Forum
Projects Stranded III Dev. Blog - CommentsSecond, you really 'dont' know how modding works, hence the "release da gaem so i can mod it" comment.
And third, you dont update said software, you replace it;
your PC is garbage.
and also.. About how you were talking about floating things and caves just not happening. Maybe you could rethink this a different way. For ex. Above the terrain make a polygon with lower coordinates and upper coordinates while the terrain only has upper coordinates, meaning that the polygon can start and stop above the terrain making caves or floating islands WITHOUT using up tons of space or performance power. Of course if you want to.
Also (this idea is crazy i know) the absolutely awesome possibility of adding a 'model' info with which you can make a 3d model in sort of the same way that you can place things on titanium version (scroll=up and down. Mouse to move things in 3d space) I have absolutely no idea how you would integrate this, but I do believe you might have the skills and background (correct me if im wrong)
edited 3×, last 11.01.19 09:29:30 pm
while i assume the S3 AI will be leagues ahead of S2 ai, i would imagine the AI would probably still be fiddly and probably wont have good pathfinding in enclosed spaces like a cave.
@ DC has written
Further Optimizations
Another thing I'll change is the resolution I'm using for the random map generation.
I was insane enough to work with a 1 m² precision for all operations. Now I plan to reduced the base resolution to 4 m² or maybe even 8 m².
Imagine a 256 m² island. With the current calculation this makes 256x256 which is 65.536 pieces of data (each representing 1 m²).
With 4 m² precison the values change a bit: 256 / 4 is 64. 64x64 results in only 4.096 pieces of data. That's only 6.25% of the original data and therefore allows a much faster world generation. With 8 m² it would be only 1.024 pieces of data.
Another thing I'll change is the resolution I'm using for the random map generation.
I was insane enough to work with a 1 m² precision for all operations. Now I plan to reduced the base resolution to 4 m² or maybe even 8 m².
Imagine a 256 m² island. With the current calculation this makes 256x256 which is 65.536 pieces of data (each representing 1 m²).
With 4 m² precison the values change a bit: 256 / 4 is 64. 64x64 results in only 4.096 pieces of data. That's only 6.25% of the original data and therefore allows a much faster world generation. With 8 m² it would be only 1.024 pieces of data.
Dont you mean 16m² and 64m²? Since it's 4*4 and 8*8 meters?
What data type are you using to store the height values btw? Looking at the height differences on maps in S2 I guess byte could be enough, giving you 256 values to work with. Scaling them to 1 unit = 0.1m would allow a total height difference (bottom of sea to top of mountain) of 25.6m. Or 51.2m with 1 unit = 0.2m. Or something in between.
Adding a chunk system to that could allow setting up even larger maps at the original 1*1m precision. Loading only a few 16*16 chunks (256 byte each) around the player should help keep the memory footprint low enough.
Right now I'm using a short (2 bytes) for heights because a byte is too limited. I will however probably change the format a bit because I don't need all of these 16 bits for height and therefore I can use the remaining bits to store other map data.
I'm already using a chunk system so I don't have to load/transfer/set up the entire terrain at once.
DC has written
use the remaining bits to store other map data
Right. There's probably also a need for a color per vertex and some texture weights...
Do you have some compression/decompression algorithm in place yet? I can't remember whether this was already covered in another blog post right now. Might be nice to zip the whole chunk though so at least on the disk it won't take as much space
Did you have a look at the new Terrain features in Unity 2018.3 btw? Although I haven't used the new Terrain yet myself some of the new features look really interesting, especially GPU Instancing
So basically all the ai stuff would be done in 2d while the actual virtual world is created in 3d. Or maybe you've already done this...
Assassin moder the ai might be a problem. Maybe two height maps, one for the ground and one for added islands (this would only allow one floating island in one place, but that's pretty much all you need). The ai would be able to sense the height difference and decide that (unless the animal is a kangaroo or something) it could not jump that high.
I'm planning for stranded II Enhanced to be a pre-stranded III mod, kind of like an appetizer before a meal.
DC : I'll try to get permession to get more models from stranded II mods into stranded 3 to contribute to it's development and hopefully save you hours of work.
Quote
DC: I'll try to get permession to get more models from stranded II mods into stranded 3 to contribute to it's development and hopefully save you hours of work.
WHAT?
I was trying to not be agressive but I am very irritated by your posts more and more. Stop acting like you know everything when you know nothing.
Because I only know how to do stranded II things, I'll stick to stranded II and use the skills I do have to organize models and add them to my mod, clear out junk, and I am planning to make maps for my mod after I'm finished with it. I wish I could give DC permission to use all the models on my mod, but I myself don't have full permission for most of the models (but I can't really get it because half of the good modders left years ago).
My hope is that they'll be happy to see their models going to good use as their mods have gotten old and dusty and are pretty cluttered and links are expiring all the time which may cause modders who barely remember stranded II to lose all the work that they did years ago. I honestly don't really want those models to die so I'm offering them a second chance. Anyway, since I already realize DC probably won't be able to use the models for Stranded 3 I will probably turn my stranded II mod into a stranded 3 mod after it is released.
So basically it isn't stealing models in my case but simply giving them a new purpose. I know I haven't had the time to mention their names in the credits. I'm planning to do that in my next update which will include a lot of cool stuff, including my first unit extension.
You expect a game developer to use a past game's mod developer's resources? are you insane? no artist will EVER use fan art.
DC is the type that wants to do everything himself, i cant blame him, having the creative input of every little detail in a game is very fulfilling if not very challenging.
You need to think about what you say before you say things, you AREN'T a game developer, you aren't DC's advisor, stop assuming you know what your talking about, its driving us all up the wall.
Besides that, Malzar kind of has a point, it is taking a while. Because of the fact the progress seems to still be going as slow as always, I will soon try to keep people sane by trying to keep my S2 mod going. Of course, although it looks the same as always, that doesn't mean i'm not making progress. It's just that I have a specified release date.
edited 1×, last 02.04.19 07:17:54 pm