Getting into the camera, this is a pretty solid feeling body, feels firmer than the previous 5D or 7D bodies (and it’s about 50g heavier than the 5DmkII) and with a battery, you are looking at almost a 1kg camera body. The button layout has changed a bit, but the buttons itself feels soft on the fingers and they are all very responsive. It also now features a few buttons from the 7D (video switch and AF-mode selections) and a few buttons that didn’t exist before (like the “Rate” button). The menu system has changed a bit with a few “pages” for every section. slightly different, but not difficult to get through. Pressing the
If you have been shooting with anything in the xxD or xD range, you will find your way on this camera very quickly.
What I don’t like, is the locking button on top of the selection dial. I don’t understand the need, and I believe it’s going get damaged. I might be a bit biased because I kept on breaking these off on my EOS5 Film camera until they replaced the dial with one that doesn’t have the locking button. Other people like it… Time will tell.
The handgrip has changed a bit in shape and the whole body is ’rounder’ with softer lines. Picking up the camera gives you a lot of confidence, and I agree with the comment somebody else made. It makes you want to take pictures. I am waiting for the battery grips to become available, but for the first time in a very long time, it doesn’t feel like the camera needs bulk to give it confidence.
Performance wise, I like it!
AF is crisp. Faster than the mkII (which is not difficult) but it does feel markedly snappier. With the 70-200 f/2.8 L mkII on both the mkII and mkIII bodies, all of a sudden the 5DmkII feels slow. It has never bothered me in the past, but side by side, it feels like the AF is sticking on the old body. Another thing that was pretty nice is the AF locking in near darkness. This is not something I could do with the 5DmkI or mkII. The 5DmkIII has the same focusing system as the Canon 1Dx with 61 focus point and even 5 double cross type points. How well these function in a day to day working environment, I am not sure… I have not yet been able to test this effectively at all. I believe it will be good, given that almost everything else on this camera is impressive. The new Digic5+ processor running this camera is quite a bit more powerful than the combined Digic4 processors of the 7D, for example.
There are various focussing modes, like the 7D and then, there is a complete menu section allocated only to AF functions and setting up AF function according to your shooting environment and type of style. You can set the camera to rather keep focus on a subject even if something else snaps in front of the image, or the camera can snap forwards and backwards as it ‘sees’ new things entering the frame. You can fine tune your style to get the best performance out of the AF. Setting up shutter release is also variable. In AI tracking mode, you can set the camera that the first shot will have priority on focus acquisition or getting the shot (or anything inbetween) then the second shot can be set up to fire as soon as possible after the first, or re-acquire focus before taking the next shot, and again, something in between. This makes the camera function and speed very versatile.
For the slower shooters, like portrait and fashion, advertising photographers, a nice little feature is that you can set a focus point for landscape shooting different than portrait shooting. So if you are switching fast between orientations, you can set a different focus point for each, or keep a single point, whichever way the camera is orientated.
What about image quality? Well… this part is difficult to really give an honest opinion on because there is not a straight forward way to work with the RAW files. You need a beta version of a DNG converter and then use that in your RAW converter, but funny things happen when you start adjusting settings, so I am not convinced we have seen the best this camera can give… that said, looking at the standard RAW files compared with the 5DmkII I’d say we have won at least 1 stop of low light performance.
5D mkII on the left — 5DmkIII on the right – No Noise Reduction, Straight RAW files (Click to open actual pixel size in new window)
This is where things get interesting though…
I may just be converted to shoot JPEG when low light, high ISO images are at the order of the day. I shot JPEG + RAW files while playing around, just so that I have something to work with if the RAW files were a complete failure, but then I saw the images coming off the camera at high ISO in JPEG. They are really, really very good! Even pushing all the way up to ISO12,800 you can get a seriously decent working file. Resize down to 12mp and you actually end up with a very good image. The noise is pleasant, it looks like film grain, so leaning much more to the Luminance scale than the chromatic side like the 60D. Event photographers, Wedding photographers and newborn baby photographers are going to love this camera (did I mention you have two silent shooting modes?)
JPEG Shot… ISO comparisons for detail retention (click on image to open actual pixels in new window)
ISO12,800 JPEG shot, straight out of camera, std Noise Reduction. Size reduced to 12mp to improve sharpness and detail
When the 5DmkIII was announced, I didn’t think I would upgrade. The features looked nice on paper, and I was excited, but I didn’t think I would win enough by upgrading. Unfortunately, just like with the 70-200mm f/2.8 L IS II – Actually working with the body changed my mind so fast.
