[mopacfan]

Hi,

Will TSRE5 generate procedural track in the ScaleRail or USTrack profiles? Could the procedural track option give TSRER users the ability to convert most if not all of the track in a route built with XTracks to USTracks or ScaleRail? I have a route project that has many customized track pieces in the XTracks profile. There are no ScaleRail or USTracks equivalents to these pieces. I would like to give the width of the track in the route a more prototypical appearance as seen with the ScaleRail or USTrack railhead profiles.

[Goku]

 Genma Saotome, on 17 June 2017 - 07:26 PM, said:

Goku, rail can probably be done very well with a UV scaling of 1m. Unfortunately ballast and roadbed are highly unlikely to come out as well. 1m of something as visually obvious as ballast will very likely descend into obvious tiling patterns

There is no 1m tiling patterns in my solution becauase UV coords are expanded by actual track length.

 Genma Saotome, on 17 June 2017 - 07:26 PM, said:

Further, things like what's between multiple tracks might be another issue -- is ballasted or not, if it is ballasted is it flat or is there a depression? That too must be addressed.

That is why procedural shapes, not .s fixed shapes, are so important.

 Genma Saotome, on 17 June 2017 - 07:26 PM, said:

What's needed is for the profile data to define all of the UV scaling. In the ideal the template should actually be a finished .s shape.

It is finished .obj shape. I don't see reason why use .S file format instead.

[Genma Saotome]

 Goku, on 25 June 2017 - 05:43 AM, said:

There is no 1m tiling patterns in my solution becauase UV coords are expanded by actual track length.

Shows you have no idea what you are doing. This is the worst possible answer. 500m stretched ballast texture in one case, the same ballast texture stretched 3m in another? Pluuueeeze.

 Goku, on 25 June 2017 - 05:43 AM, said:

It is finished .obj shape. I don't see reason why use .S file format instead.

And for those of us who do not have or use a shape editor that produces .obj files?

[Goku]

 Genma Saotome, on 25 June 2017 - 08:24 AM, said:

Shows you have no idea what you are doing. This is the worst possible answer. 500m stretched ballast texture in one case, the same ballast texture stretched 3m in another? Pluuueeeze.

Shows you have too small imagination to understand my ideas.

One meter 3d shape:

http://i.imgur.com/xWrIMH3.png

And extended UV in TSRE (it still needs some work):

http://i.imgur.com/ZZNeNh4.png

 Genma Saotome, on 25 June 2017 - 08:24 AM, said:

And for those of us who do not have or use a shape editor that produces .obj files?

Obj file format is the most common 3d file format in the world. And most important it is very easy and fast to make simple shape. S file is diffucult even for simple shapes. Using S files as templates would vastly increase shape generation time.

[Genma Saotome]

Ok, I see your comments were very poorly phrased. You are not stretching the texture you are doing a step and repeat. Which brings me back to the original complaint about tiling. 1m is fine for rails but is too short a distance for ballast and subroadbed.

IMO what you want to do is let the modeler define the template model to whatever size he wants it to be and so a step and repeat of that which may include a truncation to fit the specification of the track segment being set down. That way the modeler will have complete control over the appearances (plural, as there will be many in each route) and the programmer is left to do what programmers do best: figure out how to implement it.

[Stephen Hjellum]

Question about stretching: will it be simple click and drag, or something else? Are you aiming for something similar to what RSC/DTG has done with RW/TSW? If so, I'll be very grateful (more so than I am now).

[Goku]

 Genma Saotome, on 25 June 2017 - 01:02 PM, said:

Ok, I see your comments were very poorly phrased. You are not stretching the texture you are doing a step and repeat. Which brings me back to the original complaint about tiling. 1m is fine for rails but is too short a distance for ballast and subroadbed.

You have a 12 m texture above. Look at the numbers. It is not simple 1m texture repeat.

 Genma Saotome, on 25 June 2017 - 01:02 PM, said:

IMO what you want to do is let the modeler define the template model to whatever size he wants it to be and so a step and repeat of that which may include a truncation to fit the specification of the track segment being set down.

UV coordinates are linear, so there is no difference in final result if you make template 1m or 5m. But it is nice to have one standard length.
Also, 1m shape is used only for stretchable templates. For "repeat" templates, you can make any 3d object.

[vince]

Hi Goku,

I'm all for any new track as long as I have a way to convert existing track in my route.

I'll take one section at a time at first working up to convert existing track one tile at a time.

Yes, I did push for OR back compatible and I feel the same way about a new way of railing a route.

regards,
vince

[Jovet]

 Genma Saotome, on 25 June 2017 - 01:02 PM, said:

Ok, I see your comments were very poorly phrased. You are not stretching the texture you are doing a step and repeat. Which brings me back to the original complaint about tiling. 1m is fine for rails but is too short a distance for ballast and subroadbed.

I believe what he is saying is that the mapping V coordinates are scaled to the length of the shape. The template .obj is read in and, say, its 1m length is seen to have coordinates of V₁=0 and V₂=0.25. To create a 25m shape, for example, V₁ stays V₁=0 and V₂ is scaled to V₂=6.25. My only problem with this is it seems inevitable that the scaled coordinates are going to commonly end in the middle of a tie/sleeper. It looks odd when dynamic track does that, and this would be no better.

[Goku]

 Jovet, on 26 June 2017 - 04:36 AM, said:

My only problem with this is it seems inevitable that the scaled coordinates are going to commonly end in the middle of a tie/sleeper. It looks odd when dynamic track does that, and this would be no better.

That is why I think it's better to use 3d shape for a tie, not texture.