Atomic Systems

I would not be surprised if portals do not like stop-at commands. They are meant to be destinations, so you are supposed to set them as destination of the train. At least this would sound plausible to me.

phat2003uk wrote:Yes I know that two and a half hours is not unreasonable, it is unreasonable for a scenario on the Paddington - Oxford route due to Rail Simulator. You can keep trying but I'm just saying that you maybe wasting your time.

We all may well be wasting our time with Rail Simulator in general.

AndiS wrote:While I wanted to have AI working satisfactory long ago, I decided for myself to just give it a rest until they release a fix of the AI. We all know the AI is inferior by far to any other product. We know that RSDL are planning to fix it. We know that the fix is not available yet.

Given that (1) the problems are fundamental and any changes would require the entire pathing code to be rewritten more or less from scratch (see analysis here for a full explanation); (2) RSDL is, by its own admission, tiny; (3) despite nearly a year passing since release, not one thing has been done to remedy these defects in two upgrade cycles; and (4) the dismissive remarks made by RSDL people about these issues (such as that documented by cwgaskell earlier in this thread), combined with a wilful persistence in refusing to discuss anything to do with the pathing whatsoever (even to explain how it is supposed to work in documentation or on the forums so that people can create scenarios within its limitations) makes it manifestly unlikely that there will ever be a solution to these issues.

On the subject of dismissive remarks, cwgaskell noted that it had been claimed that Rail Simulator never goes into "infinite loop": that is flatly contradicted by the output of the programme's own logging, which, with the pathing tab enabled, reported, "]Trace C:\Dev\RailSim\Code\DLLs\DriverManager\cDriver.cpp : 2097 = Unable to resolve pathing due to infinite loop" (see here for full documentation).

I have also seen repeated remarks by RSDL employees that Rail Simulator "can do everything [we] want", if only we knew how, and then repeatedly refusing to answer questions about how things that I and others had already stated that we wanted to do were possible, or how the problems that I have expressly identified on this thread are consistent with that claim.

Searching through the posts of RSDL employees for key terms such as "path" and "dispatcher" gives the following results:

"Dispatcher"

RSDerek:

6th of August: "Hi, while the work for Mk2 has finished, planning for the future has not. We do have a meeting with the Kuju coders who worked on the dispatcher this Thursday.
Once we have discussed the issues we will schedule the work.
I'm sure Adam and I will keep you posted as best we can."

4th of May: "Hi, I'm not saying their are not issues with the dispatcher, we know it has it's limitations, and where we can we are improving them, however there is more than one way to skin a cat".

RSAdam:

31st of July: "Hiya, Stalled AI is not fixed in Upgrade Mk2. Dispatcher fixes are still being finalised".

22nd of July: "Directionality and Line Type are key. If you have a passenger train, but your track is freight, the Dispatcher will have a problem with that to."

8th of July: "Bear in mind that if all tracks on a four track mainline are bi-directional, the dispatcher will have no concept of correct line running and may well end up pathing trains around the insides of corners, meaning lots of unusual use of junctions."

9th of May: "you have laid your track using the Default1 TrackRule, this does not feature any levers at all. So if your junction is Automatic, which means the Dispatcher will control switching it, what do you mean when you run the Scenario and the junction 'wont budge'?".

1st of May: "Hi AndiS,

I guess my earlier comment fell on deaf ears. The signalling in Rail Simulator is NOT impossible to fix, infact its not even hard to implement new functionality, or even a completely new system." (In response to a comment by AndiS pointing out that the signals are impossible to fix because it is impossible to test the dispatcher properly)

4th of May: "Above you mention that unless the player follows the set timetable to the second (in the scenario set out in the opening post of this thread), everything will grind to a halt, and that this is somehow the fault of Rail Simulators dispatcher. I would diagree. That to me sounds like an extremely challenging scenario, and if the player fails to do as required, thats the players fault not the Dispatchers. This only proves that the player does not get ultimate priority (unless set that way specifically), unlike some other train simulations I could mention.
...

