Continuing with the Verdi work, a couple more notes about the process:
Divide and conquer. One of the strongest advantages of working in passes, is that it allows you to reduce complexity by focusing different passes on different aspects of the work (this idea is explained in more detail in "The Art of Passes" chapter of the book). In the Verdi example, pass 4 was a "profile pass": I constrained myself to only scrutinizing the profile of the sculpture. Similarly, Pass 5 was a "front pass".
Point of no return. Often the work presents a strategic challenge in the form of a "point of no return", which dictates that we deviate from the "pure" passes approach. In the Verdi example, the way the Sculptris software works created such a challenge: you can work on the sculpture in symmetry mode (everything you do on one side is mirrored simultaneously on the other side), but once you break the symmetry, you can't go back to that mode without losing details. Ideally I would have liked to refine the eyes a bit later in the process, while the general shape of the hair would have been defined much earlier; but because the shape of the hair is asymmetrical, that would have meant doing a lot of work on the eyes twice, probably with a lot of symmetry and proportions problems. This called for a methodological change. Here is the way to deal with a "point of not return":
- Recognize it as soon as possible.
- Define the specific aspects of the work that need to be perfected before you cross the "point of no return". In the Verdi case it was the facial features, and most importantly the eyes, that I wanted to get right.
- Work in passes towards that milestone, then work in passes towards the final vision (or the next milestone, if there's another "point of no return" along the way).