Open Rails does not follow player path file

[eric]

Nobody was attacked, Geoff. They were just told where to report the problem if they expected any action to take place.

The reason the discussion takes place on Elvas is because it's a more secure forum, and the operator/owner donated the space for them to have a closed forum for the development team.

This forum is freely read without having to bother creating an account, and it's pretty clear that even banned members still manage to participate via sock puppet accounts... That simply doesn't happen on Elvas.

Elvas is closed to anonymous members, however Dave has been known to make exceptions for people known not to be a sock puppet.

[lateagain]

Nobody was attacked, Geoff. They were just told where to report the problem if they expected any action to take place.

Well you might not consider your comment "Pissing and moaning about it here won't fix it." an attack after someone has gone to the bother of submitting such a detailed report but you and I aren't even in the same ball park if so. I'm a Londoner by birth and we're used to plain speaking and grew up with having to take it if you dish it out .....but we know the difference

The reason the discussion takes place on Elvas is because it's a more secure forum, and the operator/owner donated the space for them to have a closed forum for the development team. .

I completely understand the reason and value of such forums having been a member of two or three of them over at UKTS. For those not familiar with the concept it allows those most closely involved in a project to discuss feedback within a private forum, where they can discuss contentious issues without flames and digressions.

This forum is freely read without having to bother creating an account, and it's pretty clear that even banned members still manage to participate via sock puppet accounts... That simply doesn't happen on Elvas.
Elvas is closed to anonymous members, however Dave has been known to make exceptions for people known not to be a sock puppet.

During the development of Mid East Plus (which I did work on permissions and the stock pack for) we posted update or progress shots on the General forum at UKTS. This prompted discussion, and some obvious flames. The point is that the Flamers and trouble makers stick out like a sore thumb. You don't need to indulge them or waste time on them. The "Good Guys", as in this post, are obvious?

I reiterate my point that it's taken Geepster time and effort to compile a constructive and detailed report which he's posted (IMO not unreasonably) in a forum entitled "Open Rails Discussion". Here the point can be readily seen (as you point out) by anyone. Now if anyone involved with developing Open Rails can't use that information I'd like to see an official response from them as to why not?

The whole concept of OPEN RAILS was that it would utilise the expertise and experience built up over years of support from the MSTS community. If it is to succeed those now involved, who have achieved so much already, need to embrace as much feedback as possible?

I want to see Open Rails succeed as much as anyone in the Train Sim community but when I search to see if issues I'v observed have been reported elsewhere I seem to find too many dismissive comments from what I can only assume are "Fanbois" of the Sim who think that aggressively defending what has been achieved will in some way further development? ....when in fact it does nothing but hinder that development.

I don't want to pursue this thread any further. I've made the point and I hope that someone will appreciate Geepsters Efforts.

[baldwin]

There is a problem with reporting problems on the ET Open rails forum, unless you are already an ET member, because new ET memberships are closed until this problem with Mr X goes away. Further it iw wrong to say the OR dev team do not monitor this forum. Look through the threads and you will see quite a few posts by one or other of them at various times. The problem we have with signalling is that first of all the person who developed the OR signalling system is not very active atm and secondly while MSTS happily accepted errors in the scripts and displayed correctly, OR will display exactly what the script calls for which is why you see reversed switchstands for example.

The other problem with pathing may be related.

[OTTODAD]

The main problem with reporting BUGS directly to ORTS is that nobody else in this forum is aware of it and having similar problems will do the same, duplicating reporting them, albeit what has caused them possibly being different in every case !

Readers of posts in this forum will also not know what is being done about it by ORTS and judging by their 3 weeks X-updates since the New Year only 7 fixes have been released ! Only those who have reported them directly would know all of what they have covered.

For ORTS all this is a Labor of Love in their free time and the team will have to accept the consequences of not doing what most of its users expect from it !

[buttercup]

For ORTS all this is a Labor of Love in their free time and the team will have to accept the consequences of not doing what most of its users expect from it !

What do most users expect and what are the consequences? I'm curious.

[lateagain]

The main problem with reporting BUGS directly to ORTS is that nobody else in this forum is aware of it and having similar problems will do the same, duplicating reporting them, albeit what has caused them possibly being different in every case !

Excellent point Otto. If everyone is aware of a problem those with time can test it out and possibly add valuable additional feedback?

Readers of posts in this forum will also not know what is being done about it by ORTS and judging by their 3 weeks X-updates since the New Year only 7 fixes have been released ! Only those who have reported them directly would know all of what they have covered.

If everyone knows what fixes have been released and why it will stop unnecessary duplication of reports.

