Procedural Content Generation

[SomeoneElse]

I was hoping to see some more advanced approaches to PCG described in your article rather than a limited example consisting of random placement and one way of doing simple room descriptions. If that is state of the art in terms of PCG for text based muds, its rather disappointing.

To do PCG well, I believe you would have to go a lot further than that. An ontology and a reasoning engine. A fully 3D world (rather than room-based). Graphical editing tools so that you could edit your world in 3D. Theres no limit how far you could take it. Unfortunately, getting a code base to this level is a lot of work and I am nowhere near skilled enough yet to do so.

[angelbob]

Originally posted by SomeoneElse
I was hoping to see some more advanced approaches to PCG described in your article rather than a limited example consisting of random placement and one way of doing simple room descriptions. If that is state of the art in terms of PCG for text based muds, its rather disappointing.

It's not the utter state of the art, but it's not too far off.

To do PCG well, I believe you would have to go a lot further than that. An ontology and a reasoning engine. A fully 3D world (rather than room-based). Graphical editing tools so that you could edit your world in 3D. Theres no limit how far you could take it.

You know that doing a full world editor in 3D is defeating a lot of the purpose of originally doing the game as text, right?

I'm not sure what you mean by an ontology.

A reasoning engine could mean any of many different things, but I assume you mean "some kind of AI".

There are graphical editing tools out there, but the state of graphical editing tools for PCG in text games is no more advanced than the state of graphical editing tools for PCG outside of text games -- which is to say, they barely exist.