Getting started with a new track, few questions.

[Sam Moss]

After splining a layout using Smooth type of line for the initial nodes, then I converted them to Bezier and then used breaks in the segments with more bezier nodes. I then created a plane 4000m long by 12m and altered the segments so I have 4 x 3m width planes across the track and 5m length planes. Next i used Path Deform Modifier and used the created spline as the path but it doesn't really follow it.



Also is this a correct way of lofting a track? It seems a better way to get the track split up into specific lengths compared to the loft tool which doesn't seem to space things very evenly. Will I still be able to edit elevation on the spline afterwards?

[Alex Sawczuk]

Use the normalize spline modifier on your path, and then loft ontop of that spline.

As long as you don't collapse the stack you can edit the path elevations

[Sam Moss]

Thanks Alex, got a bit further now

[Sam Moss]

Another quick question. I made my spline all good and smooth but when I normalize it after lofting it goes into an odd shape on tighter corners

Spline


Normalised Spline

[Alex Sawczuk]

looks like your either normalizing at 20m or so where there's going to be issues, or you spline isn't actually smooth and is just being hidden by the low number of steps in the spline rendering roll out.

[Sam Moss]

Thanks Alex I put normal spline to 1 so it does it in 1m steps then altered my loft to Path Steps 0. Before I had path steps 10 and Normalise Spline seg lengths to 16m. Looks much better now. Is 6300 polys ok for a 4.3km track? I set width to have 3 steps as with real road needing to use more polygons for the live track I presumed it would give better effects?

[Luc Van Camp]

How wide is the track? Ideally, you'll want the polys to be between 1-2m wide. Typically, the ISI tracks use about 1.5m wide polys.

[Sam Moss]

The track is 10 metres wide so 4 polys would make it 2.5 metres wide per poly. If i used 6 wide that would be 1.5 metres and increase from 6300 to 9300 before adding extras in to smooth out the hairpins.

Thanks for all the help guys.

[Luc Van Camp]

Personally, I would go for 6 polys wide in your case. The difference between 2.5 and 1.666 can be quite substantial depending on the roughness of the surface.

[feels3]

I have made very simple oval circuit, just for testing and I have problem with materials.

I have two roads with the same material ID.
Like you see one is very dark (in the midle) and second one is very bright (on the left side), both with the same material.
Both should be bright.

[Luc Van Camp]

Hmm, not sure I understand the problem correctly. You say you have a problem with materials, yet they use the same material, right? In that case ...

• Could it be that one object is RaceSurface, and the other not, making one of them react to RealRoad?
• Are you using vertex colouring and/or alpha on either of the pieces?
• Do they both have correct mapping coordinates for all channels used by the material?

[feels3]

Both objects are RaceSurface.
There is no vertex colouring.
Mapping looks ok.

[Luc Van Camp]

• And these are both using the exact same material? Not a clone with different textures?
• What happens when you attach the second object to the first temporarily?

[feels3]

Material is exactly the same for both objects (ID1)

When I have attached both objects rfactor crashes on startup (track doesn't show up)

[Luc Van Camp]

OK, that's an issue I have never seen before. It isn't a material issue since it works fine on one object. As a test, remove the correct object from the SCN and see how it affects the dark object. Then put it back in the SCN and remove the other one, just to figure out if the objects aren't interfering with each other. also, what happens if you export both objects with a simple T1 shader? The good old methodical approach .
Perhaps another long shot which probably won't fix things: try exporting both objects as 3DS files (File, Export), create a new scene and import the 3DS files. Reassign the material and export.

I'll ask Scott to take a look at this thread.

[Scott Juliano]

First, when you attached them did you remember to remove the .scn reference for the no longer existing track piece? Just checking...

Otherwise, that's really odd--never seen that before except when I hadn't mapped the detail stages properly (map channel 2). But it LOOKS ok in your image there. You might want to run an STL Check modifier on the object and see if there are any warnings (you'll get open edges warning obviously), but it could point out some other faults in the geo. Baring that, try a Reset Xform.

Beyond THAT, make a backup of your file, try exporting the object as a .3ds (keep uv coordinates), move the old object out of the way, import the .3ds (do not convert), weld all the verts and reapply your material. Export THAT to gmt and see if it's the same. If it is, then I would actually have to take a look at the object in question to see if I can find something...

[Jka]

Don't wanna hijack the thread, but...

Actually I'm struggling kinda similar problem, or probably I haven't learn the new lightning system influence to material/texture appearance in sim yet.
weirdness.jpg

Upper left corner on that pic shows huge difference between main racing surface (darker asphalt) and escape road (lighter asphalt).

Upper right is max viewport more or less same spot.

Lower left is sample of main road texture lightness/hue/saturation.

Lower right is sample of escape road texture.

Main road is setup as RealRoad material and escape road is setup good'n'old BumpSpecularMapT1. Both materials contains same specular color and -power.

As you can see, difference in mid-day lightning is huge. One solution might be to get rid of escape road material specularity (map and color) and adjust texture lighning to match RealdRoad. Downside is that then I lose sunset and dawn lighning effects (and I don't want to do that). Other solution might be to set escape roads as RealRoad, but it's waiste of resources (and this track will be on the limit anyway). So I dont want to do that either.

Do I need to dump down escape road texture's lightness a lot (and lose texture's contrast as well) to match lightning compared to RealRoad material? Is there any better way?

[Luc Van Camp]

My suggestion would be to use the RealRoad shader, but export it as a different object, and don't use the RaceSurface_ prefix and don't tag the object as Deform. That would disable the RealRoad interaction (and calculations for these vertices). Perhaps we need a separate shader that does exactly the same as the RealRoad shader, but without the RealRoad texture stages.

[ethone]

If you add new shaders, could you consider a RealRoad shader without marbles/groove but with the wet weather reflection?
That would be immensely useful for tarmac runoffs like I have them at Fuji where I want the reflections and could do well without the calculations on the vertices.

[Luc Van Camp]

Yep, I wouldn't mind having a shader like that myself. It's on my personal wishlist .