6.2 Exposing to the Right

As already mentioned in section 2.5, there’s an attractive method for setting exposure levels which is just now starting to gain popularity among large numbers of digital photographers.  The inventor of Photoshop has been credited with first proposing the idea.  It’s called ETTR, or Exposing To The Right.  The idea is simple: when in the field, you try to achieve the brightest possible exposure without obliterating any details via clipping.  Clipping results in regions of the image that are pure white—that is, all the pixels in those regions have exactly the same pixel values, and therefore encode no detail whatsoever.  When that happens, we say that we have blown the highlights.
    The goal in ETTR is to maximize the exposure (i.e., image brightness) in the field without blowing any highlights.  There are several reasons for maximizing the exposure, and it’s very important for you to understand all of them.  First and foremost, if you can maximize the exposure without resorting to high ISO values (i.e., if you can maximize brightness via aperture and shutter speed alone), then you’ll be reducing image noise, by maximizing the number of photons collected by the sensor.  Recall from section 2.5 that underexposure in digital images results in small photon counts, which in turn results in noise.  And since noise becomes more apparent the more you blow up your image (to larger sizes), more noise in effect translates to less usable detail when sizing an image.

Fig. 6.2.1 : An extreme case of ETTR.  Notice that all the data is positioned
near the rightmost end of the histogram, with just a little space to ensure that
highlights are not clipped.  This photo also happens to be an example of a
photo—i.e., overwhelmingly white. Don’t confuse ETTR with high-key:
they’re not the same.  (1/800 sec, f/7.1, ISO 500,
600mm, manual mode, no flash)

    A second benefit from increased brightness during image capture is that the image will be encoded (in the RAW file) using more bits per pixel, on average (a bit is the fundamental unit of information in computers).  This is simply due to the standard practice of encoding image data in such a way that brighter pixels are allocated more bits than darker pixels, which means that there are more shades (or hues) of each color that can be accurately represented at the higher end of the light spectrum than at the lower end.  Thus, by exposing to the right you should, in theory, be better preserving any fine color gradients that may be present in the scene you’re capturing.  Some commentators have questioned whether this benefit really has much effect in practice, but there at least seems to be no harm in the allocation of greater numbers of bits to images (other than slightly larger image files), and if it does occasionally improve your images then it’s worthwhile.
    Finally, some have suggested that a further reduction in image noise may be achieved by producing larger electrical signals from the imaging sensor, in that the downstream electrical circuitry (especially the analog-to-digital converter, or ADC) may—at least in some camera models—generate relatively less noise (relative to the signal) for the higher pixel values.  Note that even if this turns out to be true, it may be very camera-specific, so don’t assume that it necessarily applies to your particular model.  Also be aware that on some low-end camera models the the full-stop ISO values (100, 200, 400, 800, and 1600) can be less noisy than the between-full-stop values (125, 160, 250, 320, 500, 640, 1000, 1250), since on some models these latter ISO’s are achieved via digital multiplication of the discretized bits, rather than via analog amplification.
    A slight variant of the ETTR philosophy calls for maximizing the amount of information in the subject, rather than maximizing the overall information in the image as a whole.  In the case of bird photography, I call this BETTR, or Bird Exposed To The Right (not to be confused with Rush Limbaugh’s pet canary).  The difference between ETTR and BETTR is that in BETTR we allow the blowing of any highlights that aren’t in the bird (or in the immediate foreground around the bird—such as in the branch that the bird is perched on).  That is, we try to maximize the exposure of the bird without blowing any of the bird’s highlights, but if we have to blow highlights in the background, then so be it.  Since the background is often rendered out of focus anyway (by a shallow DOF), trying to retain details in the background serves little purpose.  There are, of course, exceptions—such as when the background has rich color gradients that you’d like to preserve.  In those cases, BETTR might in fact not be better than ETTR.

Fig. 6.2.2 : A bird exposed to the right (BETTR).  In order to maximize the information
in the bird, I had to blow the highlights in the background, which I usually don’t mind doing
since I prefer blurry backgrounds anyway.  That allowed me to maximize the detail in the
bird, which is what I usually want.  Just be careful not to clip any highlights in the bird!
(1/640 sec, f/5.6, ISO 1600, 600mm, manual mode, manual flash at 1/4 power)

