Electronic shutter: advantages and disadvantages

Jacques Cornell wrote:.

OK, all this talk and theory and not knowing first-hand was driving me crazy, so I just now did a test, shooting a white wall under fluorescent light with my GX7. Theory says that readout speed is twice line AC frequency divided by number of bands. AC frequency here is 50HZ, and I get 7 bands, so readout speed is 1/15, same as others have reported for GX7. Starting at 1/250 and moving down in 1/3 increments:

1/250 - 1/160: obvious banding

1/125: mild banding

1/100: none

1/80 & 1/60: mild banding

1/50: none

1/40 & 1/30: mild banding

1/25-1s: none

The light from a fluorescent bulb takes the form of a rectified sine wave as shown in the red line below.

The period is 1/2*AC or in your case 1/100 of a second. The easiest way to visualize the electronic shutter is to imagine a slit moving left to right over the graph above with a width that corresponds to the shutter speed. At 1/100 the slit would be be pi or 2 divisions on the graph. No matter where you position that 1/100 slit it always captures the same amount of light, 1 full cycle. At 1/50 the slit would be twice as large and capture 2 full cycles resulting in no banding. At 1/33 it would capture 3 cycles, 1/25 would capture 4, etc.

But at 1/200 the slit would only be pi/2 or one division on the graph. If you position that slit over the the peaks on the waveform it will capture a lot more light than if you position it over one of the zero crossings and you get banding.

If you use the mechanical shutter with a fast shutter speed you will get inconsistent exposures under florescent lighting depending on whether the picture is taken at one of the peaks or a zero crossing.  It can be as much as a stop or two variation but since it is even across the frame it is much easier to deal with than banding.