A few months ago I used two images of the same scene to compare how well the Canon EOS R images at 30 megapixels would print, compared to the 50 megapixel Canon EOS 5Ds R camera. For these tests, the results of which you can see in episode 660, I had the EF 24-105mm lens on my 5Ds R and the new RF 24-105mm lens on my EOS R. I was surprised to see that the image quality from the EOS R was so good and that that lower resolution images could actually be printed as large as the higher resolution images and actually looked a little better!
Since doing these tests though, I started to wonder how sharp the EF 24-105mm F/4 Mark II lens would be when mounted on the EOS R using the Canon Control Ring Mount Adapter, so this week, I’ve done a few tests, pitching the EF 24-105mm lens against the RF 24-105mm lens, but this time both on the Canon EOS R. A few people have asked me about this as well, and I’m sure there will be some people out there that are using the EF 24-105mm lens via the mount adapter and wondering if it’s worth switching. Value and worth are very subjective, so I’m not going to tell you what to do, but the results of my tests will give you everything you need to know about the difference in sharpness so that you can make up your own mind about this.
About the Tests
For the test, to keep things equal between frames, I set up my camera on a tripod and photographed a printed test chart attached to my whiteboard. I turned Image Stabilization off for all of these images so that it didn’t wiggle anything around as the exposures were captured and I shot three sets of images with each lens. The first set that we’ll look at was shot at 24mm, the widest focal length of these lenses, then the second set at 50mm, almost in the middle, and a final set at 105mm, which is the longest focal length of these lenses.
Below is an iPhone photo of my set-up, as I shot the 105mm focal length set. You’ll see that I was also using a studio light with a softbox to light the test chart. I left the ISO at 100 and the shutter speed at 1/200 of a second for all of the images, and I adjusted the exposure by making the studio light brighter as I increased through the aperture stops. This way we are able to rule out ISO and shutter speed as a reason for any differences between the images because these settings were the same for all 36 test shots. It also helps us to rule out camera shake, because the studio light’s burst is way faster than the shutter speed, so I’d literally have needed to swing the camera around on its strap to get any camera shake. OK, so a slight exaggeration there, but you know what I mean.
Note too that I moved the camera back and forth so that I could just about fill the frame with the test chart at all three focal lengths. The RF lens has a slightly shorter minimum focus distance than the EF lens, so I had to pull back slightly from the minimum focus distance of the RF lens to ensure that the EF lens could focus on the chart.
EF 24-105mm Lens @ 24mm (Center)
So, here first are six images which cover the entire aperture range in full stops using the EF lens, so we start wide open at f/4, then move through f/5.6, f/8, f/11, f/16, and down to the smallest aperture for both lenses, which is f/22. This first batch is shot at the widest focal length which is of course 24mm.
All of these images have been cropped down to 1440 x 960-pixel images at 100%, so if you ensure that your browser window is wide enough and then click the images, you will be able to see the image at its full resolution.
You’ll notice that the EF 24-105mm lens is pretty sharp at 24mm from wide open at f/4 through to f/16, and you will probably be able to detect just a little bit of softness caused by diffraction as we stop down to the smallest aperture of f/22.
RF 24-105mm Lens @ 24mm (Center)
Here now are the same six apertures but using the RF 24-105mm f/4 lens. Again, open them up in the lightbox to view the difference. Note that the images aren’t aligned perfectly because the EF 24-105mm lens actually has a slightly longer focal length, despite them sporting the same numbers, and it is really not that important for the images to be aligned for this test.
I think you’ll agree that at 24mm the images from the RF lens are incredibly sharp from f/4 to f/16, and there is a little bit, but slightly less diffraction kicking in at f/22.
To enable us to make a direct comparison, here now is another group of images, which are the same images that we looked at above, but I’ve alternated the EF and RF lenses so that you can make a direct comparison.
At 24mm though, I think it’s safe to say that there is very little difference between the EF and the RF lenses. They both perform admirably throughout the entire aperture range.
EF 24-105mm Lens @ 50mm (Center)
Here now is the EF lens at 50mm, again, showing the center of the image, just from further away, and zoomed in to 50mm.
I was actually surprised by how soft the EF lens is wide open at f/4, and I think I can also detect diffraction starting to kick in from f/11 and very slightly worse through f/16 to f/22. It’s not a huge amount, but usually, it’s the extremes of the zoom range that suffer, not the middle of the range, so this was surprising for me.
RF 24-105mm Lens @ 50mm (Center)
Here too is the RF lens at 50mm, for comparison.
Once again I think you’ll agree that it’s sharper throughout the entire aperture range, and there is just a tiny bit of diffraction at f/22, but otherwise it’s tack-sharp.
EF 24-105mm Lens @ 105mm (Center)
Now let’s move on to the longest focal length of 105mm. The long end of a lens is usually where the image quality suffers the most.
