World file format

[jonpd]

Is there any documentation anywhere about the format of the world file, especially track section entries?

I am trying to track down a problem where some underground track laid many years ago using MSTS RE has ridiculously large y values showing up when I place the pointer over the track, like -800 meters, where the actual elevation is around -9 meters. The track shows up fine in the TDB (has the correct coordinates) and the yellow lines etc. look correct under TSRE, so I assume it is a world file issue. Most of the problems are on 1 or 2 tiles.

I am trying to figure out where TSRE is getting the large negative y values from. As I mentioned there are no problems running over the track, but I cannot place interactives on it (signals, speed posts), they just don't appear.

Jon

[jonpd]

This is an example of one section, a A2t250r5d which shows as a position of around -3 meters but moving the pointer over it shows y values around -860 (varies a little as you move around). I cannot place an interactive on it.

TrackObj (
UiD ( 69 )
SectionIdx ( 112 )
Elevation ( 0.0261799 )
CollideFlags ( 527 )
FileName ( A2t250r5d.s )
StaticFlags ( 00200180 )
Position ( -590.224 -3.18136 237.881 )
QDirection ( 0.00333531 -0.96691 -0.0126575 0.254782 )
VDbId ( 4294967294 )
StaticDetailLevel ( 0 )
)

The position looks valid but I wonder if there could be an issue with the QDirection? I don't know what the format is for those.
I wonder where TSRE gets the pointer vector from that shows up in the TSRE window. I guess only Goku could answer that.

Jon

[eolesen]

I've suspect a problem with the QDirection calc in TSRE for a couple weeks. If I manually calculate the QD in Tangent, I come up with different values, and some swapping of negative/positive values.

Have you toggled the terrain view? (alt-V, H) I've found I can't select or work with anything underground as long as the terrain shape is visible.

[Vince]

I experienced this on the PRRv2 rebuild. It was always a rogue tracksection which sometimes appeared untextured. They showed up mostly underground but some were at at a high altitude.

++In TSREv699 run the error checking routine, DO NOT allow auto fix at first! Go looking at the errors called out via the [JUMP] button.
++Another way I used on PRR (per v699) was to select the track and then dive underground looking for the bright blue 'selected' outlines.

regards,

[eolesen]

I also found a bunch of track underground and floating... Showed up plain as day in TrackViewer but couldn't find it in TSRE without diving hundreds of meters in either direction.

[Vince]

I never thought of using TrackViewer! Thanks Eric. I did use it to locate rogue signals tho'.
The new Error function (v699) makes it easy if object is selected. Blue outline is visible a long way off (>~2km)

regards,

[eolesen]

Yeah, that's how I first noticed the bug. Was taking a screen grab and found some whiskers where they didn't belong... went to the world coordinates and didn't see any blue poles and got curious.....

[rdayt]

The TSRE Errors and Messages window under the tools tab will show you a list of misplaced items and give you the option to jump to their location.
This will allow you to find these floating pieces and delete them. You may have to look around to locate them but they will be right where you jump to.
I've used it a couple times, seems to work pretty good.

[jonpd]

I don't think this is my problem as if it was, the submerged track would have to exactly parallel my visible track for a considerable distance and you would think it would show up in the TDB and/or world file with the bad y coordinate.

I am still hoping to find information about the world file format, especially the QDirection. Since I see posts where people play with the QDirection it must be known by someone.

Jon

[eolesen]

Well, QD is a quaternion value and I wish it were easy to explain. It defines the 3D vector that the shape takes starting at the XYZ position value. The math to explain the values isn't simple by any means... that's why there are calculators to derive it from two sets of known XYZ coordinates.

0 0 0 1 is due North. That's the only value that is easy to explain....

I'll explain the rest later...