Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Sentientdeth]

Thanks =)

[Lord Dullard]

If you could choose randomly from more than one tile for displaying these textured regions, that would help to reduce the griddy visual artifacts further.  (I don't know if libtcod imposes font size limits on you or whatnot.)

I could do this, and it wouldn't be extremely difficult to implement the code, but I'll probably wait. I'm not too concerned with making things look visually perfect right now - especially when it's just crappy developer art I'm using (i.e., my work). If/when somebody who can actually work with tiles properly wants to get on board with this, then I'd be happy to get more in-depth with the graphical end of coding.

[Lord Dullard]

Been working very heavily on creature spread code for the last few days (when I have time, anyhow). I have the next two days off after a double today, and I'm hoping to get creature spread into an acceptable state during that period and move on to the next phase of development. I'm going to leave worldgen in its current unfinished state, since that seems like it's going to be a necessary evil, and begin work on the game engine itself. Theoretically, that will allow me to integrate new changes to the worldgen with new changes in gameplay itself.

Appended: I added a link to Sentientdeth's blog to the Cult page.

[Sentientdeth]

Sounds awesome! =D

[Lord Dullard]

Other than my needing to add a lot of complicated ecosystem-forming algorithms once I can delve into worldgen more, creature spread (for amphibians and terrestrial creatures, anyway) is now finished. I've got it working functionally, if not perfectly, although it's definitely a resource hog and will probably end up being the most time consuming part of worldgen as it is. I'm thinking I'll eventually tweak the algorithm a bit so that the actual expansion process takes place over blocks of tiles rather than just one-at-a-time, to speed things up.

But the most important part of it - the fact that it will give  adventurers something to hunt/kill/interact with/whatever as a placeholder until I can do more work on it - is functional.

Which means I can now, officially, get started on gameplay. Whew.

Also, I did more work on the terrain tiles; there are now four per tile type (mountain, water, grass) which are randomly selected for each tile space to make things look less repetitive. Hopefully it helps; I'll let you be the judge:

Warning: HUGE IMAGE

[Neonivek]

That image reminds me of SimEarth

Where the world actually seems somewhat nondescript until you strip away all the creatures and plants. and see that the map is actually quite detailed.

I hope you work on forest tiles a bit more so it doesn't hurt the overall map.

Edit addition: MIND you, I am not saying it is the same situation. Just similar idea.

[Askot Bokbondeler]

much better tiling now

[Lord Dullard]

That image reminds me of SimEarth

Where the world actually seems somewhat nondescript until you strip away all the creatures and plants. and see that the map is actually quite detailed.

I hope you work on forest tiles a bit more so it doesn't hurt the overall map.

If I can find somebody decent enough at pixel art to make multiple varieties of not-horrible-looking 10x10 pixel trees for each type of forest/swamp/etc, awesome. But *I* certainly don't possess that ability. 

[Lord Dullard]

So, I accidentally got to work an hour early today. Since I've got an hour to kill, I suppose I should be productive and lay out a rough outline of how I'd like to proceed at this point.

1. Get the basic framework for movement on the world map in place. This includes passage of time and weather effects, and later integration with random encounters and ambushes and such.

2. Begin procedural generation of different terrain types on a close-up scale. This is presumably going to suck up a LOT of time, since there are so many different types of terrain and biomes to cover.

3. Once 1 and 2 are complete, get basic gameplay mechanics like pathing and LOS into place. I'm not terribly worried about this step, necessary as it is, because those mechanics have been so widely hashed and rehashed that documentation on how to handle them is widely available.

4. Handle creature spawning and basic combat, plus dungeon generation for simpler dungeon types.

5. Get a basic crafting and building system into place, including modifiable terrain and resource gathering. This will probably go hand-in hand with early vestiges of a recruitment/group-organization system so that you can send villager flunkies and peasants out to do your menial work for you.

This should keep me busy for quite a while, or so I can only imagine.

[Sentientdeth]

Alrighty, I've got the Summer track *done*.  I put that in asterisks because I'm not completely satisfied with the result, but I've tinkered with it for three days in a mostly completed state, and this is the best I could come up with for now.  I may end up scrapping it later down the line and re-writing it, but for now I'll leave it, as there are a few more tracks I'd like to get done before gameplay comes around, namely Winter, Spring, and combat music. 

I have some experience doing pixel art, but honestly, I don't think I could really do a better job than what LD has, at least in a 10x10 image.  If the images were larger, I'm sure I could make some pretty little icons, but I'm sure that would probably require a lot of coding to change, and at this point, I'd rather see gameplay get started than prettier pictures. 

You can hear the Summer track at my blog.

And LD, I really like what I see!  Keep up the great work man!

[Lord Dullard]

Loving the new track. It's very... for lack of a better word, dwarfy-sounding.  I'm spending this morning doing some more work on code; mostly some tweaks to creature spread to speed it up a bit before I get started on gameplay basics. I'll post some screens of creature spread in a moment.

[Lord Dullard]

Spoiler (click to show/hide)

There is a shot of creatures populating the world. The current problem I'm facing is that while it works exactly as I'd like it to, it takes too damned long. And I know that if it's taking too long on my computer, it's going to be completely unacceptable on older rigs.

[shadenight123]

what is the "way of thought" for creature spreading?
maybe you could try and make it simpler, or less memory requirement (aka random with a bit of constriction. [like, next to salty water, only fishes or crabs] )

[Lord Dullard]

It's a bit more complicated than that. There are plenty of restrictions on biomes that creatures can inhabit already, and anyway, the goal is for the world to be widely inhabited so that there are no 'empty spaces'. There'll probably be a few of these anyway, but I've thought of some shortcuts for eliminating completely barren wastelands. Everywhere should be inhabited by something.

The problem is that there are hundreds of creatures spreading out over the world, and each one needs to test each square that it's spreading to in order to see if that square is a compatible biome. If there's a solution, it's going to be syntactical, not theoretical. I need to rearrange how I'm testing each tile on the grid in some way that's more efficient (i.e., avoid searching a whole dictionary or list for each test).

[shadenight123]

maybe go "matryoshka" style.
divide things in really really simple sectors, and then complicate them afterwards.
so to say:
"hospitable"-->"greenery"/"watery"-->forest/jungle - lake-rivers
"non-hospitable" -->"deserty"/"mountainy"-->desert/beach -peaks/tundra
and you "could" place a random call everytime a creature is there that, in the event of it being impossible to live in a "non-hospitable" biome, it would avoid checking for the things underneath, thus cutting in half the work of the processor...