The making of "Once Upon A Claw"

Before things get all Sauropody around here, thought I'd throw up a quick post about one of my 4 pieces for the Anomalocarid gallery.

The piece in question is "Once Upon a Claw". This is the second piece from this year I'm quite proud of. The first being of course "Karoo Sunset", and another of my Anomalocaris pieces "The First Great Predator" has this becoming a trend. I'm not used to being happy with my pieces, so this success that ART Evolved has been motivating me towards is most welcome!

You can get a close up look at the details here (for some reason blogger tends to not allow zoom ins on pics I upload), and frankly I put so much work into them I insist you check them out.


There was no single trick to the making Once Upon a Claw, and in fact no tricks at all really. Just a lot of fine detail modelling and composing. The success of this piece is not any single brilliant element, but rather the combination of several alright ones (supplemented by a few substandard ones too). This being in the 3D modelling, their texturing, lighting, and placement.

This little slide show takes you through the 8 days of its creation.

Lastly I thought I'd introduce you to the population of the claw individually.

One of the first things I added were these "Brachiopods". Up close they are probably the weakest of my elements as they are actually clams, but the texture gives them a Brachiopod like look far away.

I could probably get away with calling them clams as primitive ones are known from the Cambrian, but realistically they were uncommon and rare. The early Paleozoic was the heyday of the Brachiopods so I'm sticking with them as my desired outcome. I just didn't do a great job on them! If I were ever hired to do a similar piece, I'd have taken the extra time to tweak these into proper Brachiopods (making a calm was easier!).

It was when I built this Hallucigenia that I actually started work on Once Upon a Claw. I'd been toying with the idea of a scene full of discard Anomalocaris claws, but had abandoned it when I had trouble coming up with a way to show a claw that'd been on the seafloor forever (it needed to be matted and disappearing under algae and other growth). I suddenly had the idea of instead showing a single claw and how all the other lifeforms of the Burgess Shale were assimilating it into a new habitat.

My Hallucigenia is quite functional, and I especially like the stylized spines on the back. I did neglect the hose like nose on the front end, but honestly that could be remedied in about 1 minute of modelling. So I don't consider it a big issue.

I then grabbed my old 2004 Raydream (3D program... precursor to Carrara that I use now) era Yohoia model to flesh out the population early in the pieces development. I reasoned if I could get an early sense of life around my claw then I'd start snowballing on the creativity. My thinking was correct, and I quickly started doing a lot of new modelling after grabbing this old model.

As the Yohoia wasn't going to be a central element in the piece itself, I opted to merely update its shaders and otherwise use it as was. From far away it looks just like it should. Up close though, it looks like an amateur's model (which it is). Making it another thing I'd put more time into if it with a "real" piece.

The giant predatory worm Ottoia was my first "modern" Burgess model, though to be fair I did build it last year (making it old skool compared to my modern efforts). Worms are pretty darned easy, and I won't be boasting much about it, other then the shaders. I'm quite pleased with getting segmentations without have to build them into the worm's actual model (which would have made posing it a nightmare!).

I needed something for the worm to eat, and so I had my choice of the dozen or so obscure tiny Arthropods from the Burgess Shale to choose from. I ended up picking Habelia simply because in Stephen Gould's amazing book Wonderful Life the drawing by Marianne Collins was such a fantastic reference for me to work from.

Despite it almost being a cameo model, it is one I'm most proud of. I not only paid attention to most of the details (apart from gills... which admittedly I'd need to add for a more legit piece) but I rigged the model for posing and possible animation. As a final touch I put a lot of effort into making sure the shader gave this guy camouflage on my sea floor.

Though the piece didn't need it to be complete, I HAD to build a 3D Leanchoilia as it is one of my all time favourite Burgess critters (here it is eating one of my others Yohoia!). So I figured why not put it in this piece. In the end it actually served an important purpose of framing the claw with life, and thus making use of all the space I had towards the top.

I'm split on this model. The arms and tentacles are a big achievement in model rigging. They are all rigged as one unified piece, and I can move the arm as a whole or just manipulate any of the 6 tentacles individually. However the back end of the body needs more work, and the shaders aren't great. So I may revisit him at some point.

Put all these critters together along with the sponges, borrows, algae, ground, and most important Anomalocaris claw models, shade these all, and finally light them we end up with Once Upon A Claw.

Making a Prehistoric Flood

By Craig

So the first thing I do when sitting down to do a composition is think to myself what is the story I want to tell today.

_

When I approached my first piece for the Ceratopsian Gallery, I had the idea of showing the moment that one of the great Ceratopsian fossils sites was created. The particular site I had in mind was in northern Alberta at a place called Pipestone Creek, where a massive bonebed containing the remains of hundreds of Pachyrhinosaurs were discovered in 1972 (and worked until quite recently). These animals had all clearly been killed by a flood, and as it was pretty much just them in the bonebed they probably were all alive together before being killed (aka a herd).

