FT3+ How to get realistic worlds

I apologize for the complex questions. I've read the user guide, and tome of ultimate mapping, and I still cannot get Fractal Terrains to make anything remotely realistic or Earth-like. As a newb I assume human error and my own ignorance are in play. So before I sink a ton of time manually tweaking a map that has a close-enough coastline but totally weird features, I wanted to see if there is something I am doing wrong that can adjust.

What I mean by "realistic" is a world where continents have mountains down the sides not the middle (Rockies/Andes on the West coasts of the Americas, the Alps in the south of Europe, the Himalayas on the North of India, etc.), and where there are large flat expanses alongside the ranges.

Every world I generate, using every combination of settings I can think of, places mountains in the dead center of each landmass or island, and only gives teeny-tiny areas of flatlands. Nothing like the central European plains much less the vast North American plains or the African Savanna; I can't even get something the size of the Gobi desert.
So what are the recommended settings? Or is this something that has to be manually tweaked each time?

Secondarily, are there any settings that will generate trenches, abyssal plains, and large continental shelves around island chains? On Earth's oceans there are the shallow areas (littoral) and the benthic (bathyal, abyssal, hadal) regions with hadal being the really deep trenches, abyssal often being immense flat plains, and bathyal being the slopes from the edge of the continental shelves down to the abyssal plains. FT doesn't seem to model oceans with these features. If I understand it, the software just does altitudes then fills the globe with water and some land ends up above and some below. So I'm guessing that the only possible way to get oceans with these features is manually? Or are there some settings that can do it?

Last, the random climate models don't seem to incorporate rain shadows. So all my mountain ranges have jungle on both sides whereas realistically one side would be dry and the other wet. Did I miss something in the settings or is this another thing that has to be added manually?

Thanks in advance for any help!

Comments

  • jslaytonjslayton Moderator, ProFantasy Mapmaker
    I like to recommend the tutorial at www.fracterra.com/CGTutorial/index.html for general FT3 usage. It's a little old, but the basics are still useful, I think.

    What you mention about the automatically-generated parts of an FT world are, unfortunately, all true.

    The altitude model is a scale fractal evaluated on a sphere with a few minor tweaks like an exponential operator for continental shelves. This model won't generate tectonic features like mountains at the edge of continents, large continental plains, deep ocean trenches, or mid-ocean ridges, all of which arise from tectonic and erosional processes. Simulating tectonic results will require manual editing (there are some minor erosion simulation tools in FT such as basin fill and incise flow, but using them to get good results requires practice).

    FT's climate model is extremely simple: it doesn't incorporate heat or moisture transport via wind and water, so features like rain shadows won't be present without manual editing.
  • Your comments about mountains in the center, etc are all problems with a fractal terrain generator.

    I have recent been thinking about using tectonics to generate the map, for example see or Lauri Viitanen thesis (located here: https://www.theseus.fi/bitstream/handle/10024/40422/Viitanen_Lauri_2012_03_30.pdf?sequence=1&isAllowed=y). Of course both of these methods would need to be improve considerably, especially for spherical worlds. Then we would need some way to import the data in CC. As I'm finishing up a Masters in Applied and Computational Mathematics I've not had time to fully investigate these ideas.

    Also, if people can simulate weather in Middle Earth (http://www.bristol.ac.uk/news/2013/10013.html), I would think we should be able to do something similar given that we would have accurate elevation data. Might be able to push some of the calculation off to the video card during the run? Then again, if the simulation takes 30 mins or so (or longer) it might be OK if done off-line.

    Also, http://procworld.blogspot.com/, has some blog ideas about water and city locations.
  • Thanks for the feedback and suggestions to check out the older tutorial and procworld blog.

    Looks like I've got a bit of work ahead of me. But it's still a lot better than doing everything by hand.
Sign In or Register to comment.