Hi Dick,perhaps I understand it finally little bit, so what we see on the generic ground in one LOD13 quadrant isn't one texture, but the blend of the textures from the corners, aAHHHH beautiful! Thanks for your patienceKind Regards

Hi Christian.If you edge the LOD5 area with #254 ( transparent ) assignments, then the landclass blends into the underlying ( default ) landclass at the edges. That gets rid of the sharp line transition.I think the actual runtime display is 256x256 groundtiles, defined by 257x257 verticies. Dick

Hi Dick,this is why I made the above described test. FS2002 only displays 255 cells fully. The border cells only get displayed half (which of course adds up to one cell). This is important, because the borders of two LOD quads will have to match. I actually checked that now, and indeed if you split a larger area with resample into single LOD quadrants, the borders do match...Cheers, Christian

Hi Christian.If it were true the edges of the LOD5 LC areas only displayed 1/2 texture ( and then the corners would display only 1/4 texture ), then why can I place a custom texture in it's entirety at the corners?The custom textures tell me it's 256x256 cells defined by 257x257 verticies. There's no guessing as the names locate the cells absolutely.It may be hard to see, but the red crosshair is exactly at the corner vertex in each screenshot.Dick

It's because you used a non-blending texture, silly :)As I said above, it won't work with non blending textures like water...Try a texture with a blending mask (eg forest) and you'll see what I'm talking about. Cells 1 and 257 will both only be displayed half.What happens is this: cells 1 and 257 share the border texture and will blend. Of course if one texture has no blending mask, it won't blend and take control. Incidently, cell 1 and 257 have to have the same value, otherwise you get an ugly border. I really should have taken a screenshot, but it's easy to replicate...I'm only referring to landclass here. I don't know what consequences this has for custom textures, since they don't blend. It may be the case if both cell 1 and 257 are nonblending, cell 1 wins, and that's why 257 doesn't show.This got me a bit thinking also. If we have 256 cells placed on top of the vertices, then the elevation values on the borders also have to match! This could be tested, though...Cheers, Christian