In the world of camera gear, teleconverters aren’t exactly the glamorous subject matter that most people want to talk about. However, if you’re a regular wildlife photographer, a teleconverter is a staple tool in the bag.
Eight years ago now, back in 2010, Canon released the latest version of their EF mount 1.4x and 2x extenders; the Extender EF 1.4x III (discussed here) and it’s sibling the Extender EF 2x III (which I won’t be covering directly).
Canon touted a number of both optical and electronic and mechanical improvements over their predecessors in the mark III converters. Broadly these changes are improvements to the old formula.
Optically, Canon changed the formula for the extender from 5 elements in 4 groups, to 7 elements in 3 groups. One of these elements is now an “anomalous dispersion lens”, which reduces longitudinal chromatic aberrations. The updated optical formula makes a small, but certainly noticeable improvement in image quality.
Additionally, the outer optical elements (front of the front most element, and rear of the rear most element) are coated with a Fluorine coating to reduce smudges and fingerprints, and make it easier to clean the surfaces.
Finally, an improved micro-controller rounds out the major upgrades by providing improved AF performance compared to it’s predecessor.
Worth the Upgrade from the II?
With all that said though, in many ways, the 1.4x III isn’t that much of an improvement optically over the 1.4x II that immediately warrants upgrading. Given that, it’s reasonable to ask, if I already have a 1.4x II, as I do, why upgrade to the 1.4x III?
As I said, optically there certainly are benefits. Especially if you’re pairing a really sharp super-telephoto prime with a modern high resolution body (5DS/5DSR, even the 5D mark IV, and certainly the 7D mark II and 80D) where you need every bit of image quality you can get.
However, the real benefit as I see it, and what prompted me to ultimately upgrade, is the improved capabilities permitted with Canon’s newest AF systems such as the ones in the 5D mark IV, 1DX mark II, 6D mark II, and 80D.
Canon has always taken a conservative approach to their AF systems, instead of putting the user in a situation where something may work but unreliably, they will disable the capabilities. Personally, I think this is the right approach. In my experience, I’d much rather have a camera that refuses to focus at all when the lens and teleconverter is outside of it’s capabilities, than one that focuses improperly but thinks it is right.
One of these cases is auto-focusing with a lens and teleconverter at f/8. With their latest generation of cameras, Canon has really embraced the idea of having an active phase detect AF system with lens and TC combinations that have a max aperture of f/8.
However, there’s no free lunch here. The lens and TC combo has to have good enough optical performance that the AF sensor can reliably operate, and that means using a 3rd generation TC.
This is of course a huge boon for mid-tier users and those who aren’t specialized enough in wildlife and sports photography to drop $10,000 on a 400mm f/2.8, 500mm f/4, or 600mm f/4. For those of us using 100-400mm and 70-300mm zooms that top out at f/5.6, and become f/8 with a teleconverter, this is quite a nice boost in flexibility.
I should note, the same stands with the 2x III as well. A user with something like a EF 70-200mm f/4L IS USM can add a 2x III TC and get a 140-400mm f/8 lens that will have most or all of the AF points available on their camera (depending on model).
Caveats (Supported Lenses and Stacking)
Canon’s extenders, both the 1.4x III and the 2x III, have front elements that project into the rear of the lens their mounted to. As a result, they’re not universally compatible with all of Canon’s lenses, and I can’t even begin to comment on compatibility with 3rd party lenses.
Due to this design, the Extenders are only compatible with Canon’s L series lenses that have a 135mm or longer focal length, and the following L series zooms: EF 70-200mm f/2.8L IS II USM, EF 70-200mm f/2.8L IS USM, EF 70-200mm f/2.8L USM, EF 70-200mm f/4L IS USM, EF 70-200mm f/4L USM, EF 100-400mm f/4.5-5.6L IS II USM, and EF 100-400mm f/4.5-5.6L IS USM.
Additionally Canon has made changes to the EF 2x III’s rear element that no longer makes it possible to stack the 2x and 1.4x teleconverters together to get a 2.8x increase in focal length.
Since there are image quality and AF performance differences, I wanted to do at least a couple of simple tests to look at them. As I said, I bought my Extender EF 1.4x III purely to have all my AF points at f/8 on my 5D mark IV, improved image quality and AF performance were secondary considerations for me.
To look at how the TC affects AF performance, we’ll look at 3 different test cases. The first two cases are AF hunting, once in dark conditions (with the lens cap on) and once against a light target that has nothing to focus on.
Hunting performance is a useful indication of the worse case situation. Since there’s no contrast to focus on, and in the case of the dark test no light, the camera has to slow the AF processing down to the minimum limits. The dark test is preformed by covering the lens with it’s lens cap, and blacking out the viewfinder.
