> It depends on the circuit. Some saw->tri waveshaper circuits rely on > the crappy slew rate of the 741 to mask the switching glitch. A 20 pF (or so) cap across the feedback loop would be the preferred method, I'd think.