You also mention - "the complete lack of documentation " - relating to the Dispatcher in Rail Simulator. Might I point out that the RSDevTools (required for develop scenarios in Rail Simulator) has 7 specific documents on authoring scenarios, and this is further suplimented by 11 official video tutorials on the subject. Can that really be considered a complete lack of documentation? Pathing of AI trains is based on priority and track properties. Remember though that ultimately there is no perfect dispatching system out there. If there was, we wouldnt have signalmen in signal boxes still over-seeing operation of the railway."

4th of March: "No, just dont save during the preview. Probably ensure to rewind to the start before saving if you do.

The good news I guess is that we are looking into Signalling functionality and Dispatcher enhancements for Upgrade Mk2."

4th of March: "Cheers for the detailed insight. I am currently discussing with Derek and others on the team how we should look at these going forward.

The main thing from me is that I actually disagree with many of the above points. Not from a bias perspective, although I apreciate many might not be able to see beyond this, but from personal experience. I can happily make complex scenarios in Rail Simulator without ever experiencing issues 4/5/6. All functionality as provided always works for me. Therefore I need to put effort into clarifying what I am doing that others are not.

Note 3 is a cross between signalling and dispatcher functionality. I would imagine (without going to the actual location you find the problem) that for the operation you are attempting to do, a signal has been used at that location that doesnt cater for such operation. This therefore is not a fundimental problem, but in the creation of the route not every possible eventuality may have been covered.

Notes 1 and 2, we are already putting effort into resolving for Upgrade Mk2." (In response to a catalogue of issues posted by Rabid about problems encountered when making a scenario)

"Pathing"

RSAdam:

30th of March: "Up/Down are related to single directions (Up or Down). Use this for sections you wish to force the player and AI pathing in one particular direction. If you lay a section of track that is facing the wrong way (as shown via the space bar displays), simply hit the 'Toggle' button to flip it. Both is related to Bi-Directional track. This is used for track where trains are desired to travel in either direction and where movement is not limited."

(None for RSDerek)

"Signalling"

RSAdam:

11th of August: "Remember to bear in mind that everything AndiS has said is based on if you use the default signalling system that comes with Rail Simulator where we have only provided functionality that the default routes need."

18th of August: "People were too busy deciding and in fact insisting what they believed the signalling in Rail Simulator couldn't do."

17th of August: "I believed that all of your tests and investigations were being done using the default signalling system provided out of the box with Rail Simulator."

17th of August: "Hiya, We can only apologise if Mk2 breaks current 3rd party scenarios that use the default signalling system. Alot of users were crying out for extra funcitonality on this system, but we did as best we could to avoid the new features creating problems.".

5th of August: "The signalling present in the Isle of Wight addon has not been designed for use on custom routes, and it will not be compatible with the default signals."

1st of August: "... use the same logic and setup that the updated signals provided in Mk2 use. So Mk2 will bring the default routes upto the standard of the IOW signalling."

22nd of July: "The IOW signalling is built from functionality included in the updated scripts that will come with Upgrade Mk2."

28th of February: "The term 'anything is possible in Rail Simulator' is in reference to what the end user can achieve. A public forum is not a place I wish to dive into the technicalities of work schedules and tasks at RSDL, its simply not appropriate.

The implementation of additional functionality requested by the community in Upgrade Mk2, when compared with the signalling before its release, should go along way to proving the nay sayers that the LUA system in Rail Simulator is more powerful than people realise.

To ensure that I am correctly quoted post Upgrade Mk2 as I have the feeling clarification is only sort so to use as a check list against things implemented into RailSim at that point. We make no guarrentee about fixing everything people have requested, and although we pay great attention to and make notes from, RSDL is not governed by prioritised coloured lists on forums. We appreciate the feedback and we do everything we can to enhance Rail Simulator in the way its end users wish. At the end of the day though we are a commercial business and make decisions and take directions on that basis".

27th of February: "The statement that anything is possible in the Rail Simulator Signalling System still stands, the work in this area that we plan for Upgrade Mk2 is entirely based in LUA Script system. Hopefully then, the short sighted conclusions about signal functionality will cease

Technically the only fix we are making is to a bad judgement about how much we should implement and how much we should leave for the community to play with implementing."

