Yeah that’s more my style.
Be an engineer. But go to class for good gpa. Gpa matters a lot if you are a graduate and don’t have any good work experience.
What kinda average we talking about?
3.5 will get you an awesome job.
I got tired of debugging the path generator so I’ve added a setting generator.
Maps look much better now 
Any progress is good progress, hope to see this completed some day.
It’s gonna be in beta in a month max I’ve stopped my “à distance” c++ courses to spend full time on this.
Here it is with paths working, the c-shaped things are still not doing what they should be doing but I finally figured out something to make paths work, patchy yes, but it works.
lol don’t stop doing important stuff for Graal
take a distance independent study course about graal dev.
I don’t understand how this is supposed to be a map generator? Is it just supposed to generate a base idea for a map, or what?
The graphic output is just a visual representation of what’s in the levels, which I’m using for debugging purposes.
It generates paths and walls through bytecode, that are going to be represented by a dirt path in the level and by the absence of a door respectively.
It also generates which setting will be where on the map, so a pale green square will be, once you visit it, a grass level that is going to be randomly generated on the spot as a grass level.
The main problem I ran into is that since the placements of areas is random, it sometimes happen that an area is squeezed between two other or that its center is too close to another area… making it too small to contain castles.
The world generator also generates the links to other areas, where they will be (so when you visit the level you see you need a certain item to get through) and what item will be needed to advance. Thus deciding what item the main castle will contain. It also decides the secondary items… and where houses and castles will be because every square is the representation of coordinates in the actual world.
It also chooses your starting weapon, which has an effect on the map considering the swordsman will start at a castle, the archer too and the monk will start at a small house/farm.
You’ll be able to choose between a pregenerated world(for the impatient like spooon) or between Tiny and Small world (and more sizes up to complete once I made more bosses). So you can see how this has an effect too on the world.
You also are going to be able to choose generator options ranging from only generating the first area right away to generating all the levels one by one (that is, if graal can support it).
As to the features that are not generated with the world but that I’d like to point out;
SPOILER:
[COLOR="#FFFFFF"]The game is going to keep high scores of the lowest playtimes and the lowest death counts once you finish the game.
I’m currently debating wether or not to add a speed run mode where the bow and bombs don’t take time to charge, you start with 5 hearts and the player walks faster.
I’m currently wondering wether or not I should add feats too that you obtain upon doing things such as finding all the hearts or finishing the game under a certain amount of time.
You’re going to have 3 items max that you can carry at once(there are going to be places in castles where to leave your extra stuff), with the option(chance) of getting the packsack which lets you have 5 items max.
You’re going to have also access (if the random generator decides so) to a heart finder which will take up 3 inventory spaces but show you the way to the hearts you missed.[/COLOR]
As it is an idea I like, here’s an update.. using old material, but it seems I worked on it between last year and today as some bugs disapeared. As it is the time of the year I seem to decide to get overworked and jump into many projects at the same time, I’ll try to keep this project aside until I finish more important stuff.
Somehow, the paths and world linkers started generating right since I stopped working on it.
There seems to be situations I haven’t thought of, such as a linker appearing in a square that can’t be accesed from the rest of the world because of an inside level, roads that are not generated because the linker is not the world’s main linker, some levels seem to be generated but not graphically (bottom and right of the map). Circle roads, such as in the top right world, not sure why.
I also need to think up a way for the generator to work around the trimming you get on the strings when you play online.
.
.
I also don’t remember posting the results I got with my dirt path generator. It could use some polishing but it works quite well. I used tricxta’s implementation of the A* algorithm for it since it goes around bushes.
[ATTACH=CONFIG]3755[/ATTACH]
[ATTACH=CONFIG]3754[/ATTACH]
lol
I’m mildly interested in how this will be used in any significant manner.
What’s really left to do until it can be considered playable is a workaround to the string trimming and a dungeon generator.
Then, it’s minor bugfixes and icing.
Why did you decide to use gs1 instead of generating levels outside of graal?
Because they’re generated on the fly with a different world at each new game and as far as I know, I can’t tell gs1 to run anything on a server that isn’t gs1.
I could hack the gserver but I didn’t want to do that/didn’t think of it at the time I started scripting the generator.
What we mean is: why do anything for an outdated version of Graal when we have a couple of projects moving away from it?
Well that’d imply rewriting all I did and having nothing to test my code with. It’d be pretty much a risky jump to rewrite the whole thing without testing every step… as the occuring bugs would be harder to find. Also, the newer projects all promise (I think?) compatibility with gs1… the difference is my scripts would probably run faster and that’s a plus.
Unless Cadavre’s project is far enough in development that I could code for it and test what I do, it’d be meaningless right now to start coding it in C#… I can see the advantage though.
EDIT:
Well since they’re all meant to incorporate gs1 I guess it’s safer right now to just script it in gs1 than to end up with something not compatible with the favorited client. Wheter it’s codr’s, Cadavre’s or Jatz’s. Past that, I’d have to write my own game, which I don’t want… I want to make a fun graal pw.
lol Do whatever you want. I don’t really care if you waste your time on pointless tests.