So the story I wanted to show was one of the herd encountering this natural obstacle. Though it would claim the lives of some who tried to cross it, others would make it across successfully.

Here is the final picture for reference.

The only way I could pull off recreating a flood in the timeframe I had at the time (having to set up ART Evolved along side making this piece) was to create it via a photo composite.*

For those of you who have never heard the term composite before, this means taking a photograph and adding and/or subtracting things from it. In this case I was going to not only being adding Dinosaurs, but also a flooding river.

Looking through my many lovely pictures of the New Zealand landscape (it is such a great place to sub in for a Mesozoic era Earth) I discovered this perfect riverbank.

There was only one problem. There was no river, that alone one in flood!

However in my photo arsenal I DID have pictures of the local aqua duct flooding from 07.

It was not ideal at all though. The scale was wrong, it was too close up a shot. Not to mention the angle was all wrong.*

Still beggars can't be choosers so...

After some realigning in Paint Shop (my cheap Corel knockoff of Photoshop) I was able to get the flood facing roughly the right direction. Then some creative cutting and pasting supplemented by cropping of the water in a separate layer from the riverbank I had a rough flooding river.

Now sadly I didn't save this initial phase (I wasn't thinking about making a "Making of" post at the time) so you'll have to take my word on what it looked like. The problem it suffered from (more so than this finished one) was that the river unrealistically covered up the trees at the bottom of the hill.

So I went in and cut and paste several of these (but not all) back on top (in the photoshop layer sense) of the water. It was still missing something. So I went and grabbed some random tree tops and added them out further into the water to add some small growth submerged. That looked better, but one thing was missing. So some cut and pasted random splashes from the flood photo were added as water spray along the leading plants.

Though not perfect, this flood wasn't too bad considering the elements I had to work with.

Next it was time to add the Dinosaurs.

Recently I went through a huge Centrosaurid construction binge on Carrara, my 3D program, in anticipation of a big event I'm doing over on The Tyrannosaur Chronicles (check for it in about 2-3 weeks). So I had several short frilled Ceratopsians to choose from, including the two best known from mass bonebeds related to floods. Centrosaurus proper and Pachyrhinosaurus. I choose Pachyrhinosaurus as it is among my top favourite dinos and there have already been tons of pictures of Centrosaurus in floods.

Admittedly I am not totally pleased with the baseline colouring I gave this Pachyrhino, but I did figure that out till I was headlong into putting the model into my flood!

So having a 3D Dinosaur model is only a part of the battle. Next came the long process of duplicating multiple animals from the original, tweaking and individualizing them, posing them, and finally repositioning these in relation to the picture.

This was a whole day endeavour. Some of my effort is obvious, and is easily seen. Such as my remodeling a young "monoclonius" version of a Pachyrhino (which looked like identical to other juvenile Centrosaurids when their young).

Some of my tweaks were not so clear. For example if you pay close attention to each animal they're all slightly different lengths and sizes, and their horns are differently sized and arranged. I think this isn't obvious as every animal is identical in colouring, as I didn't have time to make new shaders for each animal. I think if I had made slight changes to each of their colouration these horn and size differences would have been emphasised rather than hidden.

Regardless, here was the poses and positions I ended up with. However without more work the picture was far from realistic looking so far. Dinosaurs that were underwater were still completely visible. None of them were casting any shadows either, which is a key element for compositing to look real!

So next I constructed a "shadow catcher" set for the photo. A shadow catcher is pretty much what it's name implies, it is an object in my 3D program that "catches" the calculated shadows of objects I'm rendering as per the lighting settings in the scene. At the same time they don't block or cover up the background, merely project a shadow onto it. You can see my shadowcatcher objects in the 3D window screen shot I included before, they are the grey rectangles the Pachyrhinos are standing on or swimming through.

Though this may sound and look easy, but it was actually quite time consuming. You aren't building the set from scratch. Rather you're having to recreate and mimic the angles and placement of preset positions determined by the background photo, or as they call it in the biz the "plate".

With my shadowcatchers in place I was nearly there. However the Dinosaurs were lacking excitement and interaction with the water, and the herd up on top of the hill was floating above trees that were clearly supposed to be in front of them. (Shadowcatchers are insanely time consuming to build for trees and other none straight objects).

So I went in and photoshopped the forest back in front of the waiting Pachyrhinos, and added splashes around the swimming ones.

Giving me this as my end product.

It's not perfect, but it could have been a lot worse too.

My main regret is the lack of animals. I'd have like 3-5 times as many. Maybe when I have more time (yeah right! there's a dream commodity) I'll retry.

So the new story is the back or front end of the herd arrives at the river where the banks are too steep to descend safely, minus one gentle slope. The majority of the animals take this single safe path to the water. However on the sides we see inexperienced youngsters risking the steep hills with varying degrees of success.

I hope this "making of" has been useful to you.