Lesson 01 — Introduction/Softimage overview

Really looking forward to the next Digital Skills unit — the amount of work is vaguely terrifying (how are all those files going to fit onto one data disc?!) but it's looking to be a good one. I've never thought that 3D was really my thing but I wonder how much of that has had to do with lack of proper tuition — now that we're actually covering it properly I can see myself quite easily growing to love it.

Today's lesson provided a pretty basic overview of the software and some of the ___ navigation and transformation tools. I'm really glad I played with the program beforehand; I don't think I would have remembered everything otherwise! Softimage is such an expansive piece of software — seeing the interface for the very first time sent me screaming into a corner…

I've previously covered the interface during my own exploration of the program so I won't repeat myself here, but just to recap in summary:

• Main Softimage display is split into four 'viewports' which can be customized, expanded or switched with different views to best suit your workflow. By default, these are set to User/camera, a top, front and right view. 
• You can dock any menu or panel within the four viewports for easy access.
• 'User' viewport setting can be used to move around the environment without moving the physical camera. Very useful for checking/adjusting its position or making adjustments without botching it up!
• Use the 'Headlight' option (found under the small menu in the top right of any viewport window) to illuminate your model from all angles.
• The Explorer panel shows all detail about everything within your scene — similar to the layers panel in Photoshop. There are a few different explorers available: the Scene Explorer displays a list of all cameras, lights, geometry, objects etc. You can view a hierarchy of any materials or textures applied to a specific object.
• The Selection Explorer is identical to the Scene Explorer except that it only displays information for the currently selected object.
• Shortcuts for main transform tools are X for scale, C for rotate, V for translate. Press or hold space bar to flip to object selection mode. S is for navigation tools, allowing you to pan in and out of (middle mouse button/scroll wheel), orbit around (right mouse button) or track across (left mouse button) the scene.
• Small panel at bottom of Softimage window will remind you of the mouse functions/shortcuts for currently selected tool.

We were set a basic task to assemble Mr. and Mrs. Potatohead using the tools we'd been shown and produce a simple 4-panel comic strip using the posed figures.

(I'd just like to point out that for the majority of these screenshots I seem to be in Render mode — I'd been playing about with materials earlier today and forgot to switch it back! I didn't even notice, though, and was able to continue working without any difficulty/referring back to the menus, so I guess that's something of a sign that I'm getting better with the program...?)

My initial attempts to construct Mr Potatohead were clunky and awkward — I kept clicking the wrong buttons and forgetting which axis was which, but as I spent more time moving around the workspace I found myself adapting very quickly to the keyboard shortcuts and interface. By the time it came to Mrs. P I can say, with some measured confidence, that I was working fairly fluently within the program. I didn't struggle so much to 'read' the viewports and lining objects up became much easier.

I found that expanding the 'user' viewport on the right hand side to be most useful, leaving the other two for front and top views. This gave me the freedom of navigating within a fully 3D workspace, allowing me to freely pan and orbit around the scene and place objects pretty much however I liked. I was then able to use the other viewports to ensure everything was lined up correctly. The 2D top/front viewports are ideal for precise position, scale or rotation tweaking.

The only trouble with screenshots is that I only tend to remember to take them after I've finished doing something, which makes it look as if I work quite statically. The screenshots above would imply that I was working without moving the camera very much at all — rest assured that the big viewport was being swung around wildly, inspecting every orifice of my big plastic potato from all conceivable angles to ensure everything was plugged together properly! (I've seen enough side-views of his nose to last me a lifetime)

The caps of the eyeballs/eyelids were surprisingly tricky — once the eyeballs had been rotated into position the eyelids had to be very precisely fitted into position. If they were off in any way, the eyeball would protrude through and look a bit strange.

I made full use of all three viewports in this situation — the top and front were used to drop the eyelid into position and roughly line up the rotation, with the main 'camera' port used to fine-tune and ensure the eyelids were slotted over the eyeball correctly with no awkward protrusions. I found that choosing to display the wireframe over the shaded model (activated from the same menu as the headlight option) was amazingly useful at this point. I was able to see, at a glance, whether the meshes were aligned correctly and that the angles of the eyelids and the eyeball were matching up.

After completing the basic construction of the figures, I then had to remap my laptop keyboard to allow me to increase the mesh's subdivisions — that's tech-talk for "make it pretty!"

Once the model has been smoothed out you can really see the difference. The size and position of the pieces will have changed slightly, so I spent a little bit of time tweaking the position of the limbs to make sure everything was in place again.

(this is terrifying and I apologise)

With my models put together and smoothed out, it's now time to pose them and start on that comic!