Sunday, 18 October 2009

Canon EOS 7D Hands on Mini-Review

The EOS 7D is the latest DSLR camera from Canon. Its principle target use is for fast moving subjects, so is ideal for sports and wildlife. I principally shoot wildlife and landscapes, so that will be the emphasis for this mini-review. It should also be noted, that I don’t generally do reviews and will be comparing it to the 40D, the camera it is replacing for me. That said, this review isn’t a direct side by side comparison, but more of a report of hands on experience so far on an early release camera (firmware 1.0.7), with reference to shooting in similar conditions with the 40D. There are others much more able to do scientific comparisons, so I won’t attempt to pre-empt any results they may achieve and report on. In addition, I have very limited videography experience, so I haven’t tested the various movie modes, other than a little playing around.

General Use and Feel

The 7d feels sturdy and well-built. It is significantly heavier than the 40D, which is by no means a light camera. I haven’t tested its weather-proofing capabilities and quite frankly, I don’t intend to deliberately put it out in the rain to see how it holds out. It feels a little different to hold, which has taken a bit of getting used to. It is more ergonomically curvaceous than the 40D, which I think in time I will prefer.
The menus on the whole, are pretty logical. They are an evolution from previous EOS cameras, including the 40D, rather than a major change. Because of the additional features, the menus are much more extensive on the 7D compared to the 40D. My biggest complaint with Canon cameras remains, the mirror lockup remains hidden within the menus, instead of a dedicated button like on Nikon cameras. Also, despite the ability to reprogram the functions of a number of buttons, it isn’t possible to assign mirror lockup to any of them. One feature I do like, is the Q or Quick Access button, this enables quick access to a range of features, including the AF system and focus point selection, but again, not mirror lockup. I do find the buttons are a little harder to activate with larger fingers, because of the slight recess.
The shutter gives a solid “thunk” when the shutter release is pressed. In silent mode 1, it seems quieter than on the 40D, although it is lower pitched in sound, so a direct comparison may be slightly skewed. This gives me the impression that it is much better for wildlife, but only time with shy creatures will confirm that though.

Exposure and Metering

This is one area where I tend not to use cameras to their full potential, possibly due to previous inadequacies, so it is the feature I have tested the least. I tend to keep the camera set to spot metering, even for landscape use. This allows much more precise control of the exposure, particularly for animals in shade or when using filters for landscapes and testing the various exposure ranges in a landscape.
From the limited use, I find that just like the 40D, I need to slightly overexpose, but only by around 1/3 stop. This does help to limit any noise from underexposure of shadows. The new metering system does have the potential to solve metering problems when using graduated filters and is something I need to force myself to try, as it could shorten the time taken to respond to changing lighting conditions in stormy conditions and close to sunset and sunrise.

