Page 1 of 1
| 2010/2011 Hybrid Timetable 21/04/2011 at 19:23 #2899 | |
onlydjw
456 posts |
Currentlty having rather major issues - every train that has a X must not run if Y runs rule isn't entering, and, for example, 6N03 has a X must not appear until Y minutes after Z has left the area rule, excpet it appeared before 0N03 had left the area. I've obviously done something very wrong in the last 6 weeks of writing this, and am VERY frustrated it's not working. Anyone got any ideas??? God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 21/04/2011 at 19:23 #15284 | |
|
onlydjw
456 posts |
Currentlty having rather major issues - every train that has a X must not run if Y runs rule isn't entering, and, for example, 6N03 has a X must not appear until Y minutes after Z has left the area rule, excpet it appeared before 0N03 had left the area. I've obviously done something very wrong in the last 6 weeks of writing this, and am VERY frustrated it's not working. Anyone got any ideas??? God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 21/04/2011 at 19:46 #15286 | |
AnyFile
101 posts |
Is the timetable working correct if you remove all the rules? Try to insert just one rule, and then see if it work that way. Log in to reply |
| 2010/2011 Hybrid Timetable 21/04/2011 at 21:03 #15289 | |
|
AnyFile
101 posts |
onlydjw said:every train that has a X must not run if Y runs rule isn't entering, This maybe caused by the fact that probability are not written correctly. It seems to me to remember a post in the forum about this (but I could not remember if it was in the old forum or in the new one). If you have two train, see 1Z01 and 1Z03, and you want that in the 0% of the case 1Z01 runs and in the other 50% 1Z03 runs you have to assign the Percentage of time train runs (RRP) as 50 and 100 and put the rule that 1Z03 must not tun if 1Z01 runs. The second percentage must be 100 because if 1Z01 has not run you always want the second to run. Log in to reply |
| 2010/2011 Hybrid Timetable 21/04/2011 at 22:21 #15291 | |
BarryM
2158 posts |
Important. One thing to remember is train "Y" is the one that triggers the rule. If there are conflicting rules then nothing will occur. Barry Barry, Sydney, New South Wales, Australia Log in to reply |
| 2010/2011 Hybrid Timetable 26/04/2011 at 11:54 #15364 | |
|
onlydjw
456 posts |
What I'm trying to achieve is that one of either 1Z01 OR 1Z03 runs 100% of the time (and some rules have 3 or 4 options), and taking the thought literally, put in rules both ways round for the 2 trains not to run if the other runs. From what has been said above, I need to remove either one of the rules (but surely that means both could run?), or enter a percentage for how often train X runs (but would still cause conflicts given train X is train Y in the opposite rule)? Am I understanding this right?
God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 26/04/2011 at 12:17 #15365 | |
|
AndyG
1842 posts |
Set the first train to run 50% of the time, and the second always, with a rule "second mustn't run if the first does".
I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either. Log in to reply |
| 2010/2011 Hybrid Timetable 26/04/2011 at 13:47 #15368 | |
|
AnyFile
101 posts |
What AndyG has said is the way to obtain what you want. I will try to explain what happens and why you should put the rules and the probability has stated. Consider what you would do if you do in a different way. Suppose you put that train A should run in the 50% of the cases and train B should run in the 50% of the cases and you put no rules. What you would get is not what you want. In this case the probability are completely independent (and in this case it is irrelevant the fact that if the train A should enter is evaluated before train B is evaluate) This will give you the following possibility a) in 25% of the cases both train A and B run b) in 25% of the cases train A runs and train B does not run c) in 25% of the cases train A does not run and train B runs d) in 25% of the cases no trains run (note that the probability would be very different if I had used different probability in the beginning hypothesis) (for who has some mathematical knowledge in statistic, please note that we can discern between the two trains, this problem is different from the one where two identical coins are flipped, because the two coins can not be discerned). To obtain what you want you have to proceed in a different way, and you should take into account that the program will evaluate if a train should enter or not in a order (that depends on the entering time). (I suppose in the following that train A is the earlier train and train B is the later train) Let consider the train A alone. You want it to appear in the 50% of the case. To obtain that put no rules to restrict it from appear and set probability