17th of January: "Thought I should add a few words here. We do admit that maybe one or two more critical functions were not implemented into the supplied signalling system but it seems to be that everyone is judging Rail Simulators signalling capabilities on that one system rather than looking at the bigger picture. Many have questioned, why if such open ended functionality is possible, was it not implemented. Without going into every detail of games development, it came down to time available and what was best for every situation.

As AndiS has pointed out, the system supplied out of the box is capable of a large amount of operation with an enormous aray of flexibility when it comes to altering junctions or using the same signals in a multitude of track arrangements. This makes the signalling included as flexible as possible for the audience using it. At the opposite end of the scale, it is also quite possible to create a signalling system that is geared entirely to working on just one route. Creating each exact signal logic for every individual location. Surely that is a far more powerful system that anything before?

The thing everyone seems to be missing is that there is no reason at all why people need to stick with the system included out of the box. For the more advanced enthusiasts it is quite possible to create an entirely new and self operating signalling system to achieve anything you wish. We designed a signalling system that could cater for as much as possible, which ok doesnt meet everyones standards. However when given the ability to push it to one side and create something that does meet ones own standards why should that opportunity not be taken? I dont really think after only 4 months in the public domain, anyone is yet capable of determining in a few words if Rail Simulators signalling capabilities have no future.

We do intend to make some improvements to the system included, but that will have to come at the cost of not disrupting anything already using it. It would be a shame for you to simply side step such vast opportunities".

10th of November 2007: "Having read through james post, and admitting that I had seen this communication from him to Derek previously, but been to busy to fully take in the findings, I am just not sure how to respond – There’s no quick answer.

I will be attending the Flight & Train Simulation show in Birmingham on November 17th and will be more than happy to discuss these issues in person.

In an attempt to summarise an answer, here are some pointers:
• We accept that there has been a lack of early documentation on signalling which has lead the community to misjudge Rail Simulator, when clearly so little is known and understood about how things work under the hood.
• We admit that some important signalling operations were not included in the default routes, this is not because we did not know about them it was a choice, just because something isn’t present, don’t assume it is not achievable.
• It would also appear that we mis-judged how much of the basics we needed to implement and how much should be left for the community to play around and come up with. After all we wanted a simulation that will grow, not just by us at RSDL but by you guys too.
• There is wide spread confusion over which systems govern what operations in Rail Simulator, which again is not due to the software, but due to the lack of time that the community has had to understand just how things work. Thus more time, interaction with the development team, and documentation about these systems when they are released will resolve most of the problems people are concerned about.

Since release of Rail Simulator we have had meetings in person with people from this forum who were concerned about the state of signalling, and every one of them has been pleasantly surprised about the state of things once the full picture could be realised and understood. We only wish more people would give us the benefit of the doubt as we simply could not cope with getting everyone of you down to Guildford to chat with us personally".

10th of November 2007: "The Developer Tools are designed for creating NEW content. Not editing whats already there. So for example you cannot open up the details of the Class 47 and start changing elements about it.

The signalling however is a different matter, as the scripting is done via LUA which can be opened in Notepad (although we recommend Notepad++). So this can be changed by end users, and you dont even need the Devtools for this".

RSDerek:

13th of February: "I think you can safely say we are supporting Rail Simulator.

We will be improving a number of features for the upgrade mk2, more details on this soon, however I can say that signalling is near the top of the list.

best regards

Derek"

3rd of December 2007: "The trivial issues are far easier for me to comment on. That’s not to say we are not doing or planning anything for the signals. It's just not my area.

We also try to cover as many aspects of the product as we possible can whether it be signals, track, trains, exporters, operations, time tables etc, however I would personally like to thank you both for taking such an active interest in this area.

LIke I have said, we will get back to you on these issues and other raised.".

10th of November 2007: "Morning,

Sorry for not getting back to you James after I acknowledged your mail. I have asked a number of people to contact me about the issues people have the signalling. Once people have the signal document, read and understood it and what can be done, then we can take the signal subject from there.

regards

Derek".

27th of October 2007: "The core product meaning the 4 supplied routes, much can be done with the scripted signalling when we open it up to you.
I will try to get more information to you next week.

We, RSDL are a core team, smaller than the team that made the core product, but with resources to grown in the areas that need attention.
on another note, I am a little worried about you guys stalking me..."

