I guess that those of you who have given me improvement suggestions like "check out 'smart trees'" etc., have noticed that I have pretty much ignored them all (expect for Aleks' "photoshop a real sky in those shots" and Dac's "better names" suggestions as these are done in post-production). I hope you guys don't think, that I won't follow those suggestions because I thought that Rocane is already perfect or something like that.
Quite the opposite, actually: I think Rocane is not worth the work improving it would take. There would be so much to fix (Rocane being all finished and all - all the way to 2020) and the result still might not be worth the work.
I will rather start modeling a whole new city - which I've already actually done (it'll be a while before there's anything to show you guys). Rocane won't be my Staunton. But thanks for those suggestions anyway and keep them coming. They will help me make my next city better.
I've checked out those smart trees. I didn't actually try them in action so to speak - that is, I didn't try to replace Rocane's trees with smart trees. I did some math and I came to a conclusion that it would explode my computer. There's a total of 537 tree entities in the final Rocane model. Those trees have 2 faces and approx. 13 edges. That adds up to 1074 faces and 6981 edges for all the trees. The smallest smart tree has 850 faces and 7110 edges, so those would add up to 456 450 faces and 3 818 070 edges (the final Rocane model has a total of 76 342 edges and 26 072 faces with nested components included!). I know that it's not that straightforward, since they're all components, but every component instance slows the model down a bit, and it was already too slow for walkabouts for example (which was one my goals: to make a model so light I could walk in it).