Fred McKenzie wrote:

The beauty of the 5/8 (0.625) wavelength groundplane antenna, is that its
feedpoint impedance has a resistive component of 50 Ohms, and can easily
be matched with a small inductor. Some such antennas have been
constructed by winding the coil using the base of the antenna rod. It is
my recollection from an antennas class 35 years ago, that the best signal
towards the horizon comes from a 0.58 wavelength groundplane antenna, not
0.625 wavelength. I can see how 5/8 and 0.58 might be confused, and many
may consider the difference trivial.

Maximum gain at the horizon actually occurs at around 0.625 wavelength.
At lengths between 0.5 and 0.625, a higher angle lobe appears, but it
doesn't divert a significant amount of the total power away from the
main lobe until the antenna gets longer than 0.625 wavelength.

However, those gain figures are for a vertical antenna mounted on a
perfect, infinite, ground plane. Actual performance of 0.5 or 0.625
wavelength antennas compared to quarter wave ones varies a great deal in
most typical situations. And any kind of WLAN device I've seen deviates
in major ways from the ideal case. Gain of a longer antenna is as likely
to be due to placebo effect as to physics.

One big advantage of a 0.5 wavelength antenna is its high feedpoint
impedance. This makes it much more independent of the other half of the
antenna -- what amateurs like to call "ground", but is often the top of
a car, a person holding an HT, a router box, etc. The advantage isn't
shared by the 0.625 wavelength radiator.

Roy Lewallen, W7EL