Slow rise times and slow decay (slew rate) will make square waves into low mounds. Visualize that, with rise time for example, a component that can replicate the leading edge of a 'square wave' can replicate the leading edge of a cannon shot so replicating a square wave is good. However the converse is also true, a square wave which is perfectly vertical (or close) on the back edge, would rob the music of a more natural sounding wave. In fact IMHO the treating of the 'fall' time of the back edge is more important than the slope of the leading edge. It is sort of a finding of the proper degree of slew rate - too little and you have a muddy echo chamber type residual v too fast and the sound is too clinical and unmusical.
Hope that helps a bit.
Hope that helps a bit.