Modular Approach to Trainz like N-Trak in Model RR?

[Rivkah]

I am a newbie. Years ago I rode the "State Fair Limited" pulled by N&W's J 611. While at the Ohio State Fair I stumbled across a pavilion that housed model RR and was intrigued by modular railroads.

Is it possible to develop interconnectable and reconfigurable Trainz layouts using a variety of modules? Could these modules be taken apart and put together differently? Would this allow smaller and larger contributors to add to the robustness of Trainz? Could a modular approach combine fictional and factual modules? Does the architecture of Trainz already lend itself to such an approach? What could the core developers do to further enable such an effort? Could such an approach allow for a train running on a NEC module to flow to a Harrisburg module to Horseshoe Curve to Chicago via a number of existing projects or 'grand' modules? Though I don't know much about N-Trak could we use it as inspiration for such a concept without the rigorous rigidity? Are there standards to join one module to another?

As a newbie I have lots of questions. Would you please help me understand? Thanks for your time and thoughts!

In Peace,
Rivkah

 

[JIb228]

Yes and no. You can merge any route to any route. but you cant unmerge them. There is a group that I am part of in Trainz that is building a modular layout for MP reasons. Each of us builds a small section and adds it to the may route. This way, the route slowly grows.

 

[mjolnir]

Yes, one could develop interconnectable and reconfigurable modular layouts in Trainz. Yes, it could allow combining fictional, factional, and all combinations in between. And while the architecture of Trainz does not prevent such developments, there are some things that cause Trainz to be a bit less useful in such endeavors that it might be. One example: in an N Trak modular layout, you could rake each of the four corner modules, and place them at each of the other 3 corners. This is not possible in Trainz, as once created, a Trainz route cannot be rotated.

The other issue with a modular layout has more to do with the hardware of the individual participants. As far as I can tell, Trainz loads an entire route into memory at once, so the size of the modular route would be limited to the maximum size of a route that can be loaded into the machine with the least resources among the participants.

On the other hand, there is a facility within Trainz (which I have not personally used, and upon which I cannot comment further), the iportal. Having an iportal on both routes, if I understand it correctly, would allow a train to exit an iportal on one route, and enter another route through an iportal on that route. The iportal eliminates the need to have exact dimensions at the connecting edges of the modules.

ns

 

[JCitron]

This is very possible especially if N3V fixes the merge feature in TS12. In the past, starting with TRS2004 I think, there was a group of creators that did a UMR series. This was the "U-Make Railroad". There are a bunch of little modules up on the DLS that consist of small corner pieces, rail yards, etc. This was a great idea but it didn't go quite as far as you want it to. The modules were built around a specific standard almost like the N-Trak modules you saw at the train exposition. The tracks were all at zero height and were a specific distance from the edge. This is a great idea if you think about it because you can have anything you want from yards to cities. Basically these were just some simple decorations and some track with the thought of leaving the rest up to the route creator to use his imagination. There were a couple of bigger modules which were based on actual areas with one of them being a big yard on the Norfolk Southern in VA. I used that module, added some stuff to it, rearranged some track, etc., to make it mine.

The merging has worked out well, but the important thing to do is to ensure that the texturing and height is consistent. Texturing is easy, but the height and placement presents a problem because neither of these things can be altered on the baseboards. Oh, height can be adjusted by squishing down or raising up the surface, but the overall height difference can't be changed. In order to match different heights in and tracks facing different directions, baseboards have to be added in between. These "tween" baseboards need to be blended so there is a transparent blending between the different routes. It presents some challenges, but this can be done and done well.

The alternative, or even an adjunct to directly connecting the routes together is using the iPortal where Auran's iPortal server is used to "ether" connect the rail lines. With the iPortals, routes can be different heights and styles because no blending is needed. A train is driven into an iPortal and then comes out on another. I've never used them, but others have and will probably have a better details about these than I have.

John

 

[teddytoot]

If you search the DLS on my username (tick all boxes as they were for early versions of Trainz) you will find some examples of these modules. There are a lot more from other creators but the idea petered out as with the improvements in Trainz their limitations showed up.

 

[Fabartus]

I am a newbie. Years ago I rode the "State Fair Limited" pulled by N&W's J 611. While at the Ohio State Fair I stumbled across a pavilion that housed model RR and was intrigued by modular railroads.

