I've encountered a strange problem with V10. I can't quite nail it down, but the
Solve seems to be unable to solve equations that it could in previous versions. Here is one I'm almost positive I've done before.
Solve[q == l^a ((l w b)/(r a))^b, l]
For those who are wondering this is one of the steps in solving for a cost function using a Cobb–Douglas production function. It can obviously be solved, all Mathematica has to do is group the exponents.
Is there a change in the solve assumptions, options? I've also tried being explicit about reals and allowing inverse functions.