But you do use it. When you do a half-press of the shutter release the lens is set to it's widest aperture. Listen to what others are saying. With a 'fast' lens you get more accurate and faster auto-focus and a brighter viewfinder, no matter what you set the aperture to on the camera.
WRT using anything wider than f4, my opinion is that the 'big hole' will introduce softness. Conversely anything over f8 (certainly f11) will introduce distortion due to diffraction effects. Most lenses are at their sharpest round about f5.6.
Hi Stuart,
I agree.
I find that with landscapes I rarely need to above f8, maybe f11, if I roughly calculate the hyperfocal distance. See
here for an explanation.
The problem with hyperfocal distance is that it, just like DOF, is based on the circle of confusion (COC) criterion. There are many different COC numbers being used on various internet sites, but most are
not based on a realistic output size assumption for critical work. I use an upper limit of 1.5x the sensel pitch as the maximum diameter of the diffraction pattern, and the COC, for the most critical work (enlargements). Narrower apertures will gain in DOF but at a lower level of microdetail, even in the plane of focus.
Paul van Walree (an authority on optical matters) suggests calculating a personal COC for his Windows DOF tool called
VWDOF, based on one's output needs, as follows:
C= (V) / (1000 x Q x Mp), where:
C is the COC diameter in millimetres,
V is the viewing distance of the output in centimetres,
Mp is the print magnification, i.e. the ratio of the output to the sensor dimensions,
Q is a quality parameter Q=1 is conventional, Q=2 is demanding, and Q=3 is critical.
His 'critical' setting calculates to something similar as my 1.5x sensel pitch rule of thumb when applied to large output sizes viewed from relatively close.
Cheers,
Bart
P.S. Paul van Walree also mentions the risk of sensor dust showing up at narrow apertures. It rarely is a big issue for me, but then I rarely use very narrow apertures ...