So my brain can't stress enough the fact that, in your imgur album, there are some cliffs applied to biomes other than the one that's available on the mod, like the snowy mointains and the "rock" biome. Will there ever be the possibility that this mod can be applied to the other biomes/planet types?
Does anyone else have the problem where the sandbox planet is always a lava planet with trees and cliffs I cant figure out how to get a moon with cliffs or a earth planet with cliffs.
Lava planet with trees and cliffs is the biome he decided was his favorite, and so the one he released to the masses. The other ones would require a separate mod he hasn't released.
Any idea how long until the mod might be done im really looking forward to this I feel like the current PA terrain is lacking in variety .
Not any time soon. And IMHO, it isn't variety you are lacking, so cliffs wouldn't change much either. It's rather the lack of meaningful terrain formations which makes most planets appear so boring. Just adding cliffs won't help this, unfortunately. Even cliffs need to be placed carefully in order not to accidentally block of pathing in whole nor to leave to many gaps. The current planet generation method doesn't support this at all. It simply can't.
I disagree with you on this, and I don't think you have a solid argument for it. Whether brush placement is 'meaningful' or not is pretty subjective and contextually dependent. Even brush placement in vanilla biomes is done in a somewhat meaningful way, because it's determined by a weight map system that places brushes in certain areas with certain frequency. It *is* possible to spawn brushes along longer straights with higher frequency, such as with mountain ridges. This is of course very limited, but it's not meaninglessly random. But in the case of this mod, cliffs are explicitly placed at a specific height by giving them their own biome, and the frequency of cliff brushes is also set such that ramps are somewhat infrequent. This divides the terrain into two regions; high and low ground, with few avenues in-between (barring current issues with pathing / me not using a custom brush). While the overall layout is not deliberate, the placement of the cliff brushes in relation to it is much more deliberate (although the actual implementation is pretty sketchy for other reasons).
Only thing I could do right now is give it a custom brush (which I probably won't do until I get a good description on the import process for brushes, since I messed up my first few attempts and can't be bothered to try more blind testing). Implementing cliffs correctly with this method would require a lot more tools than are available though, so I won't be expanding much on it until I can script in brush placement behaviour somehow. Also, there are a couple other mods I need to get around to first. And some TF2 Tshirt designs I've been working on. And there's another project of mine I want to get around to finishing before that. And before that, I have a few important IRL obligations. (And between all that I've got playing videogames to let off steam, or I'd probably go nuts )
The Github repo linked in the first post, is it up to date? Because I can't seem to see how you managed to align brushes with equipotent lines in the terrain - or even how you managed to specify that cliffs should only occur on inclinations. Or is the latter one not enforced at all? That would, of course, explain what you meant with pathing issues, declaring the brush as impassable is impossible if the brush is also applied to flat areas - even when it shows no visible effect. Dynamic calculation of traversability based on inclination is yet another missing feature... Being able to define relative constraints on a second sample point is exactly what you would need, isn't it? E.g. absolute elevation constraint for the base point, and relative elevation for an arbitrary second sample point. This would also allow to align features with either equipotent lines or inclinations in all noise textures in general. And I'm not convinced by the noise based approach yet either. Even if you specify a random spacing in between the cliff brushes, you are still not guaranteed that you don't accidentally create sealed off plateaus or valleys. A "good" map is more of an exception rather than the default case (only exceptions being IMHO the corrosion free metal planet and moon biomes, both of which don't rely on noise for path blockers at all). Also, like you have noticed yourself, it would be much more comfortable if the height map did adapt to the presence of certain features - and not the other way around.
I think the Guthub files are up to date. In order to align the brushes at the same height range, I've set them in a custom biome that only appears between a narrow height range. This isn't ideal, obviously. For starters, it means brushes will appear more frequently on plateaus at that height range instead of appearing most on steep inclines. So your last concern is correct; I don't have any way of doing that yet. No matter what, I would have to declare the brush as passable because the bottom of the brush is meant to be pathable. I don't think that's correct. As I said, the cliffs in this mod are labeled 'pathable', but the steep sides of the cliff faces will end up being unpathable (so long as they are steep enough). Pathable brushes have unpathable areas calculated for them, It just so happens that the allowable incline seems to be pretty steep. It's quite common for a shallower section of a cliff, which may appear to be unpathable, to have a small section that is pathable. The problem with this is that it's very hard to tell when it is the case, especially from an RTS camera. The solution is a custom brush with vertical sides starting near the bottom, but as I said, haven't gotten importing brushes down. Not exactly sure I understand the system you're describing. Elaborate? (I've already laid out what I think I would need in order to get nice cliffs, and I doubt anything short of scripting would work too well) This is true. The possibility is interesting though, don't you think? Depends what you mean by 'good'. Do you get full control? No, you never do. But that doesn't make it 'bad'. I think there's a lot to the current planet generation that could use a re-think, but I don't think random generation is the problem. Yes, I agree with this.
Unsurprising, I heard stuff had been changed that might affect it (I uninstalled all my mods about a week ago, apparently a lot of mod-breaking changes were made before launch). I'll take a look at the files, see what I should do about it.
I just haven't tried any time recently TBH. If anybody has a good list of settings / a functional pipeline for it, I'm all ears.
I haven't had a go at it myself, but I am guessing it's to do with how the textures are set out. Unit models keep their textures in the same folder, however brushes reference them. \papatran.exe [options] files options: -o/--output FILENAME output filename (required) --no-materials exclude all materials --texture-mode 'include' (default), 'strip', 'reference' <--- --no-baked-anim do not bake animations --dependencies FILENAME log dependencies to FILENAME You should be able to export it as fbx and then convert it to papa via papatran. Seeing as brushes doesn't have bones, there shouldn't be any issues with blenders fbx exporter.
Or the exporter from blender should work correct? And do Terrain features really not have bones? Shouldn't they have at least one bone to show what their base is?
Textures are different for meshes. They are referenced in json files rather than placed with the model. As for bones, papatran sorts out all of that.
For the record, I'm a bit busy this week, so I won't be working on this stuff right now, but I'll try to work on it again when I get back next monday.
Why isn't a dev just posting here with the correct process to correctly bring in brushes? It doesn't make any sense to me why an amazingly basic but awesome feature like *FRICKING CLIFFS* is being denied to everyone because of... why? This should be a feature in the vanilla game. Wow.