Focusing System

The new focusing system is pretty complex and I’ve really only scratched the surface so far. No doubt with time, I will be able to refine my choices for different situations. Ever since the highly publicised problems with the EOS 1D Mk III, Canon has taken quite a pounding of the performance of its autofocus. Reading between the lines, there was a lot of hype and over-reaction, but there were definitely problems. There have been a number of occasions, where I haven’t been entirely happy with the autofocus on the 40D, but it was a significant improvement over the 400D, so the hyped AF on the 7D was of definite interest to me. In light of the controversy surrounding the 1D Mk III, suggesting that some people just didn’t know how to set it up properly, exacerbating the problems that have been well documented, I have been determined to do my best to test out different modes. While I haven’t completed a full test yet, I have tried out the main modes that are supposed to be the most useful for the sort of subjects I shoot.
Initially, I tried AF using my Canon 100-400mm and Canon 100mm macro lenses, but I have now tried some more extensive testing with my new 300mm f/2.8. With both the 100-400mm and macro lenses, I tested the ability of expanded point selections to focus accurately.
I had mixed results on the 100-400mm. With close subjects, focus was very good, but trying to focus on birds in flight that were small in the frame, seemed to reflect many of the experiences noted in some forums. That said, I had some fairly good results too. On one occasion, there was a marsh harrier, at some distance (about the same size in the frame as the expanded points), against a background of trees, with reeds in the foreground. Pretty much at the limits of the lens and with the 40D, it would have struggled. While I wouldn’t say that anything was good enough to sell, out of 70 frames, only about 11 were completely out of focus, with the rest being acceptable for personal use. Also, I had very good results with a kingfisher perched on a branch, in windy conditions, with leaves blowing in and out of frame, in front of the bird. I then tried photographing some mute swans in flight that were much closer than the marsh harrier, placing the central point of the cluster on the head. The results quite frankly were awful (unfortunately, I deleted them in disgust, so can't post them), but there also seemed to be a problem with the IS, as there seemed to be evidence of motion blur, despite a shutterspeed of 1/1600th second, plus they were almost “grab” shots, so the focus may not have had time to lock. Subsequent testing on a stationary subject at home (my back fence) showed that the lens actually front focuses quite badly, at least on the 7D, which presumably becomes more noticeable with distant subjects, particularly large birds such as a swan, where the AF may lock onto the wing tips, exacerbating any other problems. I have since adjusted the focusing, using the AF microadjustment (-20, the greatest adjustment possible), so I need to test further, as well as testing with the other AF modes.
For macro work, I almost exclusively focus manually, but I was interested to see how it would perform, considering some comments about the use of AF on the 7D when coupled with the new 100mm f/2.8L IS macro. As I don’t have that lens, I obviously couldn’t test all aspects, but I could test how the AF would cope with a swaying photographer, handholding at close distance with the original 100mm f/2.8 macro. Focus was ok, but not what I would want, when I used the expanded points selection. Following experience with the 300mm (see below), I want to hold off judgement until I’ve had a chance to test the spot and single point AF modes. Unfortunately, as autumn is upon us, with winter fast approaching, suitable insect targets are disappearing, so further testing may need to wait until the spring. Until then, for critical circumstances, I will continue to focus manually.
Next up was the 300mm f/2.8L IS, so far just with the 1.4x extender. This is one of the sharpest and fastest focusing lenses in the Canon arsenal, so if that couldn’t focus, then not much would. Of course, the narrower maximum aperture meant that depth of field would become an issue, making accurate focusing much more critical. Initially, I used the expanded points selection mode, just like with the other two lenses. As with the 100-400mm, if the subject covered a significant proportion of the frame (at least half), then it was possible to place the expanded points selection over the eyes for pin-point focus and in those conditions it worked very well. However, if the subject was smaller in the frame, so that the expanded selection covered more of the subject, then accurate focus was lost. At this point, I switched to single point mode, with much better success. I was able to use the single point to focus on the head of mute swans that were quite distant, to get sharp focus at the more distant ranges. The faster moving raptors aren’t around at the moment to test how it would cope with such a small focusing area on fast moving subjects and the slower moving buzzards and marsh harriers wouldn’t be much of a challenge for the AF system (other than wingspan problems and depth of field). I then decided to really test the AF system, with a very distant kestrel, that was little more than a speck in the sky with the naked eye. In fact, it was so small in the frame, it was smaller than the single point focus box. As a consequence, the single point focus showed the same problems as the expanded points for such a small subject, as I would have expected. In such conditions, I wouldn’t even have attempted anything with the 40D, as it would have failed miserably. I then tried the new much smaller spot focus, which was incidentally, about the same size as the very small kestrel. While it was by no means sharp, for such a distance, I must say I was impressed how it coped and some of the lack of sharpness is probably as much to do with compression of the moisture in the atmosphere. Initial testing with focus tracking using single point AF, gave good results, which will probably improve with practice.
In summary, the expanded points focus mode can be disappointing, but I suspect this is down to the size of the subject compared to the focusing area as much as anything else. My tests so far, suggest that for close subjects, the expanded points is fine, but for more distant subjects, the single point focus is much better for accurate focusing, provided the tracking can keep up or the photographer can keep steady enough. If spot focusing is required, then chances are, the subject is too distant anyway, so tracking moving birds etc., is more of a moot point with it, being more useful for static subjects, such as a kingfisher between leaves, which is a different type of challenge for an AF system. Also, some reported poor results may be due to lenses that need to be calibrated. Most will probably be close enough to optimum focus not to be noticeable, but some may require a significant adjustment.