It is evident from that that: (1) no specifics are ever discussed; (2) there has been a history of denial of any serious or fundamental problems, combined with repeated promises of fixes (to what must by implication be considered trivial issues or "limitations") which never materialise - note the statement that signalling was "at the top of the list of priorities for Mk2", and yet only superficial changes were made to signalling that have had no improvement on function; and (3) the acknowledgement of any problems in this area at all is consistently reluctant.

Doh, It's friday and I was hoping to nip off before the traffic built up.
Oh well.

jamespetts wrote:(1) no specifics are ever discussed;

They were discussed, but not with the forums.
If people have issues and ask for help we generally try and help them get their sceanrios working or find another solution.
While we give information and feedback what we are upto on the forums, our main interest is to help those with problems.

Adam and I spend many hours helping users that want to be helped. Responding to certain posts or defending/explaining why Kuju or EA did x/y/z takes time away from that task

jamespetts wrote:(2) there has been a history of denial of any serious or fundamental problems, combined with repeated promises of fixes (to what must by implication be considered trivial issues or "limitations") which never materialise - note the statement that signalling was "at the top of the list of priorities for Mk2", and yet only superficial changes were made to signalling that have had no improvement on function; and

We are not denying that the Dispatcher has a few issues finding paths, that we need to deal with. It is not always the dispatchers fault however, it could have been a signal set up wrong (which we have tried to fix) or the properties in the track or lack of markers, which we admit there are not enough of in the default routes.

jamespetts wrote:(3) the acknowledgement of any problems in this area at all is consistently reluctant.

James, we are aware of the issues and making plans to address them. We had stated that the dispatcher would not be looked at for mk2, however it will be for mk3.
Along with the fixes we are looking at giving better feedback to how and why scenarios fail.

jamespetts wrote:We all may well be wasting our time with Rail Simulator in general.

While I understand your frustration I'm not sure this statement is correct.

Lastly,

While making scenarios our guys sometimes come up against issues which prevent them from creating a scenario a certain way.
They do bitch about it for about ten minutes, while talking about the problems they are having they find an alternative way to make it work.
If they hit a brick wall rather than keep hitting their head they take a step the side and walk around the wall.
While I understand this is not an ideal solution it does prevent headaches till those issues are addressed.

regards

Derek

AndiS wrote:I would not be surprised if portals do not like stop-at commands. They are meant to be destinations, so you are supposed to set them as destination of the train. At least this would sound plausible to me.

This is the problem. If you want a train to go into a Portal and come out to a stop or Passenger Pick up at a Marker after a required time you have to put in a command using the stop at or pick up passengers Tabs on the Loco unless I am doing it wrong. eg. Should i be using the Consist Operations Tab on the Loco. The time the Train comes out after going in seems to be the same whatever details you put in the Portal properties at the Route Production stage.
Regards James.

Guys sorry to say but ..............why not rebuild Railsim from the bottom up!
seems to me we will never get the fixes that we need to make it a SIM not a toy game, I have stuck steadfast with this Sim
since it came out, having read James posts I am comming to the conclusion that a complete re write is the only way forward.

I think i made it clear time and time again! all signals remain at danger unless a path has been been set for a train (semaphores).....but no the same problem appears time and again.
Pete Doré

Ok, It's friday night and a bank holiday weekend..

http://www.youtube.com/watch?v=vK2by38n0kM

regards

Derek

Very Good Derek erm....wont fix the Signal bug though will it

Pete Doré

Well James quite a thorough Dissertation on the problems of pathing, dispatching and signalling which make very interesting reading and highlights many of the problems in creating a Complex Scenario. I for one have only produced simple Scenarios for testing purposes of my own Route with no AI until recently. I have hit several preverbial brick walls when trying out Portals and placing AI trains. Of course one could say that my Route is at fault but I have added many additional markers because the Pathing Engine could not seem to find the correct path without the additional help, somtimes in areas which appeared to me to be quite simple. But then I am not a computer.
To be fair to RSDL I have been offered help in areas of my Route in particularly in the Default Signals in relation to MK2 and invitations to PM them as appropriate. I do worry slightly about what may or may not happen in the future regarding the problems you mention being solved to a satisfactory degree. I do however accept the statement from RSDL that changes have to be done carefully so that people who have produced Routes and Scenarios previous to an upgrade do not find that their work is destroyed after installing said upgrade. Some alterations as I have found with MK2 had to be made to my Route but nothing to difficult.
The problem is a balancing act in trying please both users old and new, technical and non technical. Also to strike a happy medium between people who just want a good game and those who want absolute realism.
However I agree that the fundamentals have to be sorted out first. A house has to be built on a strong foundation or it will fall down. I believe this is the Core of the program and there are, as James points out anomalies in this area which affect the other areas of the program and what it is possible to do as a user.
There are of course limitations to enable the program to run on a variety of systems as with any other program.
Regards James.

