A couple of weeks ago, Samantha and I along with our friends Lori Maloney and Wayne Simpson went out to Hamish Kerfoot’s place to photograph model Talyn Stone messing around with some of the old cars on the property. Hamish gave us a tour of the ranch and volunteered to hold our reflectors but, more importantly, Hamish was the ‘official’ spider wrangler to keep the eight-legged beasts away from Talyn.
Below is a wee video of the proceedings followed by a few photos that Sam and I took of Talyn. To see more photos of the talented Talyn see our Flickr Set.
After you scroll past the photos, you can check out our first impressions of Canon’s latest camera. Thanks as always to The Camera Store for letting us play with the camera. We are big fans of that store as our credit card statements can attest!
First Impression of the Canon 5D Mark III
The Camera Store lent us the Canon 5D Mark III with a Canon 24-105 f4L lens to use for two days. On the first day I took the camera out and along with my Canon 300 f4L I went searching for moving subjects to test Canon’s new and ‘improved’ autofocus system.
For anyone who has been sleeping for the last 5 years and didn’t know, Canon has had a ‘few’ issues with autofocus on their pro line of bodies since the release of the Canon 1D Mark III and 1Ds Mark III in 2007. Autofocus foibles followed with the releases of the 5D Mark II, the 7D and the 1D Mark IV. Canon lost a large number of sports, photojournalist and wedding photographers to Nikon as Canon shooters became frustrated with the finicky autofocus of these high end Canons.
With the release of the 5D Mark III, reviews began circulating that finally Canon had corrected the auto-focus issues. I mostly shoot landscape photos and use manual focus and so the autofocus problems are not an issue for me. But lately Samantha and I have been returning to photographing more people and pets and so we are searching our options (Canon or Nikon) for an autofocus speed demon.
To cut to the chase I am super happy to report that the 5D Mark III does what it is supposed to do! I spent a few hours photographing cars on the Trans Canada Highway as well as photographing running cows, jumping deer and flying birds. I had hoped that at least 50% of the action photos would be sharp, but I was giddy that the camera locked focus every time on the moving subject as long as I had a focus point or a small number of focus points covering the subject I was tracking.
The verdict is still out on this one for us. Mostly we used the Canon 24-105 f4L lens for the Talyn shoot the following day and in hindsight this was a mistake. We had never tested the lens before and so it becomes hard to dissect how much of the ‘ready-processed’ files we saw from the 5D Mark III were from the lens or from the camera. The images shot with the 300f4L were better than the images with the 24-105 so we suspect the lens was part of the problem. Keep in mind we are used to seeing files from my EOS-1ds Mark III shot with a 24mm tilt-shift lens. The tilt-shift lens is amazing in the sharpness department, so we are spoiled with detailed files. As an aside, after seeing all the haloing and fringing of the 24-105 f4L we would never buy this lens ourselves; it’s a decent go-to product but we’re searching for something with better returns.
Even in spite of the poor performance of the 24-105 f4L, the raw images with the 300mm f4L (which is a pretty good lens) seem a bit ‘cooked’ somehow. It’s like Canon gives you a raw file but there is still some elixir going on behind the scenes. We are not sure we like that ‘Canon hidden magic’. Further testing is in order. For now let’s just say we were not overly impressed by what we got out of the camera in terms of file quality. We’ll go out with a 5D Mark III and our 24mm TS-E and shoot it side by side with the 1ds Mark III and the 5D Mark II to do a real comparison. Stay tuned!
According to many reviews, noise is well controlled in this camera and high ISOs like 3200 and 6400 are totally useable. It’s easy to impress people by showing sample photos of incredibly low noise at high ISO with the 5D Mark III, but these images are always derived from bright areas of the photo. Look at the image below: it was taken at ISO 12800. First, I was blown away that the camera focused so fast in such low light. Second, when you look at the bright areas of the photo, noise is amazingly low given the ISO.
Once you start peering into the shadows, noise is going to get worse. And if you manipulate your files at all (e.g. process them) then the shadow areas will be even more noisy. This camera (like all digital cameras) will be pretty noise-free even at high ISOs as long as you keep all your data in the upper half of the histogram.
Where we see a real test of image noise is in the blue channel. I did a quick test using the dusk scene below and exposing the image to the right side of the histogram for the best data possible. Here is the minimally processed raw capture.
And here is the image as I would process it in Adobe Camera Raw.
I took the same photo at different ISO settings and then ran the resulting images through the exact same processing regime. I checked the sky for noise (dusky blue skies really show noise if there is any) and the cropped results are below.
If you’re going to do any normal processing of your image and if your image has a histogram that is anything but highlight capture, you are going to get noise at higher ISOs especially in the blue channel. The promise of amazing ISO performance is only a theoretical possibility (highlight capture of the upper third of the histogram, and minimal processing). In the real world (and this is what we care about), we think the 5D Mark III works for us up to 1600 ISO.
Finally, we loved the handling and control of the camera (it handles like a 7D). Users of the 5D Mark II will need to get used to the new layout but overall it is intuitive and easy to understand.
So far we love the handling and the quick and accurate autofocus (finally!). We are not overly impressed with the noise at high ISO (but we rarely use high ISO anyway). The real clincher is the file quality. We are not sold at this point because the raw files look crunchy and a little ‘souped up’ somehow. We need to test the 5D Mark III against other cameras to see if what we saw in our first test is just us or the camera or both. Stay tuned!