Thread: Is there such a monster?

I understand what Gameprinter means. What the op is asking is the equivalent of an author asking for a program to generate cover illustrations for a novel, thereby skipping over the need for the artist, or a movie director wanting virtual actors so as they can by-pass the real actors.

I found this! http://www.the-elite.net/story-generator/

<<shudder>>

As a person who writes a mapping app and also some simulation software like my GeoTerSys which stands for Geological Terrain Simulation I have some experience in that area. I can see Gamerprinters point of view in that at present there does not seem to be any app which will make pseudo random terrain in which all of the features of it are plausible to reality. The Turing test for A.I. says that if its not possible to tell whether its real or computer generated then its intelligent.

Terrain is a physical process and as such it ought to be possible to model with a simulation if you include and process enough of the physical processes to a degree where it represents reality. It ought to be possible and maybe there are some research units running something like this on big mainframes but I have never seen one and I am pretty confident that there does not exist one for a home computer.

My simulation handles some processes and to some degree of accuracy to the real world but not enough to make every result plausible. Some more apps make even more random terrain or have pretty good generation for certain types of terrain and not for others. All of them tho dont do it to a level of detail that is sufficient so as not to be able to tell its computer generated (if you know what your looking for).

Community and town placement are a human decision thing but I would argue that for one person it could be based on quite bizarre decision making but for many people like a village up to city then its based on processes that can be modeled. Usually by maximizing the availability and efficiency of natural resources. I have wrote my simulation and posted about that on this forum and showed that its a very complex task to simulate the workings to a degree that are plausibly real.

Now people are good at this kind of thing so G.P is right in that in the absence of these apps a person can usually reason out where towns, villages and terrain might be located. My person feeling is that at the mo this is the best route but in time it ought to be possible to write an app that does this better than a person. For example, we have the ongoing issues with river placement as that is a simple enough rule to burn into computer code that a computer simulation usually does better than a person.

So it depends on the kind of terrain you want, the kind of villages and towns you need and crucially, the users experience and expectations about the map. Since Tolkiens books are the most successful fantasy books ever written and his map was extremely crude then for him it didn't make much odds about its plausibility and put in the mountains, forests and rivers in such a way as to promote the story that he wished to tell. And that's something that is unlikely to ever be burnt into computer code.

So is there an app to make random terrain - yes. Will it stand up to the requirements that it fits into a book story - maybe. Will it stand up to experienced scrutiny - almost certainly not.

I agree that there is certainly intuitive software, such as Redrobes app, that could emulate the creation of realistic terrain to some degree. If you have the scientific mind to appreciate using similar software, then you might qualify as a fictional cartographer relying on software at some level. Still cartographic skills by a human can enhance the results of such software.

Really the whole point of my Kickstarter project is so that writers, gamer designers, editors, illustrators, cartographers and page layout can have their work accomplished for my proposed printed books, for at least 1 book and perhaps 3 - and that all the professionals involved get paid up front for their proposed work. Nobody is left in the cold awaiting possible sales to pay for their services.

Edit: let me add that if your writing is for a personal game, or other low exposure situation - simple maps without true cartographic skill should be good enough. However, if your intent is for publication and sales, proper writing and proper cartography is the expectation. So for publication only, standards are much higher.

Originally Posted by ravells

I found this! http://www.the-elite.net/story-generator/

<<shudder>>

and it gave me THIS gem: 'MY PRECIOUS!!' she cried, as she reached for a sharpened ninja star. With a hasty thrust, she buried it deeply into her prostate.

<<shudders some more>>

Originally Posted by Korash

and it gave me THIS gem: 'MY PRECIOUS!!' she cried, as she reached for a sharpened ninja star. With a hasty thrust, she buried it deeply into her prostate.

<<shudders some more>>

Now I understand some of the more specialized fan fiction on the Internet...

"The Internet: A Dadaist paradise."

Originally Posted by Gamerprinter

I just find it surprising that writers (you're not alone, many writers start similar threads like this one) think their work requires serious effort, while the art or maps is a 'push button' solution.

Tell me about it. It's amazing the number of people who think the computer does all the work when you're doing CG animation. If that were true, my animation would be way better than it is!

Anyway, there are some tools that can automate parts of the process. String enough of them together, and you can arrive at something that doesn't require you to do much or any manual painting. As GP says, though, it will never look as good as if someone who really loves the craft had done it. That doesn't mean that such an effort is without merit, though. Bad art made with love can very often trump good art made without. Not always, but often.

For many years and many thousands of dollars I searched for the box of software that included talent. I found that I could use the toys in a lot of the boxes to make very pretty results, but each looked as imbalanced and unimpressive as the last. At the end I was forced to conclude that the creative spark, the difference between interesting and not-interesting, would require a massive neural network trained on all manner of inputs over a period of years.

I'm working on training up a neural network now, but I don't have any interesting results yet. Until I can get good enough results, I'll just rely on humans for the things that need to appeal to other humans and I'll rely on software for the other things.

On the subject of http://www.the-elite.net/story-generator/ - it's really just an old-fashioned Mad Libs template with a few machine-controlled randomizations thrown in. A human was needed to make the layout and generalize many really awful stories into one horrendous template. Not that software won't be able to do a pretty good job of analyzing bodies of existing work and making novel results based on that work; it will be and can do so in some limited areas now. Experienced humans will likely be able to tell the difference for the next decade or two. When the machines get to the computational power of a human cortex or beyond and incorporate true random processes into their basic systems, I don't think that we'll be able to tell much difference, and I'm not sure that we'll want to...