Again though, the EF 24-105mm breaks the rules, as it is tack-sharp throughout the range, with just a tiny bit of diffraction at f/22.
RF 24-105mm Lens @ 105mm (Center)
And to finish this group of images showing you the image quality at the center of the lens, here is our RF lens at 105mm.
I was very surprised to see that at 105mm when the lens is wide open at f/4, it’s a complete mess. The first image in that batch is so bad that I ran my tests again, thinking that I’d made a mistake, but the results were identical, so it’s official. At least my copy of the RF 24-105mm lens is crap wide open at its longest focal length. I’m pleased it’s as sharp as it is elsewhere and when stopped down, but I need to keep in mind to stop this lens down a little when shooting at 105mm to avoid that soft spot. In fact, I’m going to seriously consider having Canon take a look at this lens while it’s still under warranty.
Bottom Left Corner Comparison
In this next set, I have cropped out a 1440 x 960-pixel section of the image to see how the image quality fairs in the bottom left-hand corner. This is a great way to see how much the image quality degrades as you move away from the predominantly much sharper center of the lens.
EF 24-105mm Lens @ 24mm (Corner)
I’ve stuck to the same groupings, starting with the EF lens at 24mm through all of the full aperture stops.
I can see a little bit of color aberration or fringing at f/4, and a bit less but still see it at f/5.6, but it pretty much clears up by f/8 and the image doesn’t really degrade much again, even at f/22, where the diffraction is almost undetectable.
RF 24-105mm Lens @ 24mm (Corner)
The RF lens is up again next, through the same full stop aperture range.
This is again pretty amazing, especially for the corner of the images with the lens wide open. The sharpness is there until a tiny bit of diffraction kicks in at f/22 but I really can’t see any fringing or color aberrations to speak of, so that’s great!
EF 24-105mm Lens @ 50mm (Corner)
The EF lens starts off pretty nasty at 50mm not really sharpening up in the corners until f/11, but then it stays sharp and doesn’t really suffer from diffraction even at f/22.
RF 24-105mm Lens @ 50mm (Corner)
At f/4 on the RF lens at 50mm it’s a bit soft again, but it sharpens up nicely from f/5.6 and then just has a touch of diffraction at f/22.
EF 24-105mm Lens @ 105mm (Corner)
Despite the amazing performance of the EF lens in the center at 105mm, it doesn’t really settle down until f/11 in the corner, but then it’s nice and sharp again and diffraction isn’t an issue.
RF 24-105mm Lens @ 105mm (Corner)
I’d have been amazed if the RF lens was sharp in the corner at f/4 when zoomed in to 105mm after the poor image quality that we saw in the middle, but it actually doesn’t drop off very much at all, which is a bit of a bonus. It does sharpen up nicely at f/5.6 and shows only minor diffraction at f/22.
OK, so I hope you found this useful if you are interested in the difference between these two lenses when both are used on the EOS R. Note that there was no processing done to the images other than the default sharpening that Capture One Pro applies to all images.
The RF 24-105mm is definitely sharper overall, but with a surprising result at f/4 when zoomed right in to 105mm. I’m also thinking that the EF 24-105mm lens gave some very solid results, showing that it’s a top-class lens even when used with a mount adapter on the new RF mount camera system.
I personally made my decision to buy the new RF 24-105mm f/4 lens, based not only on my expectations that it would be a little sharper, but also based on the size. As I mentioned in my review of the EOS R, and as you can see in the below image, the EOS R with the RF 24-105mm is significantly smaller than the 5Ds R and the EF 24-105mm lens. When you mount the EF lens with the mount adapter, it greatly increases the overall size of the system, and it was important for me to keep this all at a minimum.
If you would like to test any of your own lenses like this, the chart I was using is from the Cornell University and you can download a copy of the PDF here. And indeed, if anyone with an EOS R and the RF 24-105mm lens decided to run these tests, please do let me know if you see the same poor image quality that I see at f/4 and 105mm. I’d be interested to hear what you find.
One of my biggest concerns about switching to the Canon EOS R mirrorless camera from my 5Ds R was the reduction in resolution, and how this might affect my large format prints, and I know I’m not alone in this, so I ran some tests, to see if the EOS R could keep up with his big brother.
As I mentioned in my review of the EOS R back in February 2019, during my Japan Winter Landscape tour I shot a pair of images of exactly the same scene, one with my EOS R and one with my EOS 5Ds R, so that I could evaluate various aspects of these images. I was very pleased to see that there seemed to be a stronger core of sharpness in the 30-megapixel EOS R images, compared to images from the higher resolution 50-megapixel counterpart, the EOS 5Ds R.
