Continuing ignition timing debate from the 416 thread.

If you load it steady state at one rpm (peak torque and peak hp for example) and swing the timing around while watching the observed output you could develop the curve much more effectively or maybe efficiently is the correct word.
To do this fuel would have to be pretty close to where its going to end up. If its significantly leaner or richer then it will show best timing for that condition. Isn't this why we try to get a baseline timing established first? Then get best fuel mix for it, then try tweaking the timing a bit.