is launched with that scenario active.If you've manually tried to create the decision then I think you are doing more work than you need to. If I've understood Clive's various postings correctly, the core code will silently "create" a background decision for each of the scenarios, closures etc. Therefore you don't need to create a decision as the core code has already done it. All you need to do is got to the train TT and enter the decision name in the format provided by Clive then the choice (Y if the train is to enter when that scenario/closure is invoked or N if it is not). Kev Last edited: 19/11/2017 at 10:04 by KymriskaDraken Reason: None given Log in to reply The following user said thank you: postal |
New decisions feature 19/11/2017 at 10:06 #103066 | |
postal
![]() 5216 posts |
Kev Which sim? I've just been running some quick tests on Peterborough using the codes given by Clive in the manual and things seem to be working OK. “In life, there is always someone out there, who won’t like you, for whatever reason, don’t let the insecurities in their lives affect yours.” – Rashida Rowe Log in to reply |
New decisions feature 19/11/2017 at 10:45 #103068 | |
clive
![]() 2766 posts |
postal in post 103059 said:KymriskaDraken in post 103057 said:That's correct. And that's why you can't create a decision with that name.When I try to add the Decision I get a dialog saying "Decision names beginning with '[' are reserved".If you've manually tried to create the decision then I think you are doing more work than you need to. If I've understood Clive's various postings correctly, the core code will silently "create" a background decision for each of the scenarios, closures etc. Therefore you don't need to create a decision as the core code has already done it. All you need to do is got to the train TT and enter the decision name in the format provided by Clive then the choice (Y if the train is to enter when that scenario/closure is invoked or N if it is not). Log in to reply |
New decisions feature 19/11/2017 at 11:31 #103070 | |
KymriskaDraken
![]() 963 posts |
postal in post 103066 said:KevIt's a sim under devcelopment. Kev Log in to reply |
New decisions feature 19/11/2017 at 19:52 #103084 | |
clive
![]() 2766 posts |
KymriskaDraken in post 103070 said:postal in post 103066 said:Get the developer to email me directly.KevIt's a sim under devcelopment. Log in to reply |
New decisions feature 19/11/2017 at 21:51 #103089 | |
KymriskaDraken
![]() 963 posts |
clive in post 103084 said:KymriskaDraken in post 103070 said:It's sorted now Clive. I was reading a | for a ! in the decision name.postal in post 103066 said:Get the developer to email me directly.KevIt's a sim under devcelopment. Kev Log in to reply |