Thursday, 6 June 2013

Building a dinosaur: the finished Triceratops sculpt

It's been a while coming, but the Triceratops sculpt is finally finished. At this advanced stage the fine detailing is added to the mesh as well as any adjustments that may be needed to  morphology.

Central to this stage of the process is gathering reference; in fact this task runs through every stage during the creation process. This involves looking for papers on integument, lesions on the skin as well as gathering references on colour and skin texture. Using extant phyogenetic bracketing (EPB) we can infer certain things about how a Triceratops might have looked; for instance we know birds have very bright colouration and the males often have structures designed for display, so we can consider whether ceratopsians used their head ornamentation for only for display or perhaps defence or species recognition too. We have fossil impressions of chasmosaurine skin so we have a reasonable idea how parts of the skin of Triceratops looked in terms of topology and we can look at extant animals and observe their colouration and how it relates to their ecology. One tool I've started using a fair amount to gather reference is Pinterest, which allows me to 'pin' images from the web to virtual boards, and works both in the browser on my desktop and in a standalone app on the iPad and iPhone. Using this tool it's possible to gather good reference quickly and easily, and store it all at one location that is accessible online.

Although we're not actually texturing this model at this stage, to my mind it's essential to consider integument when detailing a sculpt as it may well affect some of the decisions you make when adding detail will be reflected if you decide to texture a model.

I did all the detailing in zBrush, as this allows the artist to work at high resolution and at speed. I tend to create my own alphas as this gives me more control, but the built-in alphas are usable and a Google search will turn up many sources of free alphas to download. These can be resized to give different sized brushes and for the scales of Triceratops I created this alpha:

Ceratopsian scales. Oh yes.
I traced some scales from a reference image of ornithischian skin and blurred the image slightly to avoid jagged edges on the geometry. As scale size is not uniform over the body on either extant or extinct animals the ability to adjust the draw size gives flexibility when adding scales.

Next I added wrinkles, scars and detail to the horns and claws. I turn off symmetry on whatever brush I'm using for the final part of this to ensure variation which adds visual interest. I roughen up the skin here and there, add a few lumpy bits and other marks as this Triceratops is an older animal that bears the scars from it's years living in the Late Cretaceous world.

Detail of the hoary head.

Finally, I seek the comments of palaeontologists and other illustrators and make any suggested amendments. This is an important stage; it's very easy to get too close to your work sometimes, and you can end up making errors that are, when pointed out, embarrassingly obvious. Moreover, getting someone with experience to critique your work means you will have the most subtle of subtle changes pointed out to you, and this is a great opportunity to learn. It will help you immensely with the interpretation of the reference material used when creating artwork in the future.

So that's it! I'm now looking to get a 3D print made of the model, and I will texture and pose it of course, but that's it for the time being. I will hopefully get around to animating it at some point too.

Big thanks to all those who have commented during the process and especially to Scott Hartman for his invaluable input.


  1. Hi, Stu!

    No time for super-details, but I do notice two things that might deserve changing. The first is minor, but tasteful: The nostril should typically be in the lower, front part of the narial fossa, and will be close to the margin of the jaw and the edge of the upper beak. The way it is now, the nostril's too high and too far back, but this is just a taste thing: You can claim some independence on this with reason.

    The second thing is the position of the "cheek" tissues: Unlike ankylosaurs or stegosaurs, the predentary bone has a long process that extends over the anterior jaw bone which forms a broad, bevelled platform. This platform extends further back from where you have the beak terminate, and forms a sort of outwardly-facing facet, and it is unlikely the "cheek" will cover this as it is somewhat broader than the rest of the jaw except near the back. This also means the upper jaw forms its own platform that extends beneath most of the narial fossa to occlude the mandibular one, and thus the "cheek" should be pulled at least 1/5 further back from what you show it as. You've got the upper position okay it looks like, but the jaw should form a more distinct platform with the lower, rather than look like the lower jaw is a normal "birdy" beak, like a duck's.

    That all said, this is nice work!

  2. Hi Jamie,

    Thanks for the comments, very much appreciated. I thought long and hard about the nostril position, but upon reflection I think you're correct and will move it forward.

    I went back to check my reference and found a photo I'd taken at the Black Hills Institute of a Triceratops skull which shows the platform you mention. I'd missed this and will correct.

    Thanks again!