< PREV NEXT >

Sub-Div Modelling - Creating a 4D Modelling Workspace

I came up with this trick recently and it's just sooo dumbass simple and useful that I thought I'd pass it on. Yeah. I know. It isn't strictly what you'd call a sub-div modelling technique as it works equally well for other types of modelling too. But since I find it so exceptionally useful, I thought I'd pass it on ...

When modelling characters, vehicles, objects, etc. very often you'll start out with reference sketches or photos. A common technique is to scan those images to the PC, import them into your 3d software and then display them in the background or onto uv mapped boxes using a material set to display in the viewport. Then you can start tracing splines or modelling geometry right over the top of these reference images.

   

That's all fine and dandy but after a while having these reference images in the scene becomes a bit useless because you just can't see them anymore - all your geometry is in the way. Sure you can a) make the object see-through, b) drop a wireframe material onto the object, or c) hide bunches of polygons at a time but their usefulness has dropped pretty substantially.

So lets try something else.

Just get your uv mapped boxes and *animate* them - moving from perhaps the middle point of your mesh to its extreme edge over a large number of frames (I use anything upwards of 600 frames - it gives me lots of control over positioning).

Now if you scrub the time-slider back and forth in the time bar you can move the reference image to any point in your geometry (as shown in the animated example here).

This means you can move your reference box to points where it doesn't get eclipsed behind geometry anymore and you can tweak the mesh perfectly to align with your reference material.

If you have a graphics card that supports opacity mapping in the viewport you can also set the opacity of your material (to say 50 or 60%), and you can now see through the reference and carry on modelling your mesh behind it!

But one of the biggest advantages (for Maxxers anyway), is that you can scrub the time-slider and adjust your reference box at the same time as you're working with the sub-objects (verts / edges / etc). You don't have to release your selection on that sub-object just to tweak the positioning of your reference. And you can just toggle meshsmooth on and off and see what your geometry looks like. That's a BIG time saver!!!

Since I first started using this technique I've come to depend on it more and more - it's just so versatile. I've especially found it useful when doing characters as it really helps me visualize things such as what the subtle contours of a characters face are doing. So try it, and see if it helps you!

(photo used in example downloaded from the superb site http://www.fineart.sk and (c) Ross Peplow)

 
< PREV NEXT >