Print

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

[Telgin]

I wish you luck if you do get around to it.  It's indeed very tedious, though it's pretty rewarding when you do get it working.  As complex as the graphics layering system is, you can add stuff like random tails pretty readily.  Some examples from my coyote people mod:





The creature graphics were mostly a slog because back when the graphics were first introduced there weren't any palettes for them so you had to repeat the graphics for each skin / fur color.  That got overhauled and improved with the new portraits, but the portraits are at least as complex as the creature graphics because of the variations and components.  Eyes, noses, etc have their own variations, some of which are based on things like narrow vs. wide heads.  Sadly, the palettes also didn't decrease the amount of code you need, since the color checks in the raws are basically the same and repeated ad nauseum.  I think my final portrait config file is like 20,000 lines long.

Anyway, I'm not really happy with the portraits, but here are some examples I ended up with:

Spoiler: Large Images (click to show/hide)

I need to get some screenshots from in game with clothes on, but haven't played much with the mod since I only got the portraits working a few days ago.  I'm not totally sold on the hair, but that's a question with anthropomorphic animals in general.  I also feel like there's something uncanny and off about the faces, but can't quite put my finger on it.  They're better than they were a few iterations ago at least, where I was mostly using modified human eyebrows, which made them all look angry.

They also arguably don't really look like coyotes so much as generic canines, but ah well.

[Eric Blank]

I honestly don't get the uncanny valley feeling most of the time, except with like gorillas and chimps and ancient hominids. But when people make anthros with proportions too similar to humans and just tack on identifying animal features, it does look dorky to me.

Sometimes with anthros its better to leave the head features more similar to the actual animal, though, or you do run into that, or otherwise just not try to impose human features on them too much. Like, when I was doing vutchnell and paklara, my line of thinking is, vutchnell aren't animalized humans, theyre convergent evolution of mongeese on human-like capabilities (intelligence, speech, grasping, and standing/walking upright specifically.) For paklara I went like a necromancer experiment; a mage turned some sea birds into a humanoid intelligent creature by warping the bird into the desired shape, not putting feathers on a human. So "what would these organisms look like if they developed along the same lines as humans?" instead of "what if human with animal features?"

Like for yours, maybe if their cranium were more similar to a coyotes; a slightly vertically squashed spheroid as opposed to the human tall and compressed front/back and left/right. Find a canine skull and coyote face photograph to use as reference. A thicker neck, attached to the jaw. If you increased its size from that of a coyote to a human, you'd be almost there already, you could just expand it a couple pixels. You could just fudge it and say their brains got denser/wrinklier instead of expanding, keep an almost- or entirely-natural shape for the species.

For the hair issue, I instead of going for humans' head and facial hair chose to go with manes, like those of lions or hyenas. The longer fur extends from just behind the brow all the way down the neck to between the shoulders for both sexes, males' "facial hair" is more the mane encroaching on the back and underside of the jaws. The color though should match the fur color and pattern of the body. If you want it to be different, it would feel less forced if it was a fur pattern that included the mane and surrounding features, such as backs of the ears, brow and nose bridge, down the spine etc, and wasn't just relegated to the "hair". Coyotes sometimes have significant fluff around their neck already, it could just be longer on the top and back of the skull.

I havent even started on my portraits tbh, im still working on the damn layered sprites one species at a time. Just did the gnolls (mostly) last month, never did vutchnell hair styles. I gave myself a huge menagerie of weirdos to draw and no skill or energy to actually do that...

[Telgin]

Yeah, I think you pretty much hit the nail on the head with the issues.  Part of that was laziness on my part since I started with the human portraits and modified them to save time instead of starting from scratch.  I was hoping that would let me reuse the existing clothing sprites with few to no modifications, which it did.  I was also hoping it would let me reuse the sprites for things like eyebrows without much modification, but that didn't work out.  The hair sprites didn't work at all for a lot of reasons, so I had to end up redoing them anyway.

I also had a bit of an issue with the necks.  If you take the human sprite and modify it like I did by adding a muzzle you can't even see the neck, which looked really weird.  I ended up moving the torso down a few pixels and the rest of the head up a few pixels.  That meant I had to move all of the clothing sprites and lose some details, and maybe made the necks too thin.

