ogrenix

Yes, thank you for the reply! These are the types of considerations that are being pondered right now, so you hit the nail on the head. I certainly need this type of information,and the more the better! Right now, the project is still in it's planning stages; We are far from even beginning to program, let alone being able to show screenshots. However, I would be more than happy to keep you updated on it's progress. I am preparing to create a blog for the project as we speak, in fact. If you are curious on some of the finer details in the mean time, please review this article (complete with screenshots that look quite similar to what you may expect from this project) that outlines the basic proceudures used in typical world generation programs for roguelike games. It is very informative! http://www.nolithius.com/game-development/world-generation-breakdown Of course, as previously stated, while this method is a great way to get started, it is hardly a complete enough model for my liking. I think there are (obviously) a great, great many factors that are not taken into account. The results could be expanded and refined much more thoroughly with more data and some post-processing. I am looking for suggestions or observations on the origin of things such as land masses, biomes, elevation, climate...Basically, everything. Thanks again! I really appreciate it!

Hi there! My name is Robbie Ogrenix. I am an indie software developer working mainly with "Roguelike" RPGs. Many of these games feature randomly generated worlds using complex algorithms and rule sets. Needless to say, it's a complicated subject that is quite enough to keep someone busy without the help of all of the other parts of game design theory! So for this reason I have decided to undertake a new project: I want to make a detailed and variable world generator -- A program that creates a planetary map at random when given a set of rules. Of course, while I understand the typical methods used in similar programs, I am coming up with some complicated issues when trying to emulate real variables more accurately. Generally, a greyscale height map is generated using perlin noise. The luminance of each pixel is measured from 0 to 255 -- The higher the number, the lighter the color. Those high points are also the "high" points in terms of elevation. The darkest parts are, of course, pits and valleys, Then you create a water line by stating something to the effect of "if a pixel's luminance is less than the value set for the water line, the pixel is submerged," After that, you set some more rules for luminance ranges, high values become hills and mountains, with marginal values being plains, forest, hills, tundra etc. This generally is based on a temperature modifier which still is based in the luminance of the pixel, or "cell." While this approach is adequate for most, it is not thorough enough for my liking. In my planning, I have come up with many questions that need to be answered, spanning just about every field of science. Questions like "Exactly what role does elevation have on the temperature of a given point on Earth", "What types of soil are most fertile?", "Precisely what makes a vein of X ore?" and "What role does the moon play on the weather patterns of Earth?" So I am in need of a lot of information, you see! While I am educated, I am certainly not a geologist, astrologist or physicist. I really would like to better understand how all of this works. If anyone has read this all the way through, thank you very much! I greatly appreciate it! Kindest regards, r.ogrenix