Bug report for XTracks

Firt of all let me say, I love XTracks. Without you can't create a layout looking just a bit like real. But there is a bug inside (or I'm to stupid) with the curved switches 500r8 both right and left.
The 500r8LftMnl creates a right hand one and the 500r8RgtMnl creates a left hand switch. Sounds like nothing if you know this bug, you can remember it and thats it.
But the real problems comes up when you ride the route, especially in explore mode. If you encounter a curved rigth hand switch the F8 display shows you a left hand one ahead that is set on the straight track as default. In truth the curved right switch ahead is set on the very right turnout track.
The reason is, you inserted a right hand switch with she shape of a left hand one. Ok, enough with left and right? Just test it yourself, may be I'm only to stupid, but I'm sure not.

May be its to late to fix this bug before XTracks gets ready for MSTS2, but it would be nice to remember this bug for the future.Mike

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

RE: Bug report for XTracks, calr, Jun-16-03, 12:58 PM, (1) RE: Bug report for XTracks, geoffmayo, Jun-16-03, 12:59 PM, (2) RE: Bug report for XTracks, 020314, Jun-16-03, 01:31 PM, (3) RE: Bug report for XTracks, OkrasaGhia, Jun-16-03, 01:41 PM, (5) RE: Bug report for XTracks, geoffmayo, Jun-16-03, 05:16 PM, (8) RE: Bug report for XTracks, OkrasaGhia, Jun-16-03, 01:38 PM, (4) RE: Bug report for XTracks, calr, Jun-16-03, 01:45 PM, (6) RE: Bug report for XTracks, 020314, Jun-16-03, 04:28 PM, (7) RE: Bug report for XTracks, geoffmayo, Jun-16-03, 05:18 PM, (9) RE: Bug report for XTracks, OkrasaGhia, Jun-17-03, 01:55 PM, (10) RE: Bug report for XTracks, geoffmayo, Jun-17-03, 02:11 PM, (11) RE: Bug report for XTracks, trainsimming, Jun-17-03, 07:12 PM, (12) RE: Bug report for XTracks, davidbeach, Jun-18-03, 03:14 AM, (13) RE: Bug report for XTracks, griffinmt, Jun-18-03, 09:20 AM, (14) RE: Bug report for XTracks, davidbeach, Jun-19-03, 00:43 AM, (16) RE: Bug report for XTracks, griffinmt, Jun-19-03, 09:13 AM, (18) RE: Bug report for XTracks, jdtrainman, Jun-19-03, 10:26 AM, (19) RE: Bug report for XTracks, OkrasaGhia, Jun-18-03, 02:18 PM, (15) RE: Bug report for XTracks, 020314, Jun-19-03, 03:53 AM, (17) RE: Bug report for XTracks, OkrasaGhia, Jun-19-03, 12:07 PM, (20)

Conferences | Topics | Previous Topic | Next Topic

The naming of the curved switches is confusing. The 500r refers to the radius of the main line or default path. The l or r designation is the diverging route as you look at it from the points end. If it is a left-hand switch and the main curves to the left, the diverging route will go to the inside of the curve and have a radius of less than 500 meters. If it is a right-hand switch with the main curving to the left, the diverging route goes to the outside of the curve and has a radius greater than 500 meters. Reverse all the above if the main line curves to the right. It is very confusing but unless Okrasa changed it during the last revision, the nomenclature is correct.Cal Rasmussen
Beaverton, OR
cal.rasmussen@verizon.netColumbia Gorge Route (Phase 1 80% Completed)

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

The confusion (and I'm not saying who's got it wrong) is with the definition of left and right when on bends. The way it should be considered is which way the normal route goes (ie points/switch in the normal position) and which way the turning off route goes (ie points/switch in the reverse position). If, when looking at the common/toe end of the point, both tracks curve to the right, some might consider that a right. But if the normal route is to the extreme right, then it could be considered a *left* hand switch, since the deviating route goes to the *left* of the main route.See attached pic. Are they both right-hand switches just because they curve to the right? Or is the first a right-hand switch and the second a left hand switch because of the path of the main, normal position route?Geoff M.

Attachments

http://www.train-sim.com/dcforum/User_files/3eedf73c0edf18d3.gif

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

I don't know what you are talking about. Do you want to tell me the direction of a switch is determined by the mainline that passes the switch?
I hope I misunderstood you. A left hand switch keeps a left hand switch. Done!
No matter if there is any mainline passing thru or not, no matter if it is just layed or still in the factory. What would you like to order from a rail factory? A left right hand switch because your mainline goes to the left? They would ask you going home till you know what you want. But the problem keeps alive. And the F8-window understands the problem very well, a left hand switch is a left hand switch, because it is it's name and so it is displayed in the F8-window.Look at the attachment to see how switches are called. Never ask for the mainline. Trust me, I know what I'm saying.