Let me just say that most of the Sim is first rate! no doubt about that!
Loco's rolling stock, world graphics, even the sounds are very good, as you point out its the core where the problem seems to be, sort out pathing and signalling and you have an absolute winner

Have a good bank holiday Derek!

Pete Doré

I've been watching this thread, and others along the same lines, with great interest over the last few days and feel that maybe I should contribute a comment about my experiences with RS. I came new to train sims late last year and after playing around with MST for a couple of months I gave up, primarily because I was not happy with the graphics and was also struggling to get to grips with building my own route. I'm not a great one for just driving trains but want to build routes and then get stuck into issues such as signalling and operations.

I then discovered RS and spent 5 months building a reasonably accurate route based on the Gloucester/Chepstow/Monmouth/Ross area. Now building the route did not come easy at first but, taking careful note of the limitations of the sim, I found a way, with the help of this forum, around the various problems I came up against and now have all the track down and working well. On to signalling. My first attempt relied on using the signals provided which naturally did not suit the route in every instance. So, with the support provided by the RS docs, this forum (and in particular the excellent work of Mark Brinton and AndiS) I moved onto to customising signals for my route and spent many a happy hour signalling the route.

The next step was operations and after reading the various posts on this forum this area really did worry me. I was almost tempted to give up before I started. Doom and gloom seemed to be the predominant sentiment. I took a look back at MST1 but again could not get to grips with route building and the graphics. I even went out and bought Trainz but the graphics killed my interest within about 5 minutes. Eventually I took inspiration from what some people were attempting, in particular the fun that Ted (Oldpufferspotter) was having experimenting with scenarios and with help from Ted and Mark Brinton (again!) got the signalling working properly and learnt, from trial and error, how to operate portals, AI traffic etc. All the while being aware of the current limitations of RS.

So I'm now at the point where I have created a number of scenarios using portals ("in and out"), AI traffic, etc., culminating this afternoon in being able to spend a happy 2 hours sitting at Lydney Junction and watching 50+ AI trains trundle past, heading to and from the yard or the 6 portals I have set up. All done within the current limitations of the sim and for the £50 odd it has all cost me for the sim and various add ons. Signalling works fine, pathing works fine and train priorities work fine.

What's the point I am trying to make? Provided you accept the current limitations then I think you can have hours of enjoyment with what we have now. Derek's point was well made when he said...

RSderek wrote:
While making scenarios our guys sometimes come up against issues which prevent them from creating a scenario a certain way.
They do bitch about it for about ten minutes, while talking about the problems they are having they find an alternative way to make it work.
If they hit a brick wall rather than keep hitting their head they take a step the side and walk around the wall.
While I understand this is not an ideal solution it does prevent headaches till those issues are addressed.

Derek

RSDL are not perfect and I would question at times their priorities but I am confident that they want this sim to succeed, are aware of the bigger issues and will work towards a solution. In the meantime I'm going to enjoy what we've got....

Nick

npcleary wrote:I've been watching this thread, and others along the same lines, with great interest over the last few days and feel that maybe I should contribute a comment about my experiences with RS. I came new to train sims late last year and after playing around with MST for a couple of months I gave up, primarily because I was not happy with the graphics and was also struggling to get to grips with building my own route. I'm not a great one for just driving trains but want to build routes and then get stuck into issues such as signalling and operations.

