upon giving the original question some thought and research, perhaps one should minimize the amount of glass used, ie as little as necessary, which might be a different amount for the face or back.
Yes! This is why I asked about using different lam thickness a while back. The back should need less glass to carry the loads than the belly and you always want to minimize the limb weight. Since glass is approximately twice the weight of wood, using the smallest amount of it possible seems like a very good idea.
One other thing to consider (Alan Case has mentioned this) is the strains on the core wood. Thicker glass lowers the strain on the core, while thinner glass increases it. So going too thin on the glass can cause problems with set.
Like almost every real world design problem, there are a bunch of competing factors to balance off and a careful compromise is required to maximize performance with longevity.
Brads suggestion is a good one, but it only addresses a small segment of the question.
I would like someone to explain what they expect that experimental bow to demonstrate? Unless it is heavily strained to near failure I wouldn't expect it to shoot any differently either way. I'm willing to make one but it will be a while before I am set up to do that.
Do we really need a composite with 120,000psi compressive strength when a couple of classic woods like Osage and Yew, proven over thousands of years, come in at around 1/10 of that strength?
Do we NEED it? Of course not. Yew, osage, ipe and the various junipers are all great bow woods that perform extremely well. But glass bows can be strained to significantly higher levels than any wood can be. This allows different limb profiles and a bit higher performance than wood bows with greater consistency and reliability and less tillering skill required to get there.
Mark