For ORTS all this is a Labor of Love in their free time and the team will have to accept the consequences of not doing what most of its users expect from it !

90% (at least) of MSTS freeware add-ons have been a labour of love and a large majority of the community have been more than willing to help, feedback and support those efforts? Lock 'em out of the frame and you lose all of that?

There are more generations and specs of OS and hardware spec out there in use by the ts community than you can shake a stick at. The more feedback that is made public the more folk will be able to self help.

[CNW 5099]

There is a problem with reporting problems on the ET Open rails forum, unless you are already an ET member, because new ET memberships are closed until this problem with Mr X goes away. Further it iw wrong to say the OR dev team do not monitor this forum. Look through the threads and you will see quite a few posts by one or other of them at various times. The problem we have with signalling is that first of all the person who developed the OR signalling system is not very active atm and secondly while MSTS happily accepted errors in the scripts and displayed correctly, OR will display exactly what the script calls for which is why you see reversed switchstands for example.

The other problem with pathing may be related.

Yeah, that's if you can get MSTS to work at all. That's the same program which will crash and burn if a parenthesis happens to be missing in one wag file. I'd rather be able to play the game with a reversed switch stand than not even able to open the darned thing. With MSTS so many workarounds were developed the community got Stockholm syndrome. Now that there's a program out there that will perform the tasks demanded of it in 2016, people are upset and clinging desperately to workarounds and jerry-rigs of an incomplete program from 2001.

Do you know what I must do just to use the AE editor? Load/store everything in Train Store (except activity to be worked on), boot up MSTS, close it out, then boot up the AE and find the one activity which I told Train Store to run and hope it loads. Activities must be generated by Activity Generator (great tool by the way), edited with ConBuilder, and edited in the AE via Train Store (as I described above). Hope I didn't use a car with a wag file MSTS will vomit at the sight of but OR displays happily.

Though I can say I too learned an MSTS AE workaround. I use Activity Generator to create a small activity with a few cars, I edit out those cars in ConBuilder with stock I like that the AE won't reject, then I load Train Store, boot up the activity, exit out of MSTS, fire up the AE, and then insert the small variety of stock (say I want to fill a yard, I'd use an SP box to indicate a cut of yard cars which would be replaced with any stock of my choice in Conbuilder I have onto various sidings that would correspond to that industry. I might change a path or two, clone and modify an AI if need be. After that I save, unstore everything, go back to ConBuilder and edit ALL of those loose consists. Are you confused yet? That is the MSTS experience right there. Even with OR, as long as any aspect of MSTS is needed, there will be headaches.

I will tell you this though, I built an activity on the Buffalo route that is too powerful for my computer. I have a GTX970 but I will need a better processor for the AI and pathing. MSTS could never run that activity, it wouldn't even start. It won't even load in the editor. I say all that to say I'm noticing a trend here; everytime Open Rails usurps MSTS in one category the diehards move the goalposts. At first it had no features, just ran the graphics better. When most features were MSTS compatible it was, yeah, but it doesn't have enough real world train operations. Now that it's added comprehensive braking, train operations, and a massive ongoing steam simulation as well as superior AI pathing and signalling it's "well, the skies aren't pretty enough."

From people who have stuck by an unfinished program from 2001, you're a picky lot. People got PAID to produce what would be laughed out of the marketplace had they pulled that stunt in 2016. OR is FREE and a volunteer project. A lot more respect needs to be shown. Might be why most of the staff prefers ET, ever thought of that?

EDIT: And this isn't too attack you specifically Baldwin, just venting as there are some regular naysayers around these parts.

[lateagain]

Yeah, that's if you can get MSTS to work at all. That's the same program which will crash and burn if a parenthesis happens to be missing in one wag file.

............etc.etc.
Well I hope that makes you feel better?! Doesn't have anything to do with this thread though that I can see?

You seem to have decided that MSTS and Open Rails are followed like Hockey, Football or Baseball by groups of fans who, having devoted themselves to A Team, just live to rubbish their opponents? Not how most of us see the situation at all!

You also seem to be making rather a mountain out of a molehill when it comes to using MSTS ...but horses for courses.

Any one with two brain cells wants to help the development of Open Rails but CRITICAL FEEDBACK is essential to the development of ANYTHING. Remember this fundamental : "....not everyone who agrees with you is necessarily on your side, and not everyone who disagrees with you is necessarily against you".