to 50%. Now you want train B to enter in the other 50% of the cases. That is you want it to enter in all the cases but when A has already enter. This last sentence translate to use 100% as probability and a rule that prevent it to enter if A has already entered. As a further example, if you had 3 trains, A, B, and C and you want relative probability between these train as 40%, 35% and 25% you would need the following rules and probability train A probability 40% - no rules train B probability 75% - one rule: Train B should not enter if train A has entered train C probability 100% - two rules: 1)Train C should not enter if train A has entered; 2) Train C should not enter if train B has entered. In the 75% of the cases train B is considered, but has in the 40% of the cases train A has already entered, we has that train B will enter in 35% only. Log in to reply |
| 2010/2011 Hybrid Timetable 26/04/2011 at 21:14 #15385 | |
|
onlydjw
456 posts |
Thanks to those who replied - I've just edited all my rules accordingly, and hopefully it will work a bit better now! We'll find out later in the week! I think both issues were related to the same rule type. Thanks. God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 29/04/2011 at 12:59 #15425 | |
|
onlydjw
456 posts |
Next issue - I have used a lot of "must not depart from Z until X minutes after Y leaves the area" - but these don't seem to work. For example, iron ores, Lindsey OR (6N03 appeared before 0N03 got there), and coal. 4R03 has 2 options - 4R03-1 and 4R03-2 - set up with rules as above so one or other runs - that works OK. They return as either 6B23-0 or 6F03 (set up with must not depart until 80 mins after 4R03-0 leaves the area), or 6R03 (must not depart until 80 mins after 4R03-1 leaves the area). My interpretation of this is that if 4R03-1 did not run, then 6R03 won't appear - but this is obviously incorrect, as both 6F03 and 6R03 appeared! Other trains appear much quicker than their time limits in the rules too! Anyone got any ideas? God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 29/04/2011 at 13:14 #15426 | |
|
AndyG
1842 posts |
Think you might need to try "6R03 must not ENTER until 80 mins after 4R03-1 leaves the area".
I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either. Log in to reply |
| 2010/2011 Hybrid Timetable 29/04/2011 at 16:50 #15429 | |
|
onlydjw
456 posts |
Thanks - seems better now. Bit difficult to tell if everything's 100%, but no really obvious problems up to 0700.
God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 11/05/2011 at 08:59 #15711 | |
|
onlydjw
456 posts |
Up to 1930 now, and nothing major to report. Only things I am finding is that some freight has a habit of running quite early (which in reality wouldn't happen) - some of it may be my rules aren't tight enough, and some may be the way the sim is set up - is there any way I can tighten this up for "new" trains entering the sim (ie those not affected by rules from previous trains)? Also, I've noticed that the loaded freight runs quite quickly through the sim, so I need to look at the acceleration rates of these. For example, a loaded tank train takes roughly 20 minutes to do Wrawby-Holton in reality! The ones via Elsham/Appleby are a bit quick too! God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 11/05/2011 at 09:15 #15712 | |
Meld
1111 posts |
Had a similar problem with cl 8/9s dropping in infront of cl 1s @ Moorthorpe in Pascals Sheff 83/4 T/T - this was mainly due to the missing four track section down to Aldwarke where the slower trains would be regulated to the slow lines @ Wath Rd Jcn. I got round it by making the incoming freight appear x mins after y entered, in your case I would suggest that you make the freight appear after the preceding service its booked to follow Passed the age to be doing 'Spoon Feeding' !!! Log in to reply |
| 2010/2011 Hybrid Timetable 11/05/2011 at 10:45 #15718 | |
|
onlydjw
456 posts |
That works to a point, but a bit more difficult for some entry points with a limited passenger service. I presume that the passenger trains are programmed to run on time where as the freight is more random (somehow in the sim?) The only other thought is the use of the "enter on time function" for each train, but would that be too restrictive? My only other comment is that when I was sorting out the rules (see posts above), the use of "Must appear X minutes after Y leaves the area" was explained as meaning at least X minutes after Y left the area, or on time if came after the x minutes had expired - this isn't the case - if I tighten the rule up, then it won't allow for gaining time back if running late, but I don't want it so slack that trains run silly early, which in reality they would not do as they would be held at their origin. Bit confused as to what to do for the best!