Could such an approach allow for a train running on a NEC module to flow to a Harrisburg module to Horseshoe Curve to Chicago via a number of existing projects or 'grand' modules? Though I don't know much about N-Trak could we use it as inspiration for such a concept without the rigorous rigidity? Are there standards to join one module to another?

As a newbie I have lots of questions. Would you please help me understand? Thanks for your time and thoughts!

In Peace,
Rivkah

I just sent you an email offering help on learning curves per your post on the standards and code of conduct. Just so happens the Harrisburg-west through Horseshoe curve is being discussed as a group project by email by the Old Trainz group this last day or three. It looks like we may start there and do such modules in sub-groups giving us a project which while initially centered in Eastern PA, would be extended West soon or late. The horseshoe curve is off interest to many, and as a Pittsburgh area native, I will certainly be going that direction eventually.

The iPortal of TRS2006 and improved versions can send trains from one user to another on the route over the internet. So yeah!

Frank

 

[cascaderailroad]

snip~Just so happens the Harrisburg-west through Horseshoe curve is being discussed as a group project by email by the Old Trainz group this last day or three. It looks like we may start there and do such modules in sub-groups giving us a project which while initially centered in Eastern PA, would be extended West soon or late. The horseshoe curve is off interest to many, and as a Pittsburgh area native, I will certainly be going that direction eventually.

The iPortal of TRS2006 and improved versions can send trains from one user to another on the route over the internet. So yeah!

Frank

Count me in on this project, as I also predomenetly use TRS2006, and iPortal.

As with N-Trak and HO-Trak modules, fine specs would have to be adhered to, as to track placement(s).

The modules could be joined by small sections of straight tracks (Just like on N-Track modules).

Spiral Helix modules could be incorporated, not only, to raise trains to a higher elevation ... but to give extended time for a train that goes into a tunnel, that it does not emerge from another tunnel, for quite some time delay, later.

 

[SuperFudd]

You can "unmerge". I had a layout I had been making a long time disappear. No, I was not clever enough to have made a backup. Fortunately I had recently experimented with merging to the layout so I deleted the unwanted parts.

 

[PWeiser]

The Standards Module?

Having recently (re)started building a layout, I've been trying to "standardize" modules much as discussed. A suggestion (if it wasn't used in UMR): create a "standard module" of the desired mainline width (two or three boards) with a six-track mainline running across the middle and all tracks stopping just short of the edges (10m or so). No elevations, of course. The rule, then, would be that modules can be anything at all... so long as they can connect with the standard module at each end (which might involve moving the standard module laterally, etc.). Merge with from one (center) track to all six... Could even include some catenary/trolley wire to enforce a height standard...

One thing I've found in merging modules is that if a texture is set for "all boards" in Surveyor and the route saved, then the route merged into another, the texture is absent at the board joins (though it was fine when saved and, in fact, when run or brought up again in Surveyor). Of course this may have been fixed - I'm still at TS2009 SP2.

One final thought: if TS2013/14 does not fix the rotation-to-merge problem, could someone like PEVsoft develop a transform program which would rotate a (simple!) route 90 degrees? Should be simple math, given understanding of how the coordinate system works. (And of course N3V could then incorporate it into future versions - which would be fine!)

 

[mjolnir]

Having recently (re)started building a layout, I've been trying to "standardize" modules much as discussed. A suggestion (if it wasn't used in UMR): create a "standard module" of the desired mainline width (two or three boards) with a six-track mainline running across the middle and all tracks stopping just short of the edges (10m or so).

You don't need to define a standard board; you simply need to define two standard edges of a set of boards.

One final thought: if TS2013/14 does not fix the rotation-to-merge problem, could someone like PEVsoft develop a transform program which would rotate a (simple!) route 90 degrees? Should be simple math, given understanding of how the coordinate system works. (And of course N3V could then incorporate it into future versions - which would be fine!)