Because of the outstanding image quality, I went on to photograph the rest of my landscape tour and both of my wildlife tours almost exclusively with the EOS R body, only reaching for the 5Ds R when I needed to use two bodies at the same time. I still love the 5Ds R camera, but the EOS R has a much wider coverage of focus points as well as other important features, it’s more fun to shoot with, and the lower weight is a welcome bonus.
The thing that I was still not sure about though, is what we’re going to look into today. How do the EOS R images stand up to being printed large? One of the major benefits of the 5Ds R is that those beautiful large 50-megapixel images can be printed really big without the need to upsize them using a third party product like onOne Softwares Perfect Resize.
Large prints have played a big part in my business, and there have been some jobs that I’ve done over the last few years that I thought would not have been possible without 50-megapixel files, so it’s really important for me to know the limits of the 30-megapixel EOS R images. I have to add that I do not know how much of my findings would be relevant for a system such as the EOS 5D Mark IV, which has the same sensor, but does not use the new RF lenses, and I think it’s the RF mount that has more bearing on my findings than the megapixels, as I’ll explain.
In my tests, my main objective was to compare the EOS R images with the higher resolution EOS 5Ds R images to evaluate mainly the sharpness. The two photographs were shot within a minute or so of each other, using the same tripod, with the same settings. Due to variances in either the camera or the brackets and plates I used to attach them to my tripod, the EOS R image is slightly rotated clockwise compared to the 5Ds R image, and perhaps due to a change in the light between shots, or more likely just differences in how each camera processes its images, I also had to process them slightly differently, mainly with the Levels slider, and even then, the 5Ds R image doesn”t have as deep blacks as the EOS R image, but these things don’t really affect my tests.
I based my tests on three print sizes that I make a lot, both for personal purposes and to sell or display. I started by printing the entire image, without any cropping, at 18 x 24 inches. This is my regular test print size, and I apply my Fine Art Borders, meaning that the actual print area is 20.4 inches wide. This means if your largest print size is 13 x 19 inches, this first pair of prints are slightly larger than what you’d get printing borderless.
Here is a photo of each print, just laid on a table in my studio, with each side weighed down with a steel rule, to keep them relatively flat. I’ve uploaded these at relatively high resolution, so click on them to view the larger image to appreciate the detail. Or subscribe to our MBP Pro membership, and download the eBook for this post, to see the highest resolution images. You can probably tell even from these images, that the print from the EOS R image is actually slightly sharper than the 5Ds R image.
I made all of the prints directly from Capture One Pro, as that’s how I do most of my printing. Here is a screenshot of my settings for your reference. As you can see I have the Sharpening slider set to 25. This is the generic setting and unless I’m printing a soft image that needs some help, I just always leave that at 25. The resolution is set to 600 ppi automatically when I select the Highest resolution in the print drivers, even if I start off with Auto selected in the Resolution pulldown.
You can also see the width of the cell that holds the image is set to 20.4 inches, as I mentioned earlier. I’m printing with my own ICC profiles on Breathing Color’s Signa Smooth 270 fine art matte media. Matte is generally not as sharp a media as gloss, but it’s what I prefer to print on, and in my opinion, the better way to evaluate a print from a fine art perspective, and that is always my ultimate objective.
From the pixel width of my base images, we can calculate how much native resolution each image has. The EOS R records images at 6720 pixels wide, which means at 300 ppi (Pixels Per Inch) we could natively print the image up to 22.4 inches wide. At 20.4 inches wide, our base resolution that we are working with is 329 ppi. I generally set my printer to the highest resolution it will work with, but I’m looking for 200 ppi or more when possible in my base image.
The EOS 5Ds R creates images that are 8688 pixels wide, which we can calculate gives us images up to 28.96 inches at 300 ppi, and at 20.4 inches, the size of this print, we have 426 ppi, so that’s very respectable. But, the quality of the EOS R image is so much better, that the print from the smaller image is actually sharper. I had pretty much expected this based on visually comparing the base images, but it was nice to see this come through in the print.
24 x 36 Inch Prints
The next size that I make a lot of prints at is 24 x 36 Inches. Again, using my fine art borders, the actual width of the printed area for this size print would usually be 32.6 inches. To save paper, instead of making two 24 x 36-inch prints, I printed them out at 36 x 10 inches, so that I could just check the sharpness of the central band of my images.
Here is a photograph of the twigs to the right of that central large tree, for each print. On the left is the EOS R image print and the right is the 5Ds R print. As you can see, even in a 36-inch fine art print, the EOS R is slightly sharper, and this is without any additional processing. The settings are the same as I shared above, but the page size has been changed. The Sharpness slider remained at 25 for both of these prints.
