Digital Camera Timing Tests

Some friends and I were discussing digital cameras and the delay you see between pressing the shutter-release and when the picture is taken. One of them pointed me at Ten Seconds and I gave it a try. Here are the results I see with my digital cameras:

Olympus Camera timings
Setting D-520
full auto 1.3-1.4 0.8-1.2 just press the shutter when it's time to take the picture
pre-focus 0.3-0.4 0.3-0.4 "half-pressing" shutter to pre-calculate focus and exposure.
manual exposure n/a 0.3-0.4 setting exposure manually, and letting camera auto-focus.
all manual n/a 0.15-0.25 setting focus and exposure manually.

All pictures were taken at the camera's maximum zoom, with the computer monitor brightness set to the maximum, with no flash, everything else auto (similar to my normal shooting setup). The distance from camera to the monitor was about 1m (which is the closest that doesn't require "macro" setting)

The wide variation in times focusing with the C-3000 Zoom without pre-focus appears to be related to its autofocus field. Positioning the corner of the window in the center of the frame (so the camera has a strong horizontal and vertical line to focus on) cuts focus time by about 0.35 seconds. Without doing that, the focus hunts back and forth, and never actually settled on the correct distance. Also, all the images without the strong lines to focus on were out of focus.

Note that the D-520 is newer tech by about a year. It's auto-focus wasn't quite as fast as that of the C-3000, but it was much more reliable in timing, and produced consistent results without having to resort to tricks like putting the corner of the window in the center of the image (it uses a wider field for the auto-focus). But it's also a significantly cheaper camera without the manual features of the C-3000.

It looks like there's about 0.3 second of lag in the Olympus cameras that I can't get rid of if I let the camera do any thinking at all.

Setting the exposure manually and letting the camera auto-focus (using pre-focus) did not change the time from simply pre-focusing. Even pre-focusing is not as consistent as manual focusing. I suspect the reason for this is the changing image (as the numbers click by) which suggests to the camera that the scene is moving, so it still tries to do some amount of focus "fine-tuning" when the shutter release is actually pressed.

Using manual focus and manual exposure on the C-3000 (features not available on the D-520), I could get click-to-clunk speeds down to around 0.2 seconds pretty reliably, and using a little anticipation (which I do in action shots anyhow, even with fully-manual film cameras—all cameras have a little lag-time), I could get the timer between -0.05 and 0.05 in about 80% of my tries after a few minutes of practice. That's about as good as I am with my film camera, and suggests that most of the variability is due to me, rather than the camera.

But if I let the camera think at all, there's more variability in the click-to-clunk time, which makes it very hard to anticipate pictures. I'm still frustrated there, but I've also been frustrated with film auto-focus SLRs every time I've tried them, and I think it's a similar problem.

Copyright 2009, Dave Polaschek. Last updated on Mon, 15 Feb 2010 14:06:23.