Daylight focusing issues and solutions

Elvishefer

Something I ran into a long time ago:

With an SB900 and single point focus the ir assist light only comes on when using the center point.

Not sure about other focus modes.

The SB910 will activate on more points with single point af.

Ade

Pattern bars can be used by both active and passive AF systems, so it's not 100% conclusive either way.

Both Phase-detect AF and Contrast-detect AF are considered passive systems. Both of them actually rely on sufficient contrast, so both may have difficulties in very low contrast situations (such as strong backlit scenes we've been talking about).

An IR active AF is something completely different. Typically it would work in one of three ways:

1. By using "echo location". We're all familiar how bats echo-locate using ultrasound -- by measuring the round-trip time of an echo, we can determine distance. A similar measurement can be done using IR light instead of using sound. (Some very old cameras actually did use ultrasound to achieve focus).

2. By measuring "reflectance". We shine an object with (IR) light, and measure how much of the light is reflected back. Using inverse-square law, we can compute object's distance. (Assuming standard reflectance).

3. By using "triangulation". Here we shoot out two IR beams. Due to parallax, the reflected pattern from each beam will be slightly different. We measure this difference to determine the object's distance.

None of these methods depend on "contrast", so they should work in backlit situations**. Also, all of them are independent of the optical path. What is required is an IR transmitter and an IR receiver.

If there is a complete Active AF module in speed lights (having both transmitter and receiver), then in theory, we can focus the camera even with the lens cap on!! The speedlight shoots out an IR beam, measures distance using one of the three methods above, and returns the distance info to the camera.

That would be great, except it doesn't work. Put a lens cap on, let the speedlight AF light trigger, yet the camera fails to focus. So we can conclude that there isn't a complete Active AF module in speedlights. (I also looked at the wiring diagram for the SB-900 and there aren't enough wires from the three focus modules for an active system.)

We could still have an Active AF if the IR receiver is in-camera. However, the more I think about it, the more unlikely this scenario is to me.

Lastly, maybe despite what Mansurov says, we don't have an active system at all. All we have is an IR beam from the speedlight to assist the regular (passive) phase-detect sensor in the camera. It doesn't work via "echo location" or by measuring reflectance or by triangulation. It depends on contrast as before. It might not help in backlit situations.

**Note: true active IR can actually be fooled by IR hotspots, such as open flames, and the sun. So it might still fail in very strong backlight.

FlowtographyBerlin

Lastly, maybe despite what Mansurov says, we don't have an active system at all. All we have is an IR beam from the speedlight to assist the regular (passive) phase-detect sensor in the camera. It doesn't work via "echo location" or by measuring reflectance or by triangulation. It depends on contrast as before. It might not help in backlit situations.

Haven't read that article, but I'm still pretty confused that you guys assume Nikons to have active AF. How would any of your camera's fail-to-AF behavior in low-light and low-contrast situations make sense? It's one of the first things every noob learns when they get a DSLR: low contrast = low AF capability.

Plus, to my knowledge, active AF is in sum far inferior to passive systems, I doubt anyone would actually wanna use that if he or she had the choice.