It’s expensive, but it’s good. I am smiling…
[UPDATE Edit: Added Evening Test]
Before I received my 5DmkII I had a shoot set up for a dancer at the Irene Village Mall fountains with Moya Fourie (http://se.lekt.me/moyafourie) after the close of business. Typically, for I shoot like this, I have an assistant with maglight torches shining on the model to assist with focussing. I know how much the 5DmkII struggles in very low light and backlit conditions, so this was the perfect test for the 5DmkIII (we allready knew the high ISO IQ will be good, so no worries there…)
I expected the camera to be a better performer than the previous model (It has the 1Dx focussing system after all) but this is actually a difficult shooting area for any camera (well… I thought so) It’s quite dark (the background shots for the shoot was done at ISO400, f/8 at 4 seconds) with small highlights and spotlights in the background, making for backlit subjects more often than not.
How can I say it…? Low light focussing and function was quite good. With the 70-200 f/2.8 it snapped into focus more often than not. The nearest condition to this variable lighting was when I was hooting the performers on the Fifa opening concert with the 5DmkII and 7D. The 5DmkII was next to useless, but the 7D gave me a fair keeper rate. Making sure you had locked focus, was however, pretty hard work. In this instance, I think the 5DmkIII did better than the 7D would have (I did not do a side by side test)
In the begining I did fire off a few shots that lock at the back, but switching to cross-type sensors (any of the 41 of them!) sorted out that problem quick fast, and a NICE thing is that the camera shows you which are cross type and which are not (handy when you are choosing between 61 focus points)
So low light performance, I am giving a big thumbs up!
I did pick up a small bug though. I have only seen this at very high ISO, but I am not sure exactly where the problem lies. If I shoot on the motordrive (like I did at low light to ensure I have at least one shot without my movement ruining it all) I have noticed the slightest exposure change between the shots. The actuall settings have not changed according to the exif files, but there is a difference, so it might be small changes in WB (I did have the camera in Auto WB) but I have no way to effectively check the WB settings now. The work around RAW conversion just makes odd things happen to WB-values. It’s not a big enough difference that it would even bother me, but it is there. Will keep an eye on this over the next few weeks and report back.
UPDATE: After a few discussions on various forums, I have come to the conclusion that the differences in WB and exposure might very well be due to the type of lighting used where we were shooting. It was a combination of Neon tubes and EcoFriendly outdoor bulbs… both quite capable of causing shifts in wb and exposure due to the flicker of the bulb. Will test more under more constant light sources and update this post.
Tracking ability I have no idea about just yet, and will update this post in a few weeks after a visit to Swartkops Raceway.
ISO10,000, 70-200mm f/2.8, 1/125th
ISO12,800, 70-200mm f/2.8, 1/125th
[UPDATE Edit: Autofocus test]
Testing the AF system…
Before the 5DmkIII arived in my camera bag, we started arranging for a shoot with some of the polo players at Absolute Polo (http://www.absolutepolo.co.za) and I received a message a day or so ago that they will be playing two small games. Quick arrangements got me to the Crocodile Creek Polo Club, and a few introductions later, I was next to the field ready to go.
I thought Polo would be the perfect sport to test out the Autofocus abilities of the 5DmkIII. It’s a sport with hectic speeds, very fast pace and hectic direction changes, lots of crossing action (one rider crossing and blocking in front of another) and of course… I have never shot polo, so that would make it perfect for the camera to make all the mistakes due to bad setup and of course, lack of foresight as to where riders might be moving to.
On the plus side, Horses are pretty big things to focus on (but they are also very good at blocking your view!)
So lets get the nitty gritty out of the way:
Not bad. Started out cloudy and became clear as the day went on. Everything was hot during midday (11h15 to 14h00)
While cloudy I shot at ISO400 and moved to ISO200 as the sun started breaking out (I wanted to keep shutter speeds up) I started out shooting manual exposure but ended up switching to Av mode. I have never been happy with the results from Av and wanted to see what the camera will do with the patchy clouds and sun/shade/sun variations. I could, however, not help myself, and dialed in +1/3rd when the sun started shining through and +1 when the cloud got thick. I used the grass as my 18% grey marker on the histogram to double check exposures.