I would not expect that the "rotation-to-merge" problem will be fixed until such time as Trainz is ported to a new game engine, or there is significant time and effort (and money) spent in a major rewrite of the Jet engine. I had a situation where I started a route, and decided I didn't like the lighting that was going to occur when the route was oriented to the East to West axis, and wanted instead to have the route oriented North to South. I had spent enough time on the route that I (naively, it turned out) thought that I could make a route rotation utility in less time that it would take to redo the entire route. The time I spent investigating the file structures of test routes showed me that the inability to rotate a board or route is either an unfortunate limitation of the design of the game engine (if the subject of rotating a board were discussed in the development) or an unintended consequence (if it was not).

ns

 

[slabay]

You don't need to define a standard board; you simply need to define two standard edges of a set of boards.

I have actually thought of doing a similar thing myself. The easiest way for the standard board would be to create the "standard board" with the "standard track" running across it N-S and E-W. Then, whoever was making a module could simply use the "standard board", merge a few for a 2-3 board module if desired, delete the unnecessary track in the middle, and build to their heart's content. This way, whatever they do in the middle of the board(s) is fair game, as long as the edges remain consistent.

By going in both directions (and since Trainz cannot rotate baseboards), we can make an east-west mainline or a north-south mainline, we can make corner modules to change directions, we can delete the east-west to make a north-south through module, and so on. If we take the ball and run with it, I'm in!! The TERRIBLE thing about route building (look at the routes that I am working on!) is that you get overwhelmed with the scope of a project. Working on a single board module allows them to actually get finished by different people, and can be merged into a larger layout.

I do suggest that we stick to a standard naming convention for people to participate (like the old UMR), so that anyone who wanted to do anything with the system could easily search the DLS for the components: I propose creating something like "Standard Baseboard (V-TRAK)", then anyone who uploads a module could upload it as "SteelVille Steel Mill (V-TRAK)". Now I come along with my route that I want to make and can simply search the DLS for V-TRAK and find compatible boards.

I actually like this idea enough that I think I'll make a standard board, make it available for people to try out, and see what happens....... I'll have something available for people to play with in a few hours!

 

[leeferr]

I've thought about this whole process before and I'm sure that standards can be set up for e-w and n-s modules, but I'm just not sure how changes in elevation would be standardized for adjoining modules. N-Trak is flat and unless we're looking for flat Trainz modules, it appears to be a very difficult set of standards to set up. Does anyone have any ideas on how to overcome this issue?

Mike

 

[amrail2000]

You know, for once I'm going to ignore every post here and say that aside from other tell you, YES IT IS. Now it may take some work, but every route can be merged. Now if there height differences then you know that you will have to adjust that. Now on my forums I have several sets of two track modules that can be linked up with no trouble. After connecting the modules all you need to do is just put a short section of track in. These will create any route you want.

 

[slabay]

I made a sample standard baseboard and put it up on my website until we have a "standard" to put on the DLS. If anyone can think of anything that needs to be changed on it, let me know. This has a 4-track spacer on each corner and in the center of each board going each direction. This will give us a bunch of flexibility for how to lay out the modules since they can't be rotated. If you want to do a 2-track main, use the center 2 tracks of the 4 track spacer. To join the modules, simply merge baseboards, delete the spacers, and attach track between them. I fired up my old copy of TS10 so that older version users can make stuff on it as well (because I don't have TS09 installed anymore). They of course can't use a module built in TS12, but they can build in TS10 and TS12 users can use the modules.

I don't know about changing elevations: I think that in order to create a standard, we will need to stick to a flat and boring 0 height for the most part unless we go with a series of boards that get their own altitude (i.e. make a mountainside of 3-5 baseboards going up to a logging camp with the line coming down to meet the "mainline" level). We can certainly have a height change (mountains, gorges, whatever) across a board(s) as long as the track ended on a standard edge (or ends within the board). Examples here being a board (or 2-3) going over a mountainous pass or a tunnel through mountains, meeting the standard on each end or a spur ending in a port module.

Feel free to download it and play around with it, tell me what you think. We can make changes to the "standard" before uploading a final version of it to the DLS.

 

[PWeiser]

Yes, Like That!

