I am the developer of this solution. Of course - this is not for everybody, and compromizes needed to be made. The low-res on the ground is simply because everything is ZL16, and while my engine is capable of doing any zoom level, I am only one person, with one machine, and higher zoom levels would also mean curating higher-res pools for the textures - plus, most importantly, going from ZL16 to ZL17 alone would quadruple render times and the amount of files. Then to ZL18, yet quadrupled again... and so on. The goal was, and is, to see how far it is possible to eliminate issues with satellite imagery... but this solution will never be able to reproduce satellite imagery to 100%. Not now, and not in ten years I am afraid... but I still believe the tech is in a presentable state. I am rendering subtiles of a lat-long tile on ZL16, and that is then compiled into the final mesh. As just stated - ZL16 is a compromize I needed to make, as I do not have endless resources, time, and machines (yet... ?). Thanks for pointing that out. I do not use any AI for my posts, the code, and the whitepaper is also all by my hand. Thank you all for checking it out and discussing this... approach. I know it is not perfect, and I do not know how perfect it is ever going to be. But you can be assured I am working on new ways to improve the tech when I have the time. Clear skies!