Whether it’s a blessing or a curse says probably more about the owner than Apple, but I find that their non-production of documentation is a rather frustrating thing. See, I like to read the manual, forget everything it says, go on a voyage of discovery, then go back and refer to the manual when I have a specific problem.

I find the Apple missing manual phenomenon frustrating because I can sometimes own a product for two years and not know that it does something really funky. This post is about one such thing.

In the iOS Camera app, on devices with autofocus cameras (iPhone 4, 4S, iPad 3, perhaps some of the Touches) you can tap on the viewfinder screen to create an autofocus point. The camera refocuses the image on that point which is dead useful when the device seems unwilling/unable to get a focus lock on the image.

My discovery is this: The tap-to-autofocus also causes the camera to re-meter the scene, using a point-weighted average.

Take this shot for example. Yes, it’s presented as a grainy black & white, but it’s actually a composite of  two images. (see bigger here)

The reason for this is that the human eye is actually a pretty trick bit of kit. The difference between what we can differentiate in shadows and what we can differentiate in extreme lighting (our dynamic range) is massively greater than that of any camera. Film or digital, multi-thousand-pound-pro SLR or wretched phone camera.

For photographers, this is a bit of a nightmare. We set the camera up to show detail in the ground, and it just goes & gives up on the blue sky with its beautiful fluffy clouds, showing the whole lot as a white sheet. Or we set it to show the sky, and the ground features disappear in a shadowy murk.

The way to get around this is with compositing. Either in-camera with a graduated filter, or in Photoshop (other editors are available) with multiple shots blended together. Take one shot where the sky is great (murky ground) and one shot where the ground is great (white sky). Then put them together in Photoshop.

The shot above required such a technique. The only way to get the white of the snow, the malevolence of the clouds, and the ground detail (all of which the eye can perceive) into one shot was to composite it. But who knew an iPhone had the right software built in to give you the control you need to make such a shot? Note: using “HDR mode” will not achieve this.

Compositing is hard though. The most common method to get the phone to change the exposure is to recompose the image. It runs what photography manuals call a “centre-weighted average” metering mode. It looks at the whole scene, and pays particular attention to the centre of the image. If you want it to change the brightness of the image, just recompose the image to move the centre point.

But this will cause you all kinds of problems in post-production, because you’ll have a much harder job matching your multiple exposures up to create a coherent scene. Changing the angle of the shot alters the perspective, so you won’t necessarily be able to match sky and ground consistently. This’ll show itself up in weirdness around the horizon, with some trees or buildings ghosting in the composite.

What you need is the ability to make the same shot over and over again at different exposure levels. And you can actually do this in the iPhone with what I’ve come to refer to as “tap to meter”.

Look at the following sequence of photos. In each, the shot is unedited, save for being resized. I’ve placed a red box in the zone I tapped to meter the scene.

In each case, the overall brightness of the image is radically different, as the phone takes my tap instruction and tries to weight the overall image brightness to make my tap point the mid-level of the image. But the composition of the image itself is unchanged.

This makes compositing easy. If you can hold the phone relatively still while you tap to re-meter and then reshoot, your matching exercise (precisely overlaying each image) should be simplicity itself. Then you can pick the brightnesses you need to make the image level correspond with what you saw, compensating for your camera’s ineptitude.

If you want some more information on how to actually do the compositing, leave me a note in the comments.