Someone asked for this on Whitechapel recently, so I thought it would be good to post here as well.
For those that haven’t seen the initial DDH post, you can find it here. Direct links to the full-sized final images for the Ghoul and Troll are directly embedded in their respective headers.
All of the pieces in that batch were done on two sets of heavy time-constrained deadlines. They were also all done gratis. The first deadline was on a day that I had to leave for a convention, and as such, the completed versions of the 4 pieces I finished were rushed, and I wasn’t happy with them. Especially the initial versions of the first 2 pieces in that post.
I’m going to tackle the drone piece first, as that was an odd circumstance. I was given the description, and a reference picture provided by the author of that particular article, stating that was what it should look like. The drone wasn’t gotten to the first time around, as I was pushed for time and they had said default image. However, when I found out that the publication had been pushed back, due to various technical difficulties, that was what I tackled first. I took the original image the gave me, blew it up, flipped it, and distorted it in Photoshop (henceforth referenced as ‘PS’). Then, I moved it to ArtRage (henceforth referenced as ‘AR’), and started a paintover of it, like one would do with a matte painting. I created areas/parts, removed areas/parts, and worked on distancing it from the appearance provided by the initial image, while still keeping the basic shape and functionality qualities. One of the things I love about using AR is that you can imply complexity and detail very quickly just with a few bits of contrasting color and knowing how to make the palette knife tool work for you. After that was done, I drew in some quick line details, saved it, and ported it back to PS. At this point, I started playing with layers and filters, pulling in a couple of photo textures/overlays in the process, as well as some photo-manip to create both the turbine caps and the open air “turbine” in the middle. One more port into AR to create the sensory array, minigun, and their mutual cover plate. Then, a final run through in PS, texturing those items, filtering, layer alteration, and such.
I’ll make sure I had more time with the next one, and render it up from scratch between PS, AR, Google Sketchup, and/or Alchemy. Maybe Manga Studio if I get more comfortable with it. It would still get finished out in PS though. Not mangle the perspective. Oh, I almost forgot– I would have done the removing and modification of areas/parts in PS instead of AR, as it would have been much cleaner.
Before and after versions:
The one on the left was the initial version I did. While I liked it, I wanted to overhaul both the background (which was a 75% photo-manip shortcut that I took due to the time constraint) and the ghoul figure itself.
In the case of the background, I went back to the layer with the base B&W photo, and started working it over from scratch. Ported it into AR for creating a subtly textured color underpainting, then ported back to PS. I removed some areas, modified others, and then started playing with a mix of varied layers, filters, and opacities. Then I took the layer that I created the background shadow and lighting in from the initial version, and reworked it a bit as well, helping push/pull certain aspects of the background. At this point, while still founded in the manipulation of said photo, it was at a point where I was more satisfied with it.
In the case of the ghoul itself, I wanted to flesh it out, give it depth. So, first I ported the image back to AR to focus on it, specifically the color and painted texture under the linework. Then I started playing with varied opacity overlay and multiply layers to pull out more depth. Added more detail to the face. Then I ported it back to PS, and started doing more filtered layer work at varied opacities. In addition to some minor skin texture that I pulled from random photos, I also brought in some very low opacity overlays, distorted and warped for individual parts of it’s anatomy, to bring the depth out more. Finally, I integrated the lighting and shadowing on the figure via a few different means.
(wherein I pause for a moment to tell people reading this that I’m totally distracted by the smell of freshly baked bread wafting through the house’s heating vents… *drool*)
Before and after versions:
I actually had a WIP (work-in-progress) image of this saved, so I thought I’d include that as well on the left, with the initial version in the middle. Like the other pieces, I wanted to overhaul this one as well.
The background is relatively bland, and like the rest of the image backgrounds in that batch, I took a shortcut for the time constraints. In this case, I rendered a quick under-layer of color in AR, and did a mix of altered photo textures, overlays and layer filtration at varied opacities above it. Aside from determining lighting, it was of menial importance compared to the troll, so I didn’t spend a whole lot of time on it.
As for the troll itself… The WIP image (left picture) is strictly AR. At that point, I had done the initial sketchwork, and created a relatively basic underlayer of color to go with it. After that, I built on things a bit, giving it a little more depth while fleshing it out with other colors and paint textures in AR. These are still mostly evident in the first version of the piece (middle picture), where I’d imported it to PS for minor filtration and layer work. Upon revising, I took it back into AR and started detailing out the colors a bit more. I did more of the whole photo texture/overlay, filtration, layer opacitiy thing. I changed some aspects of the clothing. Then I really cleaned the clothing up, adding some details, mixed from photo-snippet overlays, painted details, and custom brush textures.
Therein gives a detailed version of the creation process of those images. These processes apply to the other two images in the batch as well.