Footprint decal stretch regression

[GoogleFrog]

/give striderbantha

The decal only looks different for units with trackStretch != 1. I suspect the sign of an exponent was flipped somewhere in the new decal system.

2457

2031

[GoogleFrog]

Well, more of an unintended(?) behaviour change than a regression. But I'm not going to account for it until I know it is intentional, because it impacts a lot of places that are hard to reach (ie modding and other games).

[sprunk]

I think it's a good change because now a 0.5 stretch factor (like bantha has) makes each segment 0.5 the size. Smaller factor -> smaller segments.

[lhog]

Yeah I may have made it reciprocal accidentally. Since you're first to notice tell me if you want it the old way or is the new way fine?

[sprunk]

I prefer the new way because it is more intuitive and a gameside fix is trivial ZeroK-RTS/Zero-K@00c47d4

[GoogleFrog]

I have no preference for ZK, but it would nice to have one fewer thing to fix when updating an old game to work on modern spring.

[lhog]

Well seems like you have posted the fix to ZK already with the parameter read as is, so I assume we will move forward with that new semantics.

[sprunk]

There are other games than ZK though, but I think I can just PR the trivial fix for them approximately this weekend

[sprunk]

FluidPlay/TAP#11
techannihilation/TA#1678
MF and SF seem to have nothing to fix

Others TBD

[sprunk]

Looks like that's all.