Ron: But heavier DS spokes won't alter/improve the NDS tension... which is the main advantage of triplet lacing... no?
For radial loading, the bottom spokes will lose tension and the rim will distort until the aggregate change in spoke tension matches the load. A 1.8mm spoke (for instance) has 44% greater stiffness than a 1.5mm spoke. That means that for the same deflection it will take 44% more load. If you apply a radial load, the stiffer DS spokes will take a majority of the load... compared to having the same spokes on both sides, where the load will be ~equally shared... and the rim deflection and detensioning of the NDS spokes will be less.
For purely lateral loading, using heavier DS spokes doesn't provide a lot of benefit since the stiffness goes up with the square of the bracing angle and the NDS spokes provide most of that. But lateral loading is always way less than radial, and it is usually the combination of the two that results in NDS spokes going slack.
It is hard to say exactly how much it helps unless you make a FEM and try out some scenarios. The rim will also bend sideways when a radial load is applied (unless the difference in spoke stiffness matches the difference in bracing angle), so the rim stiffness will play a part. You won't be able to take a 44% higher load for sure... maybe 20%?
Any engineering students interested in making some models? Shouldn't be very difficult if you have access to good software.