Attachments

http://www.train-sim.com/dcforum/User_files/3eedfe3729e3e935.gif

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

> Trust me, I know what I'm saying.
No you don't. See my reply message #4.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

LOL!>I don't know what you are talking about. Do you want to tell
>me the direction of a switch is determined by the mainline
>that passes the switch?
>I hope I misunderstood you. A left hand switch keeps a left
>hand switch. Done!
>No matter if there is any mainline passing thru or not, no
>matter if it is just layed or still in the factory. What
>would you like to order from a rail factory? A left right
>hand switch because your mainline goes to the left? They
>would ask you going home till you know what you want. This is a game, not real world! We don't have factories, unless you call Okrasa a factory, which I'm sure he'd be less than impressed with!Let me try to explain in simple terms about normal and reverse. In signal control tables, signal logic requires points (switches) to be controlled, locked, and detected in a certain position - either normal or reverse. The UK uses the terminology normal and reverse, as does (at least parts of) the US. Some European railways use Left and Right positions. The Normal position is usually the fastest route, the mainline path that I referred to earlier. The exception is derailers, catch points, and trap points where the normal position leads you the fast way to biting the ballast (ie off the track).As far as Real World (TM) is concerned, no there is no distinction in a curved point between left and right. But, again, if you look in control tables there will be a normal and reverse position for control and detection purposes.In the ideal MSTS world, a left hand, right curved point (ie both paths curving to the right), MSTS would show a left arrow if you were on the reverse path. A right hand, right curved point would show a right arrow if you were on the reverse path.>But the problem keeps alive. And the F8-window understands
>the problem very well, a left hand switch is a left hand
>switch, because it is it's name and so it is displayed in
>the F8-window.
>
>Look at the attachment to see how switches are called. Never
>ask for the mainline. Trust me, I know what I'm saying. I'm not entirely sure you do... have a look at my credentials on http://www.simsig.co.uk and http://www.theraileng.co.uk . If you don't want to, basically I developed a signalling simulator used by 40% of the UK's computerised signalling control centres to train and assess their signallers.Geoff M.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

This is not an error but a matter of nomenclature.
As Cal and Geoff are saying you should not be confused by the main track going left or right. The left or right designation of the switch refers to which side the diverging track goes. In the 500r8dLftMnl the main line is 500r to the right and the diverging track is slightly less to the right (actually a 700r8d curvature) making it a left hand switch. Think of a crossover made out of these switches on a two track curve and then picture straightening the curve and it might become clear.Lastly don't be fooled by the F8 switch indicator. It does not work as it should, only reason it works with the original switches is because Kuju made a 'hack' to show those correctly (I've investigated it).Thus it is not a bug but by design.Ready for MSTS2? I'd say MSTS2 is still at least six months away. Sure they say fall 2003 but that is the marketing department and they are at least three monts early. When was the last time you picked up a game the date it was innitially told to ship? Before it is released and I can get my hands on a copy I have no way of telling what needs to be done to make XTracks work in MSTS2.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

The F8 switch indicator doesn't work properly on some of the default switches!Cal Rasmussen
Beaverton, OR
cal.rasmussen@verizon.netColumbia Gorge Route (Phase 1 80% Completed)

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Thanks Okrasa,your explanation is surely right, but looking for a piece by name referring to the name it has in the real world, it still looks like a mistake. Let me describe: in the real world the standard is a switch with a straight track and a curved track going either to the left or the right. The direction of where the curved track goes determines the name of the switch either left hand or right hand. A special shape is the curved switch, if I would translate the German (thats my native language) name it would be called as 'inside curved switch', where the straight track is substituted by a lesser curved track that goes to the same direction as the more curved track. If you follow this definition you might understand my trouble, because the lesser curved track is considered as 'nearly straight' to let the more curved track piece (following the same direction, otherwise it is a Y-switch ) show the direction left or right. The description is nearly 1:1 translation of a school book for railroad employees. If I uderstood it right you had to call the switches as they are called now to make them working proper. So I have to live with this.You also wrote that F8 often does not work proper. That I never saw. I just biult some routes, first only with MSTS standard track pieces ( and now with XTracks ) . All together some hundred switches and F8 N E V E R showed any wrong sign beside these curved switches. And also here F8 tries to show the right sign.
Again imagine a 500r8dLftMnl which turns both tracks to the right (I hope we agree ). Because of the more curved track is the basic one it is determined as default direction and be shown as a green arrow. Because of the name the red arrow shows to the left, so let's say the green one is the one 'more to the right'.
Now look outside the front window, you see a right hand switch where you will run over the more curved track which is on the very right.
Still now F8 tries to work fine, only handicapped by the name of the switch. OK, enough with switches.
Have you ever thought about deviding ballast and ties into separate 3D-objects which lay close together - lets say 1 mm - to avoid drifting the ballast under the ties when you move the train. I thought about it because I'm very unhappy - like many others - with the disappearance of the ties when you put tracks together building -oouuch! - switches. I know that some people try to solve this problem, but the best way looks to avoid it. Would it slow down the frame rate very strong? Is there any other problem ( impossible to connect track pieces )? Mike