I made a sample standard baseboard and put it up on my website until we have a "standard" to put on the DLS. If anyone can think of anything that needs to be changed on it, let me know. This has a 4-track spacer on each corner and in the center of each board going each direction. This will give us a bunch of flexibility for how to lay out the modules since they can't be rotated. If you want to do a 2-track main, use the center 2 tracks of the 4 track spacer. To join the modules, simply merge baseboards, delete the spacers, and attach track between them. I fired up my old copy of TS10 so that older version users can make stuff on it as well (because I don't have TS09 installed anymore). They of course can't use a module built in TS12, but they can build in TS10 and TS12 users can use the modules.

I don't know about changing elevations: I think that in order to create a standard, we will need to stick to a flat and boring 0 height for the most part unless we go with a series of boards that get their own altitude (i.e. make a mountainside of 3-5 baseboards going up to a logging camp with the line coming down to meet the "mainline" level). We can certainly have a height change (mountains, gorges, whatever) across a board(s) as long as the track ended on a standard edge (or ends within the board). Examples here being a board (or 2-3) going over a mountainous pass or a tunnel through mountains, meeting the standard on each end or a spur ending in a port module.

Feel free to download it and play around with it, tell me what you think. We can make changes to the "standard" before uploading a final version of it to the DLS.

Yes, just like that (I do still have TS2009, though, so I don't think I can use your sample). All corner modules will be 90 degrees and of four kinds (E to N, N to W, etc.) In fact, every module will be one of 6 kinds (add W to E and S to N to the above) unless the builder decides to do a 180 - 270 would be problematic since it limits what it can be attached to.

Incidentally, it takes 4 boards to go up just 16m at a 0.7% grade - very mild by logging railroad standards!

 

[Ricke82]

I guess it’s time to come out of the closet and admit that I’ve been playing around with the modular concept off and on for the last couple of years. It’s all based aound two or four board straight sections which can be merged on the flat edge or the corners:

Uploaded with ImageShack.us

Any straight section can be connected to any other straight section by curved or junction modules of 22.5, 45, 67.5 and 90 degrees (the angles are only approximate in most cases). Because you can’t rotate baseboards, the number of modules is twice and, in some cases, four times what is would be if you could (I currently have almost 250). I’m not suggesting this as any kind of standard, but it may give someone an idea that’s useful.

 

[cascaderailroad]

I think the NS, EW, is the best way to start out, designing modules, with single, double, or quad tracks running directly down the exact centers of the baseboards, possibly with a 4m track spacing ?

Or a 90 degree curve baseboard module, entering on the E, and exiting on the S ... etc ... etc ...

Or a 4 point cloverleaf baseboard, with N, S, E, W, exits

Or "S" curve baseboards, with exits running off the baseboard centerline

I think there are some old ones on the DLS

 

[Rivkah]

I have noticed that some folks are very good at building bridges or structures while others are good at track work and still others good at scenery. My ideas of modularization would capitalize on those strong points.

For example consider three modules: 1) a center "Track Module"(1) and two flanking "Scenery Modules" (2 & 3).

North (or Up) "Scenery Module" (2) One or more boards wide. Could be a cityscape, mountains, or rolling farmland with one potentially substitutable for the other.

Middle "Track Module" (1) with track running E to W One board wide.

South (or Down) "Scenery Module" (3) One or more boards wide. Possibly many modules could be combined to match track module length.


As another example, consider short "Track Modules" for a highly detailed bridge, trestle, or other amazing feature. The match between modules would be one board edge.

Another thought would be to have a two by two board (or more) "Track Module" with an S-Turn. Such a module might begin on the W side of the Westmost top (or NE) board and ends on the E Side of the Eastmost side bottom (or SW) board. These S-curves would break up the extreme linear view that would result from many Track Modules being joined in a line.

Being able to rotate a module by 90, 180, or 270 degrees when assembling a "temporary" route would be a plus. One could be easily do this by swapping X and Y axis (or a negative X or Y) and wouldn't require complex mathematics.

Anyway some fodder for thought. What do you think?

 

[Fabartus]

Count me in on this project, as I also predomenetly use TRS2006, and iPortal.

As with N-Trak and HO-Trak modules, fine specs would have to be adhered to, as to track placement(s).

The modules could be joined by small sections of straight tracks (Just like on N-Track modules).

Spiral Helix modules could be incorporated, not only, to raise trains to a higher elevation ... but to give extended time for a train that goes into a tunnel, that it does not emerge from another tunnel, for quite some time delay, later.