I then discovered RS and spent 5 months building a reasonably accurate route based on the Gloucester/Chepstow/Monmouth/Ross area. Now building the route did not come easy at first but, taking careful note of the limitations of the sim, I found a way, with the help of this forum, around the various problems I came up against and now have all the track down and working well. On to signalling. My first attempt relied on using the signals provided which naturally did not suit the route in every instance. So, with the support provided by the RS docs, this forum (and in particular the excellent work of Mark Brinton and AndiS) I moved onto to customising signals for my route and spent many a happy hour signalling the route.

The next step was operations and after reading the various posts on this forum this area really did worry me. I was almost tempted to give up before I started. Doom and gloom seemed to be the predominant sentiment. I took a look back at MST1 but again could not get to grips with route building and the graphics. I even went out and bought Trainz but the graphics killed my interest within about 5 minutes. Eventually I took inspiration from what some people were attempting, in particular the fun that Ted (Oldpufferspotter) was having experimenting with scenarios and with help from Ted and Mark Brinton (again!) got the signalling working properly and learnt, from trial and error, how to operate portals, AI traffic etc. All the while being aware of the current limitations of RS.

So I'm now at the point where I have created a number of scenarios using portals ("in and out"), AI traffic, etc., culminating this afternoon in being able to spend a happy 2 hours sitting at Lydney Junction and watching 50+ AI trains trundle past, heading to and from the yard or the 6 portals I have set up. All done within the current limitations of the sim and for the £50 odd it has all cost me for the sim and various add ons. Signalling works fine, pathing works fine and train priorities work fine.

What's the point I am trying to make? Provided you accept the current limitations then I think you can have hours of enjoyment with what we have now. Derek's point was well made when he said...

RSderek wrote:
While making scenarios our guys sometimes come up against issues which prevent them from creating a scenario a certain way.
They do bitch about it for about ten minutes, while talking about the problems they are having they find an alternative way to make it work.
If they hit a brick wall rather than keep hitting their head they take a step the side and walk around the wall.
While I understand this is not an ideal solution it does prevent headaches till those issues are addressed.

Derek

RSDL are not perfect and I would question at times their priorities but I am confident that they want this sim to succeed, are aware of the bigger issues and will work towards a solution. In the meantime I'm going to enjoy what we've got....

Nick

Thats a good post I have to say I feel the same.

John

Nice clip derek. Blackadder is a fav.

The only intractable pathing problem I have is to get the Council to repair the pavement outside my house.

Have a great weekend.

Just to chuck in my tuppence....

... Trainz has been around for 8 years (correct me if I'm wrong) and as a one-time follower of this particular religion I remember buying 4 versions at £30 each and each of these had up to three patches which fixed certain things and broke others, it wasn't until Trainz 1.3 that we could even set a path for an AI train. Also can anyone also remember the version/patch that made the wheel animations play in reverse? At least RS had this right!

So that's what, 7 years more development than RS so comparing the scripting and AI pathing to RS isn't quite fair. It would be nice if TRS had the same flexibilty as Trainzscript out of the box but I'm willing to wait patiently and tinker in the route builder to see what the next version brings.

Cheers.

Jon

peterdore wrote:Let me just say that most of the Sim is first rate! no doubt about that!
Loco's rolling stock, world graphics, even the sounds are very good, as you point out its the core where the problem seems to be, sort out pathing and signalling and you have an absolute winner

Pete,

I'm not getting at you but unfortunately everything in the rest of the RailSim garden is not as rosy as could be supposed from your post. The locos and rolling stock are not very good in my opinion. Whilst they may look better than some produced for MSTS, they have flaws in the way that their "physics" are modelled that are just as annoying to some as the flaws in the pathing and signalling are to others. For example the HST and Class 166 should have notched brake control; graduated train brake release should be modelled in those, the Class 47 and Deltic; the bogie brake gauges in the Class 47 and Deltic should react to the train brake; the Class 66 throttle apparently has an incorrect number of notches.

All these bugs, except the last were pointed out prior to Upgrade Mk1 and have still not been rectified. Agreed they are not "stoppers" like broken signalling but on the other hand they should be easy to fix and I am disappointed that they have not been. I do sometimes wonder whether RSDL has anyone with the knowledge to get these things right in the first place, and perhaps of more concern, rectify them once errors are identified.

Andy L