I shot the whole day on a Canon 70-200 f/2.8 L IS mkII at f/4 on a Benro Carbon Fibre monopod and DJ90 monopod head. Ideally this sport requires much longer lenses, but lucky for me, a lot of the action happened on my side of the field and cropping the 10-20% of the 5D’s 22mp shot will still leave a seriously nice image.
The AF menus in the 5DmkIII has an explanation of what every setting does and luckily, Canon also created a document with image examples of where the best kind of settings will work, and how they will influence the camera. I skimmed through this document quickly before going to bed, and then thoroughly re-read it in the morning before I went out.
I set up my AF priority for first and second shot to Focus instead of shutter release (take the shot only when focus is achieved, as opposed to taking the shot when the button is pressed) and started of tentatively with Case1: Versatile multi-purpose setting. I soon switched over to Case 3: Instantly focus on subjects suddenly entering AF points. In afterthought, I should have made a few more changes during the day, but case3, in it’s default mode seemed to work fine. so I stuck with that.
I Started of on the AF point Expansion setting (8 surrounding points) but ended up going full 61 point automatic selection. What this does is to start you of on a pre selected focus point, just like AF expansion, but then it tracks through the frame… the speed at which is will switch AF points or groups can be set up as well, depending on what you are shooting, but again, for today, the default settings worked fine.
Word of warning! I have selected a supercool feature on the camera: Orientation Linked AF points – Separate AF points. What this does is to remember the last place you put an AF point when you were shooting that way last (so, different points for horizontal and vertical) this is great, except, it also remembers the AF type you selected, so make sure you have changed them all to the same type before you start your shooting day!
Now… disclaimer time.
I am not a super wildlife or sports fundi, and the stuff I generally shoot never moves too fast, and apart from the Canon 7D I owned I have never had a camera with any kind of advanced, high speed AF abilities. I went into today with the idea that if I hit a 50% keeper rate out of any sequence of shots, I would call it a success (honestly, you are just going to use one shot any ways, why bother with having more?)
This camera blew my mind! I have never, ever had so much confidence in a piece of equipment as now, while writing this short review. I would lift the camera, touch the shutter, and the image snaps into focus, I fire a sequence, and let go. Apart from chimping (and who believes their LCD display for sharpness in any case) I had no idea if I was getting anything sharp, but the response was so solid without hunting, that my confidence grew with every pass.
My technique for multiple shots has it’s roots in my 4 frame buffer Canon 10D that wouldn’t start clearing the buffer before you let go of the shutter button. I would shoot two or three shots, release and double tap again. This has been how I always shoot, double tapping. Inadvertently 1 would be in focus, one slightly soft (sometimes the first, sometimes the second) For this test, I forced myself to also run 7 frame sequences, just to see what the camera would do. Well… it did just fine!
Out of just over 600 shots, I have the absolute minimum out of focus shots. Were they all perfect? Heck no! but they weren’t bad. Out of any given series of shots, I wouldn’t have more than one out of focus shot. If the action stayed more or less predictable, I had a 100% hit rate. When the action drastically changed direction, I would have a dodgy shot, but the very next shot would be pin sharp. How much was the fault of the camera, and how much my bad technique, I am not sure, but since I was basically shooting default settings, I assume I can go a far way to set this camera up to perfection!
I know I am repeating myself here, but I can’t tell you how much confidence I have in this camera right now. To be honest, on the shorter side, you can hardly ask for a better combo than the 5DmkIII and the 70-200 mk II (all AF points function at peak and active – not true for all lenses and lens/lens+converter combinations) but… Considering all this, with the knowledge that the 1Dx has everything the 5DmkIII has + a dedicated processor to AF plus all the new AF assist features that is not in the 5D, I can’t see how you can miss a shot with the 1Dx… really, I can’t! Ok, I can see it, but then it was because of you, not the camera!
Ok, so the big question: Is the Canon 5DmkIII better/faster/more responsive/more accurate than a Canon 7D?
Tough question. but I am going to have to say yes. I have not had the opportunity to put them side by side and do a slightly more scientific test than pure gut feel, but in the year that I have been shooting with the 7D (same lenses) I have never had this keeper rate or this confidence in the camera. Maybe I didn’t set my 7D up perfectly, maybe I was just low on caffeine, but that is what I think and feel today.
All Images Copyright Sean Nel