Small Question About Populations

[darthvader]

when can we expect tg2 to be able to create different sized instances in a population again? This isn't really a big deal and the images still look fantastic but in my opinion it is easier to have different sized trees in a simgle population than multiple populations.

[RArcher]

Unless I am really mistaken, you can set a range of sizes by modifying the minimum and maximum object scale values.

[darthvader]

hmmm, strange, tweaking that around never worked for me before now...i must be losing my mind

[Oshyan]

It's the "number of variations" setting that's gone. But that never worked anyway.

- Oshyan

[rcallicotte]

...and we want it to work, because...?

It's the "number of variations" setting that's gone. But that never worked anyway.

- Oshyan

[Oshyan]

Well, the underlying concept is that it would solve the "sameness" of instanced objects. However it is difficult if not impossible to do this with imported objects since you can't programatically determine an appropriate way to create variations in a realistic and consistent way. Simply distorting the geometry in some arbitrary way is going to look very bad. I believe the function was originally implemented to handle populations of *internal* objects, such as rocks, grass, etc. in which case, being procedural and generated by TG2, the creation of variations would not have been so difficult. However it is not a system that we have been able to develop properly yet, so it was removed for now. We realize of course that solving the variation issue is very important to the realism of populations so it's something we'd like to devote time to in the future.

- Oshyan

[rcallicotte]

As long as I can turn a tree randomly in a population per tree to keep this randomized, I'd be happy with this for now.  Do we get this already?

[cyphyr]

As long as I can turn a tree randomly in a population per tree to keep this randomized, I'd be happy with this for now.  Do we get this already?

Sure do, its the default setting, you might also want to set the min and max scale to something like 0.8 and 1.2 (or thereabouts, you decide the limits). This is about as random as you can get in a single population for the time being. A year or two ago someone was working on a way of randomizing the colour over populations, dont know how far that developed but it would cirtainlt add to the randomness overall.

Richard

[rcallicotte]

Yes, I mess with the min and max, usually 1 min and 2 max or 3 max.  Depends on the object...usually it's trees or bushes.