Quite a bit of talk about voltage decrease under low light conditions, but the real problem is the decrease in Isc. Consider the fact that shading could be thought of as localized reduction in irradiance. Most manufacturers publish charts indicating performance at various irradiance levels. I haven’t quite figured out if these are specific to a particular panel or panel series or are more generic in nature. They are virtually identical for either “12-volt” or “24-volt” panels.
Below are a couple of such charts. The one on the left shows performance at discrete light levels. The one on the right is a normalized continuous graph.

It can be seen the Voc drops somewhat with decreased irradiance, but Isc has a greater than one-to-one relationship with irradiance. A 50% drop in irradiance results in slightly more than a 50% drop in current resulting in a corresponding decrease in power. I don’t believe either PWM or MPPT can “make something from nothing”.
I also found a chart for a
Unisolar thin film “panel”.

It doesn’t appear much different from the above. It’s not that I disbelieve the claimed superior low light performance, but it would be nice to see some published data that verifies the assertion. Perhaps someone has some further info?