new map generator

[DrPraetorious]

That fixed it.

Thanks!

Making maps now.....

[DrPraetorious]

A few RFEs -
* control for the frequency with which different terrains are seeded. At the moment, these maps have no mountains, which is a game balance issue.

* border mountains. Not sure how to do this, as border mountains don't get seeded like other terrains.

Anyway, it works great, it's amazingly fast. Nice work!

[paradoxharbinger]

sweet.

im going to remove the other attachments, and post a new zip with an updated readme in it.

[paradoxharbinger]

as far as how terrains are seeded, oceans, land masses and mountains are determined strictly by a height map, so everything below x is ocean, everything above y is mountains. rivers are created in a similar manner. the other terrains are sort of 'grown' in the space in between.

try using the tlevel command

the default is tlevel: mountain 120;

so try using a lower number

impassable terrains, border mountains and the other dom3 terrains are something that i intend to work on after i get the sprite functionality in.

well i'm glad that the thing is working, though i'm quite annoyed that it can't be simply copied to another machine.

[DrPraetorious]

Oh, I get it. That sets the water level! Okay, that'll work fine.

I think I found a bug in one of the maps it generated (attached). Provinces 10 and 11 appear to have their terrain assignments reversed. There are a few other terrain reversals.

If I had to guess, I think that the script is numbering provinces wrong whe the center (white) pixels have the same y-coordinate. That's just a guess though.

Thanks for the awesome tool!

[paradoxharbinger]

that may be. i will check it out.

btw, can you post the debug for this map? i'd be interested to see ho long it took to generate it

[DrPraetorious]

Oh, sorry, I overwrote it already.

This one is about the same size - took 1 minute and 14 seconds, it says. Gah, that's just for the last part.

Total time is about 15 minutes, counting all the steps.

[paradoxharbinger]

well i took a look at the one you posted, the centers are very close to having the same y coordinate, but they differ by 1. they may have had the same y-coord before the province centers were adjusted, but i cant tell without the debug, so if you see it again, make sure you save it.

speaking of which, first priority is making debug as verbose as i can to address any issues anyone finds