    The crucial point of ETTR (and BETTR) is that you’re not trying to make the image look as nice as possible on your camera’s LCD.  When employing ETTR/BETTR, the images that appear on your camera’s LCD might look downright unpleasant.  When you get the image onto your computer you can then adjust the exposure digitally back to a more pleasing level (during RAW conversion—see Chapter 12), while retaining more details in the image (or at least retaining a greater potential for accentuating subtle details via further postprocessing).  The idea behind ETTR is to maximize exposure in the camera, even if you have to turn down the brightness once you get the image onto the computer.  This will help you to avoid systematic underexposure of your photos in the field.  As long as you don’t blow the highlights, darkening the image in postprocess is always better than brightening it in postprocess, because artificial brightening of underexposed images in postprocessing typically brings out more noise, whereas darkening images that are overexposed (but not clipped) doesn’t.
    In order to employ the ETTR/BETTR technique, your camera needs to have one critical feature: a highlight alerts function.  This function causes blown highlights in the captured image to be detected and displayed prominently on the back-panel LCD of the camera, typically via bright flashing.  Most of the newer DSLR’s on the market today have some form of highlight alert.  What’s important is that the highlight alerts be available during the
image review period—i.e., as the image is automatically shown on the back LCD panel just after capture (typically for 1 or 2 seconds).  Having the highlight alerts automatically show up immediately after each capture allows you to rapidly assess your exposure during the heat of intense shooting, so that you can make quick adjustments (if needed) to the exposure in-between shots, without drawing your attention away from the bird for too long.

Fig. 6.2.3 : ETTR helps retain subtle details in bright areas.  Notice the many subtle
details in the front/underside of this white-eyed vireo.  Some of these are due to the use
of flash (see Chapter 7), but much of this detail would likely not be present if the photo
had instead been underexposed and then brightened in postprocess.  ETTR ensures
that you rarely have to brighten images in postprocess.  Darkening images in postprocess
is always better than brightening them in postprocess, for a variety of reasons (see text).
(1/80 sec, f/11, ISO 250, 840mm, Av with -1 EC, evaluative metering, TTL flash at -2/3)

    As mentioned in Chapter 2, If you can perform ETTR by adjusting only the shutter speed and/or aperture, while keeping ISO very low, you’ll be both maximizing the number of bits allocated to your image’s detail and at the same time minimizing photon noise (i.e., minimizing sampling error by not underexposing the image).  Exposing to the right via shutter speed and aperture (but not by using higher ISO’s) reduces photon noise by allowing the sensor to collect more photons, thereby reducing sampling error and ensuring a more accurate measurement of color information from the incoming light.  Remember, collecting more light allows a more accurate measurement of the individual colors making up that light.  If instead you expose to the right by increasing the ISO setting, you’ll be getting the benefit of better bit utilization (since you’ll have brighter pixels, which are generally allocated more bits in the RAW file than darker pixels), but you won’t be reducing photon noise at all, since you’re not collecting more light (you’re just amplifying the signal after it’s already been measured by the sensor).  Depending on the noise characteristics of your camera’s downstream electrical circuitry, turning up the ISO after you’ve already exhausted your options for maximizing exposure via aperture and shutter speed might result in additional benefits (besides better bit utilization).
    Just remember that ISO doesn’t affect the sensitivity of the silicon atoms in your sensor; it simply multiplies the photon counts after the photons have already been counted.  When applying the ETTR technique, if you have to do it by increasing the ISO, that’s OK: you’ll still get some benefit due to better bit utilization.  But if you can do it via shutter speed and/or aperture instead of higher ISO, you’ll also be reaping the benefits of lower photon noise.  So, only turn up the ISO once you’re sure you can’t increase the exposure level via aperture and/or shutter speed.
    It’s important to note that achieving ETTR via the use of flash is generally a very bad idea, for several reasons.  First, since the use of flash actually introduces more light into the scene, the potential for feather glare (see section 7.11) increases as you increase the flash output level.  Although extreme feather glare may show up on your highlight alerts, less extreme glare that doesn’t cause clipping won
t show up in the highlight alerts but may still degrade your image aesthetics and can be very difficult to correct in post-process.  Second, because flash illumination falls off nonlinearly as distance increases (see section 7.2), reducing exposure digitally in post-process can be very tricky, since most exposure adjustment tools in software perform linear adjustments (and they certainly can’t compensate directly for in-the-field distances).  A good rule of thumb is to keep the flash output moderately low when finding your initial (non-flash) exposure parameters, and then once you’ve found the parameters which give you an ETTR exposure, increase the flash only as needed to illuminate shadow regions on the bird.  You may then need to adjust other exposure parameters to mitigate clipping introduced by the fill flash.  They key here is to use flash for fill only: don’t keep increasing the flash output to achieve ETTR, because you’ll be inadvertently affecting flash ratio and possibly introducing feather glare.
    Postprocessing of images captured via ETTR/BETTR is addressed in Part III of this book.