You're also right about manes instead of hair.  I tried to halfway do that as a compromise since I left the default hair styling tag on them and that meant that there needed to be enough mane / hair to braid or put in a ponytail.  It makes sense that if they had enough mane to braid that some would.  I also left a bit of fluff on the backs of the heads for the "clean shaven" style.  Anyway, I'm having second thoughts about it and may remove the styling tag, which is a bit annoying because of the clean shaven style anyway, which feels weird for a species like this.  I could just put in a few variations on the sprite to give them different mane styles I guess.

And colors are another sticking point.  Real coyotes are mostly sandy with darker backs, but it's hard to work the dark part of the coat pattern into the sprite due to the angle.  I think you might be right about doing that as part of the mane though.  Anyway, I may need to reduce the intensity of the color variations a bit.  Not sure if real coyotes can be melanistic or have white coats, but probably not.  Also unsure about the manes having a different color from the rest of the fur.  It again helps with variety but does stand out too much.

I actually wonder what happens if you change the available color tags on a creature if you have a save already.  I do have a game I've been running for about a game year that I'd like to keep but if I make major overhauls to the creature definition I might have to regenerate the world.

[Eric Blank]

A quick Google search shows coyotes can indeed by melanistic or dark colored or albino/white or very light blond coloration, so that's not off the table. Honestly I think nearly any animal can be melanistic or albino.

Also, graphics tokens do appear to update with mod updates just fine from what ive seen. I haven't had to reset a world for that alone.



Ive updated all my currently "finished" race graphics to use palettes for armor and weapons! It took forever!

[Eric Blank]

Havent actually made a mod here, but a question on FOTF inspired me to follow my dream of making combined animal people species. So I made a spreadsheet and split the animal people into groups based on common relations. There are many that include only two species, and many more that are entirely alone, because they don't have any close relatives below the order level. Hyenas for instance are feliformes, but do not belong to felidae. They are alone in DF. Hedgehog men are alone, unless you go up to the order level and include molemarians as a "animal person." None of their other relatives are included in DF. Pangolins straight up don't have any living relatives unless you go all the way back to "not ungulates", including all carnivorans. The largest grouping of species of mammals in Df are carnivorans, but I wanted to divide them up into families. Skunks are no longer considered mustelids. They're a relative of mustelids, but are not themselves mustelids. Red pandas are also divided off into their own family.

Bird taxonomy is fucked up. The highest classification I can create where there are no individual species is dividing it between "neognathes" and "paleognathes." There are four species of Paleognathes in Df, which is about right as theyre about 5% of all extant avians. Trying to divide Neognathes was a mistake.

Im not touching insects. And I threw all the molluscs, arachnids and amphibians into the same groups. I discovered cave fish men are the only fish people. And what the fuck am I supposed to do with sponge men? They're all alone! I couldn't even include coral men if coral men existed because those are completely different animals! Include different species of sponges??

I could pad out many groups with extinct relatives, both recent and ancestral. But if I'm going to do that for animal people but not add the animals theyre based on (which would be completely conjectural because, of course, they are extinct.) then that's just weird.

Its a pipe dream.



I suppose I could just say fuck it and smoosh them all into like six or eight creature definitions based solely on which DF body detail plan they follow. It IS possible for different castes of a creature to use different BDPs, I can do that and just smoosh them all into one single species. But I don't know what that's going to do with their behavioral tags as well. And then a squid woman could give birth to a muskox boy and a black mamba girl and a leech girl. Because castes are selected at random.

[Telgin]

The way castes work and an inability to have multi-species entities is the reason I've never tried it myself, despite having very similar ideas.  There's no satisfying way to play DF as a general anthro fort right now.  I think the best you could do is make separate animal person entities for each species, but that would probably cause lots of issues in worldgen.

I did consider trying to write a DFHack plugin once to fix it by only allowing pregnancies if the parents were the same species and to fix the baby's caste to match theirs, but it would be very ugly code.  Since males and females of the same species would be different caste IDs, it's hard to do unless there's a lookup table hardcoded or something, and you'd still end up with migrants showing up with all kinds of messed up family trees of unrelated species.  I'm also not sure that there's any way for DFHack to trigger code on a birth, so it would probably have to just run through every unit periodically and change their caste if it didn't match the parents or something.

Technically doable, but I'm not in the mood to learn enough to do it.

Then you'd run into the inevitable problem of most citizens marrying unrelated species and producing no children, though some might consider that a bonus.  That's a problem a lot of anthro settings have though, and few address the consequences that should have on populations.