The light AF test was preformed by attaching a piece of parchment paper over the front of the lens hood. This is aimed to present the camera with a signal, thus preventing the camera from being stuck in slowest mode, but not one that can be resolved into an in focus image.
The second test will be focusing on reasonably high contrast target under mid-afternoon daylight conditions at a distance around 40 feet. This test will be looking at the performance of the AF system in snapping to a target that can actually be focused on. For this test, I also adjusted the focal length so that the subject under the AF point retained the same angular size.
Before I present the numbers, I want to reiterate that this is a very crude test done in my back yard, under natural light and with varying conditions.
I recorded the data by recording the distance scale with a GoPro set to 240 FPS mode, then counted frames in the resulting video in premiere from the frame before the scale started moving to the frame after the scale stopped moving. Beyond that though, the test was very unscientifically done once for each of the test cases, and it was shot outside under scattered cloudy skies so the light of both the test target and my light hunting test case are not inherently identical.
Finally the test was done with an EF 100-400mm f/4.5-5.6L IS II USM as the test lens on an EOS 5D mark IV. I chose this since my primary interest in this teleconverter is because of the AF capabilities at f/8, I wanted to see how things played out there. I may in the future go back and repeat these tests with my 70-200mm f/4L which would keep everything at f/5.6 and faster.
|11.72 s (4.63x)
|0.73 s (1.03x)
|0.13 s (1x)
|0.36 s (1.21x)
|11.08 s (4.38x)
|0.71 s (1.01x)
|0.2 s (1.55x)
|0.35 s (1.17x)
The table above summarizes the test results.
I don’t have a tremendous amount of confidence in this numbers, since I only conducted one reputation of each test. Therefore errors in my judgement in the analysis or other potential sources of error during the test (different lighting due to a cloud, etc.) will have a disproportionally large impact than over many tests that are averaged.
In the vast majority of cases, with the exception of focusing from infinity to a subject ~40 feet away the TC III focused faster than the TC II. In this one instance, I the apparently lacking performance of the TC III is likely a result of changes in testing conditions or a measurement error.
All told though, the AF performance of the mark III, is better than the AF performance of the mark II.
I’m of the opinion that statistically relevant objective image quality measurements are an important part of any lens review. Of course, then I go on to weasel out of having any and throw excuses like how expensive it is to do that.
For just a regular old lens, to get a statistically relevant sample size for testing you need to have 10s of lenses. At $1000-2000 a lens in many cases that’s $10,000-20,000 in just testing costs. Which, quite frankly is way out of reach of my budget, and apparently out of the budgets of most places doing lens tests. The one exception to this is Lens Rentals.com, since they test their entire stock of rental lenses and that reaches numbers of copies that is a statistically relevant size.
However, for a teleconverter, the testing is just as simple as plunking down a lens in front of a test chart and being done with it. Teleconverters don’t work on their own, they need to be attached to a lens. This means that you may get appreciably different results from different lenses depending on a number of factors (such as how well the TC and lens are matched, and just what the optical characteristics of the lens is to start with).
That said, there’s one saving grace here. Currently the Extender EF 1.4x III (and 2x III) is the only new manufacture teleconverter made by Canon. If you’re buying a first party Canon TC (and based on my experience with 3rd party TCs in the past, I only buy first part Canon TCs now), there’s no other choice.
Fortunately, image quality from the Mark IIIs are better than their predecessors. Moreover, when paired with the supported lenses, the image quality holds up very well.
As with any other teleconverter, because they increase the focal length of the lens without being able to affect the diameter of the aperture, they result in the combined lens’s aperture being a smaller f-number. For the 1.4x teleconverter, the resulting combination has an aperture 1-stop lower, so an f/2.8 lens becomes an f/4 lens, and so on. For the 2x teleconverter, the change is 2 stops; making an f/2.8 lens an f/5.6 lens.
For a first party teleconverter, Canon doesn’t offer any other choices. As a result, there’s not a lot of reason to avoid the Extender EF 1.4x III (or the EF 2x III) if you’re looking for a TC. Optically their better than the second generation ones that they replaced. Moreover when paired with a modern Canon camera, these 3rd generation extenders enable the broadest AF coverage possible.
There are of course 3rd party options, Sigma, Tamron, FotodioX, Kenko, and Yongnuo, to name a few, make EF compatible teleconverters, and many of them are a fraction of the cost of the Canons. That said, in general when it comes to optics, you generally get what you pay for. In my experience at least, you’re better off not using an extender and cropping in tighter in post than you are using a bad extender.