Headlights in ORTS

[ericmp]

One issue present in ORTS is the incorrect positioning of headlights, whether they be too far up or down from their intended position. While tinkering around, I found something rather interesting: It seems the positioning of headlights in ORTS is related to the radius of the headlight specified in the locomotive's .ENG file. I'm not sure if the ORTS team is aware of this, but I thought I'd point it out. Here are some images demonstrating this possible relationship.

Here we see this locomotive with it's headlight set to dim in ORTS. It is evident that the headlight is positioned too far below the intended position.




Here we see the locomotive's headlight set to bright. The bright headlight, with a larger radius, appears to have changed in position. It appears to have moved farther down than before. This could possibly be because the radius of the bright headlight is larger than the dim headlight.




Here is an excerpt for the .ENG file, with the headlight radii highlighted in red.




Here now is the the .ENG file with new headlight values, highlighted in green. I've tried many values, and it seems that ORTS seems to like a radius of 0.5; any lower, and the headlight is actually positioned higher than the intended position.




Here now is a picture of the locomotive in ORTS, with the headlight on, and it appears that the headlight is positioned correctly.




Here again is the locomotive with corrected class lights. Their radii values are set to 0.5 as well.



Again, I'm not sure if the ORTS team is aware of this, but I found it interesting and I thought I'd point it out.

[twpol]

I don't think I've seen anyone link the radius to the vertical alignment before, but I sure do find that interesting. I'm adding this to my team's issue list for a proper investigation; thanks!

[mestevet]

I wonder if it's something like referencing the upper right corner of the light instead of the center...

Steve

[twpol]

It's not fixing a corner... notice in the pictures only the vertical alignment is off, where as horizontal appears stable... and having consulted the OR code, it is clear that it's positioning the top-center at the location loaded from the file.

Now, light positioning has always been a bit of a black art (there's this lovely 0.275m vertical adjustment in the code too) so there may have been a reason for doing it like this... but it certainly seems to be the wrong thing to do when different states have different radii.