When lightmaps are enabled, RSM model instances are lit in a similar color as the surrounding terrain. At first glance, the light contribution might just be a simple average, though this has yet to be verified. Using debug textures, that shouldn't be too difficult. Alternatively, it's possible the same lighting model as for the terrain needs to be used - IIRC shinryo mentioned trees were affected by some strange colors that sound like the contrast correction (based on ambient/directional light). Should be easy enough to test?
However, there also exists a configuration file called mapobjlighttable.txt that presumably controls the feature on a per-map basis. Somewhat unsurprisingly, it doesn't seem to be working (correctly)? I remember reading in some old forums thread that object lighting was bugged, but I can't remember the details... and I was unable to find the post again. Either way, the setting appears to have no effect. More testing would be needed in order to determine whether it actually does anything at all, and if so, what exactly that might be.
Doesn't really make any sense until #187 is implemented.
When lightmaps are enabled, RSM model instances are lit in a similar color as the surrounding terrain. At first glance, the light contribution might just be a simple average, though this has yet to be verified. Using debug textures, that shouldn't be too difficult. Alternatively, it's possible the same lighting model as for the terrain needs to be used - IIRC shinryo mentioned trees were affected by some strange colors that sound like the contrast correction (based on ambient/directional light). Should be easy enough to test?
However, there also exists a configuration file called
mapobjlighttable.txtthat presumably controls the feature on a per-map basis. Somewhat unsurprisingly, it doesn't seem to be working (correctly)? I remember reading in some old forums thread that object lighting was bugged, but I can't remember the details... and I was unable to find the post again. Either way, the setting appears to have no effect. More testing would be needed in order to determine whether it actually does anything at all, and if so, what exactly that might be.Doesn't really make any sense until #187 is implemented.