Sony A550 Review
The Alpha 550 has the latest version of Sony's proprietary dynamic range optimization feature, DRO (similar features are now appearing on most SLRs, most based on technology developed by Apical). DRO works by altering the tone curve of the image to brighten the shadows without significantly altering the highlights. The Alpha 550's version offers both automatic and manual control over the amount of DRO applied (though as we've seen before, the auto setting is very conservative, often showing no effect at all). The highest setting is useful for dealing with very high contrast situations (such as shooting in a dim room with a lot of light coming through the windows), but it does significantly increase shadow noise at the higher settings.
DRO doesn't interact with the camera's exposure metering, so it makes no attempt to capture additional highlight information - it's all about the shadows. As a result, users who shoot exclusively in RAW mode will not see any effect of DRO (it's a post-shot processing step so it only appears in JPEGs), and to get the best effect you really need to expose for the highlights. The best idea for more advanced users is probably to shoot raw and use the included software, which offers even more control over the DRO function (and offers independent control of the highlights and shadows).
|DRO Off||DRO Auto|
|DRO Lv1||DRO Lv2|
|DRO Lv3||DRO Lv4|
DRO effect on dynamic range / tone curve
Looking at the tone curves for the various DRO settings it's easy to see exactly what Sony is doing. The Lv5 setting produces shadows that are around four stops brighter than standard (DRO off) JPEGs, which explains the increased noise.
The most intriguing new feature on the Sony Alpha 550 is its Auto HDR mode. DRO is great for lifting shadows and dealing with high contrast shooting situations, but it can't actually increase the dynamic range of the captured image. That's where the Auto HDR feature comes in. The principle is simple: two shots are taken in rapid succession at different exposure settings (up to 3 stops different to be precise) and the powerful Bionz processor then merges them to create a single image.
This is hardly a new idea: it's the basis of all HDR photography techniques - it's not even the first camera to offer in-camera HDR merging (in fact the Pentax K-7 not only beat Sony to the gate on this feature, but it offers even wider dynamic range since it uses 3 shots per image). Where the Alpha 550's mode is unique is that you can use the camera hand-held as the processor can compensate for any small movements between the shots. Inevitably if you move the camera too much (or use a shutter speed slow enough for camera shake or subject movement to appear) you will get some weird effects (where there's too much difference between the shots).
Compared to some of the wild HDR effects produced by enthusiasts using multiple exposures covering 20 or more stops the Alpha 550's Auto HDR mode produces a relatively subtle result (with a maximum 3 EV between the exposures that's to be expected), but for balancing out scenes with a dynamic range too large to be captured in a single exposure, it works well. I can imagine landscape photographers playing with this feature a lot (when the sky needs 1/500th sec and the foreground needs 1/60th there's not a lot else you can do to preserve both shadow and highlight tones).
There are a few things to watch out for. Firstly, as mentioned above, you'll get some very odd results if there are moving objects in the scene or if you move the camera too much between exposures. Secondly there are situations where the second exposure is actually quite long, meaning camera shake is a risk (though in our experience the SteadyShot worked well enough to avoid this being an issue in most cases). Finally the effectiveness of this (and any other similar) system relies very much on the accuracy of the metering the two exposures are based on. In our experience this aspect of the Alpha 550's performance was far from foolproof, so be prepared to use exposure compensation (or AE lock) if you don't get the result you hoped for. The problem is most obvious with scenes that require more than 3 extra stops of dynamic range (such as when shooting inside a dimly-lit building with a large window looking out onto a bright day. In these cases it's almost always better to meter off the brightest area (i.e. the window) or you'll end up with an image that looks over-exposed.
Overall though, this is that increasingly rare thing - a novel, clever new feature that not only works as advertised, but is likely to prove genuinely useful in many situations.
Auto HDR examples (handheld, using the 3.0EV maximum setting)
|DRO and Auto HDR Off||Auto HDR On|
|DRO and Auto HDR Off||Auto HDR On|
|DRO and Auto HDR Off||Auto HDR On|
Auto HDR (3.0 EV) vs DRO (Lv5) vs OFF
There are some fairly important differences between the two dynamic range options (kind of obvious really, since they're entirely different technologies), though with careful exposure you can get very similar results out of them in many situations, especially if you apply DRO (the purely digital 'one shot' dynamic range expansion) to raw files rather than letting the camera do its (irreversible) work on JPEGs.
The most obvious difference between the DRO and Auto HDR modes is that DRO really does push up noise in the shadows as it boosts them, and it can (as seen below) result in some color shifts (mainly seen as an increase in saturation of primary colors in anything darker than middle grey). It's also worth pointing out that there's nothing in DRO that you can't do with raw files, but for the novice user it's obviously handy to have all this done for you in-camera.
The Auto HDR mode produces images with far lower noise, and the results look a lot more natural - or at least as natural as HDR images can look (I should declare at this point that I'm not a big fan of HDR unless it's absolutely essential). The biggest problem we had was anything that moved significantly between the exposures produces inconsistent results, though to be fair this only happened in low light at longer shutter speeds. We also found that the system produces some very odd images when it meets chromatic aberration and blooming (such as you get with the kit lens, used here, in overexposed areas), and occasionally with out of focus areas.
|Auto HDR (3.0 EV)||DRO and Auto HDR Off||DRO Lv5|
- 15 Photographic tests (DR)
- 16 Photographic tests (DR)
- 17 Photographic tests
- 18 Compared to
- 19 Compared to (JPEG)
- 20 Compared to (JPEG)
- 21 Compared to (RAW)
- 22 Compared to (RAW)
- 23 Compared to (Higher ISO)
- 24 Compared to (Resolution)
- 25 Compared to (Resolution)
- 26 Kit Lens test
- 27 Conclusion
- 28 Samples
Dec 9, 2009
Aug 27, 2009
Dec 5, 2012
Dec 7, 2012
- Fujifilm X-T223.6%
- Nikon D50025.4%
- Nikon AF-S 105mm F1.4E8.2%
- Olympus M.Zuiko 12-100mm F47.5%
- Panasonic Lumix DMC-G857.2%
- Sigma 85mm F1.4 Art6.7%
- Sigma 50-100mm F1.8 Art5.1%
- Sony a63006.4%
- Sony Cyber-shot RX10 III3.7%
- Sony Cyber-shot RX100 V6.3%
|Lighthouse, Bottom of the World by CelticOdyssey|
from An A to Z of Subjects- Week 12, L
|Dundrum by Rik Powdrill|