When you use internal fluorescence you typically need to use a lot of protein 10-20 uM. However with sypro you can get very nice unfolding curves, usually with very minimal purturbation of intrinsic unfolding Tm (based on Trp fluorescence), at concentrations in the 1 - 2 uM range, and I have gotten down to as low as 250 nM for a few proteins. Depends on the amount of hydrophobic surface.

These are still boring experiments, I typically multiplex and do things in 384 well plate using a Q-PCR machine. So you can generate a whole bunch of data all at once.

Another fun little path is to use the 'proteoplex' software to quantitate and optimize unfolding conditions, really speeds up assay design. (http://www.proteoplex.de/)