Default Track Replacements

[eric]

OK, the values *should* be Position (-2.6 -0.02) and Position (2.6 -0.02). If you see Position ( -2.5 0.0 ) or Position ( 2.5 0.0 ), change it and see if that fixes it visually.

The TrProfile for the USTracks textures is intentionally deeper and wider than that for ACleanTrack. That makes the TrProfile.stf's different as well.

[RobGuru]

Will you attach the fixed trackprofile?

[eric]

No need -- those values are the ones in the May upload.

[pschlik]

OK, the values *should* be Position (-2.6 -0.02) and Position (2.6 -0.02). If you see Position ( -2.5 0.0 ) or Position ( 2.5 0.0 ), change it and see if that fixes it visually.

The TrProfile for the USTracks textures is intentionally deeper and wider than that for ACleanTrack. That makes the TrProfile.stf's different as well.

The 2000 meter LOD had the 2.5 meter position but other LODs used 2.6 meters already. Changing the 2000 meter LOD obviously didn't do much good for staring at track up close. (I already replaced all the track profiles with the one included in the update.)

[paulytechnic]

A bit late, but I updated the UStracks version of default track replacements I was using and the track profile on dynamic track is not matching the shape of 'actual' track sections. (The track closer to the camera is regular track, the further track is dynamic)



I also notice some flickering, though I'm not sure if it's plain old z-fighting, at a distance on the dynamic track sections which makes them clearly stand out from regular track sections.

The Atracks set doesn't have those problems, so I'll be sticking to that for the moment despite the hit or miss nature of the original track textures.

I too am using the latest UStracks version and have noticed the dynamic track profile is different (specifically the ballast slope), even with the latest track profile. I fiddled with the UStracks profile and think I've come up with something that matches the ballast "better" than before (attached), but I'm not sure it'll ever be 100% perfect due to the variations of the different pieces it connects to. The main issue seemed that the ballast slope wasn't sunk quite deep enough. I also resolved the flickering issue with distant dynamic track by changing the "Dark Ties" ShaderName to BlendTexDiff instead of BlendTexADiff (this was an issue I encountered with the standard version of UStracks as well).

LODItem (
Name ( "Dark_Ties" )
TexName ( "US_Track3.ace" )
ShaderName ( "BlendTexDiff" )
LightModelName ( "DarkShade" )



Also, the replacement shapes seem to be missing the spike heads from the original UStracks design, while the dynamic track still displays them...

[RobGuru]

How to convert US3 to Dynamic?

[eric]

Also, the replacement shapes seem to be missing the spike heads from the original UStracks design, while the dynamic track still displays them...

I'll take a look. There's another layer for the spikes that should be there.

[oper190]

Just checked my install of updated tracks, USTrack vers. and found the same.



Normally don't get this close to the rails to realize no spikes.

[qballbandit]

Hi, may sound like a pretty ridiculous question to some of you - but, what utility is best for opening and editing the .stf (trackprofile) files,please?
Very much appreciated.

[paulytechnic]

I usually use Notepad since it's readily available & simple, but most any text editor will do.

[qballbandit]

Thanks Paul, easy enough!

[eric]

replacement shapes seem to be missing the spike heads

Found 'em. There's an additional texture and layer that wasn't included in the original profile I got from Doug.

I'll have a fix out sometime in the next week or so.

[paulytechnic]

Found 'em. There's an additional texture and layer that wasn't included in the original profile I got from Doug.

I'll have a fix out sometime in the next week or so.

Wow, that's great! Thank you, Eric! 😎

I just noticed something else this evening, which isn't really that big of a deal, but I thought I'd share my observation anyway. When compared to Dynamic or regular US3 tracks, the UStracks Replacement shapes seem to have a "harder" shadow on the ballast slopes. It's most noticeable in the dawn/dusk hours when shadows are at their most extreme, in both OR and Shape Viewer. You can see the difference in the screenshots here:





It's been quite a while since I've done any kind of modeling, but I believe it's "poly smoothing"(?) that normally would reduce the crisp, defined edge. I'm not sure if it's possible for the script generating these shapes can do any kind of selective poly smoothing or not. If so, it might help the look since ballast/gravel rarely has such a sharp edge. However, if it's not possible, I completely understand, and it really doesn't look that bad (just different).

Another option might be to update "something" in the TrProfile.stf to disable the smoothing, so the dynamic track can have a consistent shadow as everything else, but I'm afraid this is a bit beyond my manual-editing expertise.

Just thought I'd throw this out there! 😉

[eric]

Yeah, I've noticed that, and have tried a couple different things but haven't landed on a good fix. The shadows are legit, and I suspect that the reason they're not appearing on dynamic track is that it may have been designed in a way it doesn't cast a shadow.

The problem is exacerbated by the USTracks textures and how alpha's are defined. The edges are alpha'd as you'd expect, but the ballast around the ties is also alpha'd out, which means it has to be drawn as three separate objects (slope, top, slope).

As three separate objects, there's no edge to un-crease or apply smoothing to. It might be possible to make it more of an arched curve, but that's going to increase the poly count for each shape.

[NW 2156]

Eric,

Is there a readily available Dynatrax profile?

I've got a couple routes that are heavy with dynamic track that I'd like to convert fully to the new fine rail.

Thanks.

Robert