I saw your question and thought that documenting the embark process might help you understand enough to experiment on your own. As an example embark, I have selected the conflux of two streams. This is a pretty common feature so finding a similar embark shouldn't trouble you.
In my (not particularly unique) opinion, these next three tab screens are the most important for predicting an embark's landscape.
The default tab gives us a few important clues as to the nature of the selected embark, though these are easily discovered by most players at a glance. If you're familiar with the game you can safely skip this paragraph. There is only one biome, a temperate conifer forest, and from the listed traits we're sure to find a lively mix of trees and shrubs. A fairly normal assortment of forest denizens is predicted by the Wilderness savagery level. The temperature is Cold, meaning you will most likely experience an annual freeze anywhere from late summer to early winter (this is not !!SCIENCE!!, just experience talking). "Stream" is important to note if you are interested in using the water's course as a natural moat, since a "Brook" lets creatures walk right over it. Stream also implies a flow of water about three or four tiles thick. Very Deep Soil suggests a few z-layers of surface soil, which could be good or bad depending on your designs. I will assume that you have already encountered aquifers and understand what sorts of foul things they can do to your spectacular designations. The metals are best left for some other post since there is quite a bit of depth to vanilla predicting and most players will just use DFHack Prospect anyways. I'd like to point out that this embark isn't that interesting to me, it was just the first conflux I could find.
The next part is the relative elevation tab. Here, we can get a veeery rough idea of the landscape. I have selected an abnormally large embark, and to give you a good idea of what your typical conflux looks like I have selected a 3x3 teal cliff while leaving one square of 'space' on three sides. This represents the stream of higher elevation merging into the lower stream, which usually results in interesting landscapes and waterfalls.
The final part would be the Cliff Indicator, which provides a more concrete way of scouting for waterfalls. Seeing a double-thick layer of red *'s (very high cliffs) surrounding a trail of blue 0's suggests a stream or river leading to a fairly dramatic waterfall. Personally, I don't like really extreme cliffsides since they betray the underlying procedural generation too much to feel natural (you will see what I mean in a second). High Cliffs or smaller usually leave room for interesting variations in topography.
And now, I've recorded a .gif of the resulting landscape so take a look (I would insert it here, but apparently Imgur translates larger .gifs into .webms now).
http://i.imgur.com/mib5ek3.webmFrom that video you can see how the central plateau + waterfall were somewhat reflected in the relative elevation view from earlier, but we didn't know about the southeast - northwest meandering of the canyon stream nor that the northwestern portion of the plateau would have a part that juts out from the cliff (this post would be easier to write if I had payed attention in my geology courses). Overall, the 3x3 teal square was a poor indicator of the smaller details, but the larger formation of plateau - cliff - valley floor was predicted correctly.
In my opinion, this embark is potentially interesting but the squareness of the cliff base as well as the triangular shape of the waterfall's basin throw me off. Plus it's just too large to play at a consistent FPS. I also enjoy when the cliff ends on a larger, sprawling valley floor and this embark feels claustrophobic in that sense. So I've made three more embarks, all nearby this plateau, just to show you how subtle changes in embark choice can change the entire feel of the landscape.
Embark two:
Here, I have decided on a 3x3 embark to the right side of the plateau, capturing only part of the stream and leaving out the waterfall. This translates to a steep, forested valley with easy access to cliffside minerals and a surface water source.
http://i.imgur.com/ioamx7p.webmEmbark Three:
This is a 4x3 embark located to the eastern side of the plateau. Looking at the relative elevation view, I tried to capture the edge of the plateau while leaving plenty of space for a flat valley floor. You can see here how difficult it is to know what you're going to get, since that 2x2 portion of elevation that I had interpreted to be a valley floor ended up looking like a small, rounded dip. The relative elevation didn't lie, it just left out the fact that getting from a teal elevation to a dark blue elevation is a gradual process. Even considering the cliff indicator, the markings suggest dramatic, tall cliffs but we're left with a more gradual, sloping descent.
http://i.imgur.com/8xMU3s5.webmEmbark four
The final example is a 3x3 to the northwest of the plateau. This leads to something that feels like the meeting of two valleys, with a natural landbridge or separating wall in between. Probably my favorite from the four - I could easily visualize a few different fortress entrances or surface constructions. I still dislike the squareness of the valley floor but I mean this is DF, concessions must be made. You should also note how impossible it was to predict this formation without scouting ahead via the larger embark.
http://i.imgur.com/oBVISBP.webmI might make my own thread to continue posting examples (they are pretty fun to make), but for now I just want to give you the advice that taking the time to find a perfect embark is extremely important to fortress longevity. The character of the surface helps you contextualize all those crazy events, which means you're more likely to keep the fortress going when things slow down. It's more exciting to watch a goblin squad ambush a caravan traveling over a waterfall's ridgeline than it is watching the same ambush happen on a flat plain. Totally my opinion, of course, but my larger point still stands: the more time you spend finding that amazing embark that really inspires your imagination, the more fun you are likely to have.
Don't be afraid to scout by embarking with huge, 8x8 areas selected. Just identify an interesting formation or two, then kill the process (DFHack is useful for this), and embark with a more manageable 3x3 or 4x4. As I've shown, changing your perspective on the landscape can change your entire experience.
Author
Topic: Tips for picking an embark site (Read 1644 times)