Super Realistic Topo Terrain

[Peter Toth]

Dear Guild,

Here is simply another wholehearted attempt to create super-realistic terrain, without using any pre-existing satellite DEM data. For those of you who prefer the "human element" of roads, cities, and such, I'm sorry if you're disappointed. My preference for mapmaking is to emphasize nature and present pristine spaces devoid of human developments, for I'm highly dismayed by the degree to which we've polluted and developed the earth. Indeed, worldbuilding/conworlding allows me access to a world teeming with unspoiled natural wonder, even if every square meter of this earth becomes marred by high-rises sprouting unchecked, roadways branching ad infinitum, and--the ultimate anathema--gridlock and associated toxic gases!

Perhaps when I've mastered topography, I could progress to other aspects of mapmaking. Please let me know how I'm doing, though.

Thanks,

Peter

Baumann2.png

[QED42]

That is some super cool looking terrain! I'm itching to know what process you used on it.

[Harrg]

Hi Peter, I love your progress. If you want to say that this is a manual DEM simulation, then you have achieved good results. This is a rather complicated process.
If you want more detail, you can try the Pixie method. But be prepared to spend a lot of time on small areas of the map.
A few tips / criticisms that you can ignore. Sor for my eng.
1. You have very deep rivers, it feels like fjords. I do not feel the difference between the rivers in the valley and the mountains. This is due to the fact that your rivers seem to be on the same level with the sea. Perhaps in some (most) cases it is better to lower the land relief by 1-2 levels in those places where the river flows along the entire stream. Depending on the scale, we may not always see water. You are already using this method for streams
2. You can improve some rivers by adding initial meandering steps. And monitor the thickness of the river not only in relation to the mouth and source, but also the relative thickness of the flow in certain places and bends
3. You do a good job with the shelf. But for now, it simply follows the outline of the land above the water. Do not forget that in some places rivers can carry a lot of soil. The relief will be softer here
4. Do not forget about the current, it can flatten the coast in some places
5. If you want to get the best results with mountains, see how mountains have different shapes depending on tectonic patterns and erosion. This affects both the shape of the mountain ranges and the shape of individual watersheds.

[waldronate]

Your mountain lakes have rivers that flow steeply uphill, which is decidedly unrealistic. The green rims in the topography around the lakes says that they are punched down to sea level. Those lakes have rivers which flow down into them. Then the rivers hop up out of the lake and flow over the nearby higher terrain to get to the sea, which is physically implausible. I recommend having a separate layer for lakes and then compositing it over the base terrain ( https://www.cartographersguild.com/a...chmentid=70257 offers some suggestions for getting interesting lake depths without damaging the surrounding landscape ).
It's still a little hard to tell the intended scale of the map. Comparing the topography to the closest equivalent here on Earth (Norway) suggests that the map is about 800 miles from top to bottom.

[Peter Toth]

Hi QED42!

The algorithm I use is very complicated (shuffling back and forth from Wilbur to Photoshop many dozen times) and doesn't always produce favourable results, as many might have witnessed in my mediocre previous submissions. Thank you for your interest, however. Creating that tutorial is on my mind, but first I'd like to put several months of solid work into a colossal-sized project that I intend to release this summer.

[Peter Toth]

Thank you so much for that information, Harrg!

This map represents a small piece of a rather large project, and I'm glad I released it and asked for criticism before submitting the entire project (which will require hundreds of hours of work). I'm trying my very hardest to create atlas-accurate maps, although sometimes my algorithm produces unfavourable results which then detract from the plausibility of my maps. Like the rivers. Yes, many of them are as deep as the ocean, which I didn't intend. I'd love to try Pixie's method--where may I find her method?

When you said that my relief should be softer near the rivers, did you mean a gentler slope of the continental shelf?

Thank you so much for pointing out those details. I will try to resubmit this map after I've addressed each one.

[Peter Toth]

Thank you waldronate, you have pointed out my weakness--the rivers and lakes; yes indeed, they detract from the plausibility of my map.

In my process, I select flat areas using Wilbur to create a BMP file showing the lakes, which are then allocated to a separate layer. Somehow, though, the lakes became so glaringly large, that I decided to try another method, lowering the altitudes of the terrain to place the lakes. And yes, that's probably why I have this embarrassing feature of "rivers flowing uphill" disgracing my map. With regards to scale, I intended my vertical distance to span 2750 or so kilometers. This is only a very small portion of a large project that I'm working on. Any ideas on how I could reflect the scale in my topography? (Without having to "cut and paste" actual satellite DEMs?)

Thank you so much for the critique.

[waldronate]

The deep coastal canyons look a lot like fjords (glacial canyons carved below sea level), which is why I suggested the coast of Norway as an example. Being off by a factor of two using something like Wilbur is not too bad. Making it controllable is the hardest part of something like this task and I'm not sure what you were aiming for when you started. A good self-challenge (even if it's only a private one) is to generate a rough map with the general features that a customer is likely to want (coastline shaped like so, high mountains here, low mountains here, plains here, rivers along there, and so on). Make a high-level map using that rough map and then a couple of zoomed-in areas from that first generated map. If you can preserve the required features and maintain plausibility, then you'll have something. Having said that, you're doing fairly well with generating terrain so far.

An important aspect of cartography is managing expectations. If you had started out with a scale bar on the image to indicate that the map is intended to be 2750km across, then you'd have set expectations and offered hints as to the scale of individual features. Wilbur is particularly hard to do plausible things with pixels much larger than a hundred meters or so because of how the algorithms work. Fortunately, another important aspect of cartography is generalization: only show what's important to the map. Pushing unimportant details to the background while keeping the important things uncluttered is really hard, no matter what tools you're working with.

I can't recommend enough the idea of looking at the works of others. http://shadedrelief.com/ is an excellent starting point for getting some how-to ideas. Decide if you want to go with straight hypsometric shading or if you want to look at other features as well such as moisture. I've said this before (and in no way am I trying to make light of your efforts): generating terrain is the easy part of making a map. Making the customer happy is the hard part.