Actually me thinks you over complicate the modular needs of Trainz, speaking also as someone with most of my experience in TRS2006... Consider a bendy track really doesn't care how long it is. It is designed to curve gently away from it's joins. If your landscape is more or less the same height, and you end your two tracks ends at least 1/4 board from the edge, or habitually extend a portal off the edge as I frequently do during an intermediate stage (Learned that Trick from HP Trainz on Marias Pass Approach), you have a lot of distance to connect up to give a mild slope when merging. And that same distance will keep your alignment needs to 'close enough for government work'. Don't over complicate it. Do like the Russians against Napoleon, use distance to defeat the problem. What's an extra board length when the engine will run the 150 mile main route plus it's several 25-30 mile spurs like Marias Pass?

In other words, use an extra board or two and join the two ends farther apart. When the merge is done, use the track height tool to shape the slope of the roadbed. Use the terrain plateauing tool swiping at right angles to the slope of the tracks to smooth out the lands to either side. Or first raise the roadbed by paralleling the road a few squares over with a track, then adjust it's height below the road grade, preferably with an eye on watercourse installation too.

Those I shape out of valleys just like nature, and I always keep in mind that many railways follow watershed valleys, so water features are common along the tracks, even if it's only a small stream. Tracks need drainage too! Shape the land relative to that height, punch the creek under the track a couple of times winding it's way down slope and put short bridges on your joining boards.

Take for example the Tidewater North and Tidewater South maps in TRS2006, I made an intermediate extended region between them adding about 5-7 boards N2S 4-6 board across, allowing room for a wye and spur above the bridge to that south yard at Mapleton. As I worked the south section, had two portals up there ala' Marias Pass Approach usages/style, and did the same off to the East with a larger extension, and the non-tunnel style portals. That Eastern part has a reversing loop covering about 6x6 boards and cross above the portals on a bridge, that was theoretically a road to Washington D.C.--I looked up the world co-ordinates and the road would be West of the Big Six Mile B&O yard, and NW to NNW of DC, making that bay a inlet of the Potomac River. South out of the Mapleton yard I added several miles of climbing road, three portals, and a tighter reversing loop. That river section I let originate in a lower altitude marshland, so the 'port' was on the bay part of a slow backwater, not actually a big river, and that and the Maple Creek where the bridge crosses were both imaginary tributaries of the broader Potomac. Not authentic pro-typing to use Reversing loops, but they hid in the landscape and weren't obvious since I made them large enough to hold two long consists which I favor; but my AI Drivers were able to rumble back and forth creating dodge-em hazards for the unwary shuttle Drive to/fm the port, when sorting a consist and so forth. Added a station at the south end of that yard too, so passenger trains periodically complicated life inside the main yard. When I joined the maps creating a longer Tidewater Point Rail layout, bent folded, spindled and mutilated per Frank, I had a lot of operations railroading with mixed single and double road trackage, predominated by freight, but with the occasional prissy passenger consist demanding pull over waiting for passing expresses.

I did another large group of alterations on the North map, and when I joined them with that big gap patcher, the hardest part was getting the Maple Creek to look realistic. I'd have been better off using another two boards because the bends of the river were working against me. Much of the road added south of the Lumbermill I triple tracked for a long waiting siding, since coal trains servicing the Power Company needed to delay long enough for it to consume coal in their round trip repeat schedule. I also extended the Power Company road down to the bridge, introducing complications in the cross-over switching there with double cross-overs and whatnot, but the dang thing works pretty well. I had to rework that East end of the bridge several times to avoid collisions. Made me much smarter about signals placement and waypoint management.

Spiral Helix modules could be incorporated, not only, to raise trains to a higher elevation ... but to give extended time for a train that goes into a tunnel, that it does not emerge from another tunnel, for quite some time delay, later.

I used that trick several times in mountainous terrain -- it is of course an old model railroader's trick, but the height change is what I needed there, delays I manage with AI Driver waits, and triggered instructions to human AI's!

Count me in on this project, as I also predominantly use TRS2006, and iPortal.