Tonal and Dynamic Range

This isn’t easy to qualify, but I have been able to photograph some scenes with quite a high dynamic range, including some in which I was able to shoot without the need for a neutral density graduated filter. Tonal range seems to be at least as good as the 40D, certainly enough for most scenarios. No doubt a full frame sensor would exceed what the 7D can achieve, but it holds its own, to produce good landscape images, with plenty of resolution. One concern I have had, is that images from my zooms (even though they are L zooms) may appear soft, from the extra sensor resolution, but so far, I’m not finding too much of a problem, although I am yet to try out my 17-40mm f/4.0L, which is the lens that seems to be more affected on the 5D Mk II.

Noise Control

One of the features I was looking for in a camera to replace my 40D, besides a higher resolution, was better low light capabilities. High ISO capability is handy, but only so far as the quality of the images produced. If the high ISO settings produce high noise levels, then it limits the use of any photographs obtained. When I first saw the press releases, I paid close attention to the sample images in the pre-production reviews that were circulating, particularly the scant few that looked at wildlife. Sometimes I think too much is made of noise, people can sometimes lose sight of the fact that some images simply wouldn’t have been possible at low ISO. However, if the detail is obscured by the noise, then it is pretty much unsaleable. The 40D was pretty good up until ISO 800, provided exposure was good, but then it started to deteriorate, although in some circumstances you could get away with ISO 1600, if the background wasn’t uniform. So far, Adobe only offer beta support for 7D RAW files and I’m not keen on Canon’s software, so it isn’t possible to see the full potential yet, but I would say that the 7D is at least one stop better than the 40D, pretty impressive for those extra 8 MP. I’m hoping it will prove better than that, but time will tell. From the images I’ve seen taken with the 50D, it is an even bigger improvement in terms of noise control at high ISO. One thing that is certainly improved is the banding of noise that was a feature of previous Canon cameras at high ISO. The noise on the 7D is free of banding and is much more pleasing and less distracting as a result. The dedicated second DIGIC IV processor is certainly having an effect.

Conclusion

The Canon EOS 7D is by no means a miracle camera, but it has narrowed the gap in terms of noise control with Nikon. Whether or not it exceeds the capabilities of Nikon cameras is a matter of debate and may be down to personal choice, as most early indications, based on reviews suggest that the quality of noise is different and therefore not necessarily comparable. In addition, Canon have increased the sensor resolution, without any noticeable decrease in performance in other areas. If I’m honest, I would have liked to have seen the results if the sensor had stayed in the 16 MP range, but I’m happy with the results as it is. The frame rate is ideal for both sports and wildlife photography, where the moment can sometimes fall between shutter releases, making the frame rate vitally important. There are many arguments between full frame and crop sensors, but the fact remains, that the effective magnification from the crop sensor is very important for wildlife photography, where you often need as much reach as possible, plus it would have lowered the frame rate considerably. It has its downsides, but overall, the APS-C sensor in the 7D is what makes it so appealing to the semi-pro or enthusiast wildlife photographer. Professional photographers will probably stay with the 1D series, particularly when the eagerly awaited Mk IV is released, but the 7D would make an ideal second camera even for professionals, particularly where weight is an important factor.
Autofocus aside, there are very few limitations. It would be nice to have a bigger image buffer in RAW shooting, but that teaches you to choose your moment. Metering seems pretty good and with more testing it could be even more useful and noise control is also significantly improved over the xxD series. The autofocus still needs further evaluation to confirm its performance and best shooting modes for different situations, but it does look promising also. Overall, the 7D is a worthy upgrade from the xxD series and will also appeal to semi-professional and professional wildlife photographers as a second or backup camera. Will it tempt Nikon owners and is it a world beater? Probably not, but then it would have to be something pretty special to tempt Nikon owners to switch systems, having invested a lot of money in glassware. In this price range, after thinking about whether it would be worth switching to Nikon for a while, I think Canon just about has the edge, at least for a couple of months. Some of the early niggles will also be fixed by the first firmware updates too.

2 comments:

  1. For anyone interested, the marsh harrier and kestrel are 100% crops. To see the full size crop (and therefore the actual number of pixels), please click on the images. For reference, the 7D produces images of 5134 pixels on the longest side. All images are unsharpened and are created from RAW files. The photo of Bridgwater quay has had a medium contrast curve applied, the others are unprocessed (the kingfisher didn't need it).

    ReplyDelete
  2. Canon have released a firmware update (1.0.9) today as expected.

    ReplyDelete