[WIP] Castenet Regional Map - My First Map on CG

[Redrobes]

I'm not sure I understand your method. Would you care to provide an exemple, I'd be interested!

Well it is going to be a little hard to describe. But if we work in greyscale bitmaps then to make a seamless texture you can use the tutorial thing I wrote a while back:
https://www.cartographersguild.com/s...ead.php?t=1373

Now if you do that only with 16 bit height map values instead of bitmaps then you can get the same result.

So I download a load of real height maps and place them together as best as possible. Then blend them into one large one where there is no obviously large discontinuities at the edges so that you could butt them together and it wont look all that bad. You can fade out at the edges and have a different tile if you like is another way to do it.

So that tile makes your high frequency rock texture in height map. You then manually use an airbrush to generate a greyscale bitmap of where you want your mountains and set up the heights manually but very roughly. If you want a long mountain range you can spray on a line of white and fade it all down so its a blurry greyscale height map of the landscape but with no texture to it.

Then with a suitable height map tool (and thats the issue usually) you multiply them together and then where you have high blobby bits you get high rocky textured zones and where it was low (or dark) blobby areas you don't get very much of the texture coming through. When you look at the result you will have mountains that match the form of where you placed them. The detail of the mountains look real because it came from real height map relief from earth data such as Shuttle SRTM or similar.

[waldronate]

The detail of the mountains look real because it came from real height map relief from earth data such as Shuttle SRTM or similar.

And that's the major weakness of collaging for terrains: terrains are defined by their drainages rather than their landforms. That is to say that landforms emerge from drainage patterns controlled by underlying geology and the landforms are joined by those drainages. If you don't get your rivers lined up plausibly in the collage, then the results are pretty unconvincing for anything other than very quick glances.
Montcalm's map looks very nice at first glance and individual subunits are all plausible. It's an excellent example of how careful selection and manual blending can overcome a lot of the problems that often turn up when people do this sort of collage. However, I personally find the landforms a little bit unsettling because the individual parts just don't seem to flow well from one to another. For example, there are a number of glaciated mountain forms and then there's that big block on the right that isn't nearly as worn. I also wonder if maybe there was a different scale used for some of the results? The top-left mountains just feel out of scale with the rest of the map (but my eyesight has been going in the last few years, so it might be me). I also wonder if the auto-exposure feature of the Tangram Heightmaps browser might be contributing a little bit because auto-exposure will tend to make things appear more rugged than they otherwise might.

[Naima]

And that's the major weakness of collaging for terrains: terrains are defined by their drainages rather than their landforms. That is to say that landforms emerge from drainage patterns controlled by underlying geology and the landforms are joined by those drainages. If you don't get your rivers lined up plausibly in the collage, then the results are pretty unconvincing for anything other than very quick glances.
Montcalm's map looks very nice at first glance and individual subunits are all plausible. It's an excellent example of how careful selection and manual blending can overcome a lot of the problems that often turn up when people do this sort of collage. However, I personally find the landforms a little bit unsettling because the individual parts just don't seem to flow well from one to another. For example, there are a number of glaciated mountain forms and then there's that big block on the right that isn't nearly as worn. I also wonder if maybe there was a different scale used for some of the results? The top-left mountains just feel out of scale with the rest of the map (but my eyesight has been going in the last few years, so it might be me). I also wonder if the auto-exposure feature of the Tangram Heightmaps browser might be contributing a little bit because auto-exposure will tend to make things appear more rugged than they otherwise might.

I agree with that, a way to solve is to apply the wilbur erosion or other form of erosions like world machine, gaea etc , but Wilbur allows for massive multicontinental scale, while the others , well I never got convincing results.