I was in two minds about doing another arachnid so soon after the harvestman, but I thought it would be nice to work on my favourite spider. It’s a species of small “money spider” that is known for the strange head ornaments in males when they reach adulthood. It is known colloquially as the periscope spider, but the turret is not to aid vision (it’s actually likely this is involved in mating, where it produces a nuptial fluid that the female eats during copulation - these spiders mate face on, and it seems likely the female fangs sit around the tip of the turret). In life these spiders are a distinctive orange/red colour and move in a distinctive manner. I used to rear these with aims of doing research, but the project never got off the ground. They are quite tricky to rear as they only eat other spiders.
Unlike the harvestman I worked on, I actually have a couple of specimens in the freezer from my old DNA work, and there are plenty of reference photos to work from. I also have some scanning electron microscope images of this species. There’s a lot of variation in the turret shapes so which does make it a little easier.
I cracked out my microscope and took some reference photos. Not amazing given the small size of the spider (about 3mm, and my crap mobile phone), but enough to help get major shapes down. I am uploading these here for interest mainly (and my convenience).
My progress so far is rather limited, for some reason yesterday I really struggled to get the coxae joints and maxillae down. These are probably the hardest parts so I will try again today.
Blocking in main shapes. I’ll probably retopo this at some point as it’s too difficult figuring this out as I go. As long as I don’t add edge loops to the legs I can just duplicate them and start work backwards.
Wasn’t happy with the overall proportions and attachment of the legs, so I went back and redid these. This also fixed the topology, so now we are back to quads again and no (significant) superfluous edge loops.
Currently at 30K verts (subdivision level 1), although I need to create the pedipalps next.
Remade the maxillae too, I think I finally have the shape down. Hardest thing about making these arachnids is not the actual modelling as such, it’s getting your head around what the shape actually is from 2D images. The maxillae sit behind the chelicera (fangs) and are moveable pads basically that help in feeding. The pedipalps also come off these - in general the shape is a bit like a bean, but it curves around the chelicera and also over the labium.
Modelling the pedipalps. Probably the hardest thing I’ve ever tried to model. I’ve spent 3 days just trying to get my head around the major shapes of this, constantly taking photos from different angles and trying to understand how it fits together. I’m not sure if that actually helped as this thing seems to change shape depending on what angle you look at it from.
I think I am happy enough with this the overall shape for now (there are some inaccuracies, but this thing is hard enough to understand down a microscope let along modelling it by hand), but will need to add further detail and double check where the little sharp bits are, where creases are, weird membranes, etc. At least the mesh is manifold and could be articulated corresponding to where the pseudojoints are in the pedipalps. It should look convincing enough from a typical render distance showing the whole spider.
Female is now modelled, and both specimens rigged. Tried the new pose asset library for the first time as well. Next I will work on the textures and adding the hairs, spines, etc.
As I modelled both spiders to scale, I could not resist investigating if my ideas on how they mate were feasible (as in, would the limbs be long enough, etc.). It seems feasible at least - the male can easily reach the female epigyne with his palps with the female mouthparts on the turret.
I really do not like spiders. You are the only person who has ever had me take an active interest in looking at spiders. I really enjoy your progress and methodology, and your results are fascinating even to a arachnophobe like me. You should be very proud of yourself, normally when I see a spider I run the other direction
I’m pleased with how this project is progressing. It’s been a good learning experience - rigging, texturing, etc. If I can also convert some arachnophobes to be arachnophiles along the way then that’s even better
Not made a huge amount of progress since I had other projects to work on, but I am nearing completion.
Currently working on the rather tedious aspect of hairs, which need vertex groups set up individually. I also need to figure out how to have the trident hairs on the head arranged correctly. I also need to add to the armature to allow the feet claws to tilt slightly, as these do move in life so the spider doesn’t walk on tiptoes.
Once this is done, I can repeat the process on the female and start working on the scenes.
One scene I want to recreate is my lab rearing setup. I used small plastic vials (35x56mm) with plaster of paris in them and some moss. I’d like to create a small scene of a female with eggs.
I decided to rework these meshes as my understanding of retopo etc. has improved. So I’ve been working on the female - now down to 20K verts. I also made use of the new hair curves to redo the spines, which I think gives a more realistic look.
The material is also totally reworked. Instead of multiple material slots the shader is now running from a single set of 4K textures, with a couple of additional masks for SSS and transmission. This has been quite tricky as there are to my mind two variables involved, the thickness of the cuticle varies by location, but also the degree of colouration and degree of transparency. It’s still not quite right - some areas need to be more transparent, but when I alter things it tends to have knock on effects to the entire mesh. It’s also annoying how the scale of the mesh makes such a huge difference - it’s a shame there is not (?) any way to model in a relational way rather than defining actual distances. You can see this in the second render where I just scaled the model 10x smaller. Some areas are now too light, others look better.
Compared to a reference image, it’s closer than the old mesh but still work to be done.
I think the colours are about as good as I can get at the moment (it doesn’t help that the spiders themselves vary from more orange or red to yellow/brownish, and also vary depending on how you light them). If I change the cephalothorax integument to make it darker, the entire mesh changes colour.
I did a quick test render of a little scene. It was mostly to indicate the scale (most people have those clay balls in their plants). A shame the clay ball texture is a bit borked this zoomed in. The spider itself still sort of looks like a plastic toy however. I can’t seem to get it transparent enough without using the alpha channel. Another mask required it seems.
I think this is much closer. I had to use volume scatter and then totally redo the transmission settings, but it seems more like the reference image (obviously the reference is quite saturated and contrast-y so it won’t be an exact match).
I’m from a country where most of the fauna and sea-life or for that matter the very land itself, can be lethal too ones continued existence but I’ve a soft spot for our arachnid friends. Been a while since I’ve seen a Huntsman, also less commonly referred too as the Crab Spider, typically their sometimes unannounced confronting appearance inside the home, usually signals the onset of winter…tbh kind of miss them being around.
Guess like everywhere else globally their disappearance, is basically symptomatic of man made adverse effect, upon the natural world.
The male is now textured, pretty much (I might work on the palps, they seem a little flat along the embolus). Some tweaks here and there and I think they are ready to put into a proper scene for render. Working with transmission is annoying so I’ll be happy to do something different after this for a change of scene.