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Beta testing starts in July...Geoff M.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Mike the naming of the switches is not forced but by convention.
I could have called them pig and cow (though David would not have been impressed). As I said the F8 view is not working correctly. It is not reading the name trying to find a 'Lft' or 'Rgt'. It simply shows the diverging track always on the same side while you have the main path selected. When you hit the 'G' key selecting the diverging path it will correct itself. You can try this with a pair of 3d switches, when the main arrow is green they will show the diverging track on the same side.
Kuju solved this for the standard switches by doing a check in the F8 view to see if the internal id of the track section is that of a left hand switch to draw it correctly for those. I don't understand your question about dividing the ballast?Geoff, beta testing in July? Might be though most of the beta testing MS leaves to the customers (and get paid for it).
Even if MSTS2 goes into testing in July that is not saying it will be released in September/October. I maintain that the earliest would be X-mas.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Kuju contacted me about beta testing. I left the ...... at the end since I wasn't hazarding any speculation so you may well be right!I think the reference to the ballast is basically to have two layers of texture, one for the ballast and one for the ties/sleepers. The latter would lay on top of the ballast and be transparent between the ties, although there would have to be a much greater difference than 1mm else they might start meshing together and causing nasty effects once they start getting further away from the camera.Geoff M.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Surely that should be cow and pig?

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>I could have called them pig and cow (though David would not
>have been impressed). I might be amused, depending on how the mood struck, but I am hard pressed to imagine what could possibly get either pig.s or cow.s included in the standardized file.As cryptic as the file naming may seem at first glance, once you get the hang of it, it is fairly easy to find what you are looking for.David Beach.Palouse & LewistonRoute Building 1% Completed
Route Mapping 10% Completed
Route Photo Map 100% Completed (Now, Can I make the Railroad Profile match the photo map?)

Standardized \global\tsection.dat file now at build00019.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Besides, we already have a cow.s Martyn T. Griffin
griffinmt@mindspring.com
http://www.geocities.com/tcmrrc/

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

>Besides, we already have a cow.s Yeah, but its listed in the .REF file and goes in the \route\shapes\ folder, we were talking about the \global\tsection.dat file and a cow.s that would go in the \global\shapes\ folder.
David Beach.Palouse & LewistonRoute Building 1% Completed
Route Mapping 10% Completed
Route Photo Map 100% Completed (Now, Can I make the Railroad Profile match the photo map?)

Standardized \global\tsection.dat file now at build00019.

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Unless you want some static track and try putting it into the routes shape folder!Martyn T. Griffin
griffinmt@mindspring.com
http://www.geocities.com/tcmrrc/

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

We've got cows.....Where's the tornado?

Jim Duncan
Team GM&O

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Are we starting a pig & cow thread here?Getting back on track (pune intended).
If Geoff is indeed correct about Mikes suggestion for doing the ties and ballast as separate textured layers with the ties sligthly above the ballast I don't think it would work.I made experiments with similar techniques a year ago but the way MSTS handles drawing of tracks prevented success.
Have a look at this recent thread: http://www.train-sim.com/dcforum/DCForumID6/9739.html#10


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Hello Okrasa, I looked at this thread and remembered that I did already before. When I read this I noticed that one problem were substituted by another. I think that is not the way to make MSTS getting closer to the real world (I think that is what all of us are looking for).
As you wrote yourself, the problem of disappearing shadows comes up and disturbs the illusion. So I thought, why not to catch this problem by its root, and that is merging both ballast and ties in one level. I surely believe if you devide these in two layers the track pieces never would work together with Kuju's basic stuff. Thats another problem. So excuse me, I'm not familiar with creating objects, but it should be possible to tell one layer - i.e. the ties, because its closer to the rails - to be the basic one, which means it will proper connect with its neighbour. So the ballast would be only additional stuff that has nothing to do with the connecting procedure.But it is only an idea.MikeP.S. Okrasa, regarding our private communication, why do you wait for an email, don't you have THEIR email?

Alert | IP

Printer-friendly page | Edit | Reply | Reply With Quote | Top

Dividing the ballast and ties in layers does not help.
Maybee if you do the ties without alpha layers, that is single ties separately textured but then the FPS is bound to take a hit and I don't want to think of the process of creating all the track sections with individual ties.
You see the sim only draves the alphaed textured of one of the sections when overlapped. I think it decides which to draw by the order they were layed in, it does not matter which one is above the other. Yes it is real weird all this. The connecting procedure is something completely different from the shape, that is defined in the tsection.dat.About the e-mail, I'm not waiting for one. I don't expect to be contacted neither do I think an e-mail from me would lead to anything. Though if they want to contact me they can do that.


//Okrasa
Chief Engineer
okrasaghia @ yahoo.com
http://www.geocities.com/okrasaghia
http://www.geocities.com/okrasaghia/XTracks

Alert | IP

[ Home | MSTS Forum | MSTS FAQ | File Library ] Copyright © 2002 by Nels Anderson. All Rights Reserved.