After experiencing the improvement of surveyor tools in TRS2009 I'm kicking myself three times for not downloading it from Amazon/Auran sooner! Don't take this adversely, but I feel sorry for you stuck back there in the stone age. (I know, I stayed there voluntarily myself for 4-5 years!) Neither of my CPU's are spring chicken's either. 2GHz Pentium 4, and T'09 is running on them just fine.

The Laptop is just 1.87 GHz but a dual core something and has the better video card with the two math features missing on the towers G-card. Both are Nvidia G-cards with just 512 Mbytes, so not world beaters, the laptop is 7 years old and the tower over 10, so give it some consideration. If you can match those specs, time to upgrade my friend! See this for a good price on T'2010, and Amazon.com has TRS2009 for about $10.00 iirc. A steal at those prices. The Amazon site creates a infinite download 'My Software Library' private account keyed to your login ID and password, so don't even need boxed DVD sets anymore unless your internet is slow as molasses. In which case I'd surf for Amazon partners and haunt model trains stores/shows until I found it.

• My '04 and '06 content imported easily. In fact today I imported everything in '04 by clicking on the folder and most of the eight errors I got in the list of 687 warnings and errors were about 5 sessions I'd either orphaned, or hand edited and forgot to change both map kuids. Those all gave 'wrong kuid' errors, and the series was ...:1004 to ...:1008 -- so I deleted Map ...:1003 and also botched the hand edit... those were early daze, and those were modifications of tutorial 6 commodities, with extra rolling stock, locos and my first attempts at Editting Driver Sessions. Shortly after I moved to TRS2006 with CMP, so never caught on that they were orphaned. So much trouble! LOL
• The other there were missing textures - it looks like CM didn't know how to access them in TRS2004's hash file storage system. I didn't look at what those affect yet.
• And the 'real' cautions... just some obsolete tag name field glitches and a few other minor things was all, had more trouble with downloading Joggers 1-4 off the DLS-- that person wrote a license string tag named 'license', which apparently '04 and '06 tolerated and ignored.
• Predominantly, all the hundreds of separate files in the directory (even .dll, .exe, ... .ja, .gs, and .gsl) were 'This is not an asset warnings'
• But in the T'09 config.txt string parser, it barfed up 'the interpretation', unhelpfully actually adding quotes trying to patch the tag-string format and created successive 'false error' lines beginning with other false tags it'd created. Those were just words of his sentences of the license like "by", "the", and so forth... with each 'patch generating other bad tag names in the same paragraph.
• Fixes were simple, rejoin the false lines, eliminate the tabs and all the inserted quotes, delete the tag 'license' and put the rejoined license phrasing down where it belonged, added to the string already there.
• Most other errors I've seen are obsoleted tags like smoke or a missing (mis-found actually) sub-asset which could be moved up out of a sub-folder where it actually was.
• Had the most trouble updating several assets I'd cloned. The DLS had no update for them as I'd modified them to have loads for starting sessions. Those were all PRR rolling stock and two Gilpin (Northbay County) loco's I'd cloned to just have a preferred user name. (I prefix all my loco's with AaL## or AED## so FIND finds them first when I want to pop around without using the AI Drivers and especially in surveyor.

On the PRR, and Harrisburg environs project, come join and join the discussion. //Frank

 

[Fabartus]

Yes, just like that (I do still have TS2009, though, so I don't think I can use your sample). All corner modules will be 90 degrees and of four kinds (E to N, N to W, etc.) In fact, every module will be one of 6 kinds (add W to E and S to N to the above) unless the builder decides to do a 180 - 270 would be problematic since it limits what it can be attached to.

Incidentally, it takes 4 boards to go up just 16m at a 0.7% grade - very mild by logging railroad standards!

After pushing a Big Boy up the Marias Pass at some of 'those' grades (mostly 1.4% and greater. I think it topped out just short of 2.5 % for one short stretch) hauling 3900 tons, I say piffle to mere 0.7% grades anymore! That's essentially 40 miles of steady climbing on 56 miles traveled overall. I made it twice, and both the Helper and my Loco were critical on coal by the time we hit the Summit and could decouple. Fun though, you have to run both Steam Loco's yourself hopping back and forth ++, adj, --, adj, ++, adj, --, adjust and so forth. Never have time to get bored. Those steel beasts were a bear to operate. // Frank