Work flow for a large scene

I was interested in making a scene that showed fly threw that ended on a castle in the mountains surrounded by a lush european style forest.

Something very much like this but with steeper cliffs and some rock being exposed and less windows for the castle.
random pic

The castle should have stone texture smiler to this pic.
another random pic

Now that I have an idea of what it is that I want to make, what kind of processes/work flow would you recommend for making the forest and castle for this shot? I can see there being a lot of trees and I’d like to have a bit of detail in the castle but at the same point I don’t want to spend the next several months just working on this one shot. So what do you guys recommend?

I think you can make details only on parts that camera will show.
Motion blur or something like that may be usefull as well.

Depending how much you want to show in close ups, this will take a lot of work too.

I don’t know if blender supports level of detail, but that would be usefull as well. (Models’ details changes with distance, it’s used in games i think).

For the fly over, you can use a fairly simple model, with UV mapped texture images. The landscape would be a plane, subdivided somewhat, with a couple of cloud bump maps to simulate the tops of the trees, with the diffuse color from an image of tree tops, such as your example.

Likewise, the castle would be simple geometric shapes with the wall details “painted” on using UV mapped texture images.

Getting from the sky to the landing site is the trick. If you want to show that, you’ll have to model everything. Motion blur can allow you to use simpler models, but it’s still a lot of work.

You might consider doing a cut from the overhead view to a ground based camera on the landing site looking up at the flying thingy, and watch it come down to the point where you’ve avoided all the tiny stuff growing large animation, and then cut back to the flying thingy point of view with a close up of the landing site. If the camera on the ground doesn’t move, you can fake a lot of the scenery with texture images, although you’ll still need to model the things in view once you switch back to the flying thingy point of view.

Work flow:

Make a rough model of the scene.
Make a clay animation (no materials, everything blender grey, basic lighting to show shadows and shapes) to work out camera angles and motion, and timing.
From the clay animation, decide what you must model, and what you can texture map.
Do the detailed modeling, and collect or make texture images.
Test your modeling and texturing by doing still renders at various points in the animation.
Fix and tweak.
Render the animation.
Do whatever post production you want (sound, compositied special effects, etc.)

This is a very subjective question, but what do you think the minimum and maximum amount of time would be needed to make such a shot if the person working on it knew what they were doing? In regards to just modeling, texturing the environment