Thanks a lot for all who replied to my question on and off the list. It
seems that microscope and laser manufacturers now work closely together to
optimize their systems. I came to the conclusion that the remark: "short
pulses will broaden much more"  is not relevant for systems with dispersion
compensation like the current Olympus FV1000 or the MaiTai Deepsee or the
new Coherent vision. These systems have enough compensation to achieve puls
widths out of the objective which are very close to what comes out of the
laser. But the initial difference (140 fs for the Coherent and 100 fs for
the Spectra Physics MaiTai HP) remains. By detuning the dispersion
compensation i can generate longer pulses on the system i am working with
(MaiTai, Olympus FV1000 with Olympus pre-chirp), this indeed causes a loss
in fluorescence intensity on a test sample. From this i conclude that the
shorter pulses of the MaiTai can be an advantage.