[geepster775]

Personally, I don't see this as an "MSTS clingers" issue. Its just that when the old product can do something prototypical that the new product is less successful at, it is not improper to speak up about it. Maybe 80% of the user base could care less what train holds the main. Yea, that clearly shows in the activity "product" produced to date. Until now, the 20% who did care found their own solution and didn't bother the rest of you. Is that being extra picky? You bet. Get used to it.

I don't have any other comment on the ET situation other than to say it is a bit counter-intuitive when the intellectual core of the project prefers to hang out in a man-cave with restricted access that makes it difficult for those with abundant knowledge on the prototype to inject such into the discussion. More so when the restrictions are driven by localized man-cave politics and are not related to the Open Rails project in general. And as long as the OR team prefers that ham-strung arrangement, then they maintain the obligation to keep their heads up on other forums. If that becomes too big of a burden, then a gentle reminder that they are the only ones possessing the power to change that is in order.

As for the original problem, aside from Baldwin's suggestion that it may be related to the signalling failures, another possibility is we are seeing elements of the Open Rails improved anti-standoff mechanism at work here. I think deviating from the path is part of that "solution". But it is one thing to design an algorithm that successfully deals with single track and sidings less than 2 miles long, its another thing when you attempt to make that algorithm work with double track and crossovers every 15 miles or with (more challenging) 60/40 single track routes, which are common in some US locations. The Black Hills Sub portion of the Orin Line is an example of the latter, where there are double track stretches with crossovers every 8-12 miles followed by pieces of single track. The algorithm needs to recognize the 12 miles between crossovers as a potential siding. And I don't think it sees that far ahead. Absent an algorithm capable of that degree of vision, it is left to the users to define what they want done via the path and hope the algorithm doesn't over-rule them. It could very well be the case that the algorithm is okay but we simply lack the sophisticated AE to go along with it, where users can set differing priority of revenue trains over non-revenue trains and let the algorithm carry out the tasks properly.

Make no mistake, this program needs to be put through its paces and stress-tested to find its weaknesses in all geographic locales. Nobody needs to apologize for that.

[csantucci]

As far as I know, if the player path has no passing path at the diverging point and if the location-linked passing path processing option is not selected, OR runs the player train on its player path without making different decision. If location-linked passing path processing is selected or if the player path has a passing path at the diverging point, OR can let the player train run on a different path.

[R. Steele]

Yeah, that's if you can get MSTS to work at all. That's the same program which will crash and burn if a parenthesis happens to be missing in one wag file. I'd rather be able to play the game with a reversed switch stand than not even able to open the darned thing. ...

HUGE THUMBS UP!!!! Especially the part about moving the goal posts.

............etc.etc.
Well I hope that makes you feel better?! Doesn't have anything to do with this thread though that I can see?

Thanks, that really got me laughing, you really think your response is about "Open Rails does not follow Players Path"? Come on folks quit kidding yourselves...this thread is so off track it looks like an MSTS trainwreck!

Has anyone filed a bug report?
Has anyone checked the same route under the same conditions and duplicated what geepster found?
Has anyone with signal script expertise or path experience checked the route and activity to see if the problem is with either pathing functions or route problems.

Post#24
Personally...Make no mistake, this program needs to be put through its paces and stress-tested to find its weaknesses in all geographic locales. Nobody needs to apologize for that.

Back on track...except for the "man-cave" bits, but that's a funny enough and well written, so what the heck.
I really hope you find a solution...and if it does show a problem within the OR code it will require someone with the interest and coding knowledge to volunteer to fix it. Progress in a volunteer effort takes time but it is obvious, at least to this low level user, that excellent progress has been made in OR.

[baldwin]

As far as I know, if the player path has no passing path at the diverging point and if the location-linked passing path processing option is not selected, OR runs the player train on its player path without making different decision. If location-linked passing path processing is selected or if the player path has a passing path at the diverging point, OR can let the player train run on a different path.

In the Feather River activity mentioned earlier in this thread, the player is sent onto the diverging path with nothing coming in the opposite direction. This happens with or without location linked passing selected. Far as the player path is concerned, this should be followed, but it seems that OR is treating an AI running in the same direction the same as if it was running in the opposite direction. Instead of holding the player train at a red, it uses the alternative path into the siding which has a red at the far end because the exit is still blocked by the preceding AI.

[lateagain]

Mr Steele,

Are you part of the Open rails Development team and if so what part do you play in it?

[crstagg]

Mr Steele,

Are you part of the Open rails Development team and if so what part do you play in it?

He is not on the list: http://www.openrails.org/discover/project-team/

[conductorchris]

Geoff, are you asking because you saw his signature line? Note that it just identifies him as an open rails supporter, not part of the development team.
Christopher