God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 11/05/2011 at 12:36 #15728 | |
postal
5265 posts |
There might be a couple of ways to get a working solution: 1) Pick a passenger train which is due to enter before the freight and set a rule for the freight that "must appear X minutes after Y enters the area". The passenger can be entering anywhere in the sim, not necessarily near to the freight. Presuming that the passenger is running early/late within the normal sort of sim parameters and that you pick X minutes to be a bit smaller than the scheduled time gap between the entry of the passenger and the entry of the freight, then the freight should be able to enter slightly but not stupidly early (but may even enter late). If you do this and you can find a train entering fairly close to the freight entry time rather than working to the previous train leaving the area, you can probably get a better control over how much leeway you want to allow to the freight. 2) A similar process using the rule "must appear X minutes after Y arrives or passes Z". The Z seems to be any sim location, not just stopping points from the drop-down menu. There will be some perturbation if the nominated passenger has been delayed in running through the sim which would make this a bit more random than the first suggestion but it does give you a lot more choice about narrowing down the entry time of the freight as there are a lot more passing/stopping times than entry times for the passengers. I haven't tried either method so I can't vouch for any proven results, but they might be worth a try. “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 |
| 2010/2011 Hybrid Timetable 11/05/2011 at 12:44 #15729 | |
|
Meld
1111 posts |
John Both methods you describe above work - I've set rules like this in the Sheffield 83/4 t/t and success has been acheived with both methods. Hmmm now the new delay feature in the t/t editor, would be nice if one could randomise the preset delays put in by the t/t author by the click of a single button. instead of going through and having to alter each one as required. Passed the age to be doing 'Spoon Feeding' !!! Log in to reply |
| 2010/2011 Hybrid Timetable 16/05/2011 at 10:36 #15859 | |
|
onlydjw
456 posts |
Just an update to say version 1 has completed it's timetable test, and there are no further major errors. I am working on version 2 (with more coal trains and more steel trains) to put in "new" workings through Spring 2011. This will include more rules to try and get the freight not to run too early, as in reality, the freight is held to time more than the sim can simulate (it appears). Many thanks to those who've helped test version 1. God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 16/05/2011 at 12:42 #15860 | |
|
postal
5265 posts |
Meld said:Hmmm now the new delay feature in the t/t editor, would be nice if one could randomise the preset delays put in by the t/t author by the click of a single button. instead of going through and having to alter each one as required.I've got it in the back of my mind that these delays are randomised each time a new run of the sim with the TT is started. If the TT is then saved, the delays are included and you have to clean the TT to go back to the "virgin" state. Could someone with more expertise confirm (or shoot me down)? “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 |
| 2010/2011 Hybrid Timetable 16/05/2011 at 13:35 #15861 | |
Peter Bennet
5402 posts |
Your are correct that every time you save the main timetable (as a WTT) you risk compounding any delays that the previous run may have generated. Look up "Timetable cleansing" on Wiki- that should fix it. Peter I identify as half man half biscuit - crumbs! Log in to reply |
| 2010/2011 Hybrid Timetable 16/05/2011 at 15:17 #15863 | |
|
AndyG
1842 posts |
Peter Bennet said:Look up "Timetable cleansing" on Wiki- that should fix it.TT import-export and Wiki:timetable cleansing NB make sure the 'ignore extra info' box is ticked. Current simulations also have a 'Clean TT' button under the 'Development' tab in the TT editor. I can only help one person a day. Today's not your day. Tomorrow doesn't look too good either. Log in to reply |
| 2010/2011 Hybrid Timetable 16/05/2011 at 16:05 #15864 | |
|
onlydjw
456 posts |
Where do I find the download for the file I need to do the conversion? I've had a look in all the obvious places!
God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 16/05/2011 at 16:28 #15865 | |
|
postal
5265 posts |
Dan If you follow the Wiki link that Andy put up you eventually find a link to Clive's site where you can find Convdata and Instructions. Clive very kindly sent me some details about where to install convdata.exe in order to make it easy to use. He wrote that according to the Delphi and Windows documentation, it will look for convdata in each of the following places in turn: 1. The directory from which SimSig loaded. 2. The current directory for SimSig (I think that's the one holding the timetables). 3. The 32-bit Windows system directory. 4. The 16-bit Windows system directory. The name of this directory is System. 5. The Windows directory. 6. The directories that are listed in the PATH environment variable. I dropped it into my Windows/System32 directory and it is picked up from there by whichever sim I am using. “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 |
| 2010/2011 Hybrid Timetable 01/06/2011 at 15:51 #16237 | |
|
onlydjw
456 posts |
Just a note to apologise for the lack of updates - I've been having computer problems which should be resolved whilst I'm away at Didcot tomorrow and Water Orton on Friday. Hope to have this ready in the next couple of weeks. God bless, Daniel Wilson Log in to reply |
| 2010/2011 Hybrid Timetable 10/06/2011 at 12:58 #16463 | |
|
onlydjw
456 posts |
See the link below for a continuation topic for version 2 of this timetable. Mods - can this topic now be locked, as version 2 replaces version 1, which is now obsolete - thanks. {Done-AG} http://www.SimSig.co.uk/index.php?option=com_agora&task=topic&id=3183&Itemid=54 God bless, Daniel Wilson Log in to reply |