Resolution-wise, we can calculate that the EOS R image at 32.6 inches wide would have been printed at a resolution of 206 ppi, whereas the 5Ds R would have 266 ppi at this print size. Of course, Capture One Pro is doing some processing, because it’s pushing the images to the printer at 600 ppi, but that is all happening behind the scenes, and with the same processing being applied to both images. Note too that I shot these images of my prints handheld on an overcast afternoon at f/4 and an ISO of 1600. Just keep that in mind as you look at the images.
44 x 62 Inch Print Test
The next test I wanted to do was to see how the EOS R would hold up to my largest generic print size, which is 44 x 62 inches. This is the largest print I can make on my Canon imagePROGRAF PRO-4000 printer of non-panoramic, native 3:2 aspect ratio images. With my borders, the actual printed area width is 55.8 inches. I really didn’t want to make two prints of this size just to check the sharpness, so I did a bit of math to figure out how to do this still printing on 36-inch roll media.
32.6-inches is 58.4% of 55.8-inches, so if we resize our images to 58.4% of 32.6-inches which is 19.05-inches and print them at 32.6-inches, we are essentially printing at the same resolution than we would be if were printing the uncropped image at on the full target size of 55.8-inches.
Here’s a screenshot of the resize process for the EOS R image. As you can see I set the resolution to 206 ppi, which I calculated by dividing the pixel width of my EOS R image 6720 by 829.1 mm which is 32.6 inches.
Without setting the Resolution the crop size readout is inaccurate, so the recipe resolution is important. Once set, I just resized the image to 19.05 inches, and we’re ready to print. For the 5Ds R image, I did exactly the same but with the resolution set to 266 instead of 206 to compensate for the higher resolution of the base image.
I’m pretty sure this math is good, but let me know if you think otherwise. There may be a better or easier way to do this, but for someone that came bottom of the class in math, if the result is accurate, I don’t care how I get there.
From the new sizes, we can also calculate that the base resolution of each image for this largest size print is now 120 ppi for the EOS R file and 156 ppi for the 5Ds R file. In the past, I wouldn’t dream of printing something that drops below 150 ppi, but as you are about to see, the EOS R makes that possible.
Here are my two prints from the cropped images. These are the same resolution that I would have got if I had printed the un-cropped images at 55.8-inches wide. By the time I got this far in my testing, it was too dark to shoot my prints by window-light, so these final images are shot using a ProPhoto studio strobe in a softbox, from camera-right.
Here again is a pair of images for comparison, with the EOS R image on the left, and the 5Ds R image on the right. Once again, I think you’ll agree that even when pushed to the size of a 44 x 62-Inch print with borders, the EOS R has a slight edge.
As I mentioned a moment ago, once the base resolution of the image I’m printing drops below around 150 ppi, I have always pretty much automatically reached for ON1 Software’s Perfect Resize, and upsized my image to ensure that I get a nice crisp print, so I tried one last test, upsizing the EOS R image to 300 ppi, and I made one last test print, that you can see here. The quality does improve slightly, so for a print of this size, I will probably still upsize the image in Perfect Resize.
Let’s keep in mind that we are looking at photos here that are essentially mimicking the photographer’s habit of putting our nose to the print to see if it’s sharp. From a regular viewing distance, you really cannot tell the difference between the upsized version and the native resolution print.
Let’s do one final comparison here, with a before/after slider. This is the 55.8-inch print that was not upsized, on the left, and the Upsized 55.8-inch print on the right. I think you’ll agree that upsizing helps to improve the image quality, but even without it, I am surprised that the EOS R with 30-megapixels, can be printed without upsizing and still be this good.
Of course, there are still benefits to having more megapixels. Even bigger prints will still benefit from more pixels, but based on what I’ve found today, this doesn’t concern me as much as it did, with the technology we now have in Canon’s new mirrorless camera, and its accompanying RF lenses.
The other thing is the ability to crop. Sometimes I make a decision to crop an image to get the framing I want, and although I don’t like doing that, when you have 50 megapixels, you can crop away a chunk and still have plenty to play with.
So, I’m still looking forward to the rumored 5Ds R Mark II that will likely also be mirrorless, and at least higher resolution than 50-megapixels. As long as the ISO performance remains good and the frame-rate respectable, I’ll be all over that. I am now much happier that I have already shot 16,000 images with my EOS R this year, and having now sold both of my 5Ds R bodies and bought a second EOS R for my upcoming Namibia Tour, I feel much more confident that my images can be used for pretty much anything I can currently create.
Before we start to wrap this up, I should also mention that I am not using the Dual Pixel Raw feature on my EOS R. It not only imposes a number of restrictions on your shooting, but you also have to use Canon’s Digital Photo Professional software to develop the resulting images, and neither of these things i acceptable to me, so I won’t be using it.
My findings today have far surpassed my expectations and knowing the sort of results I used to get from my 22–megapixel images, I believe the quality that I am seeing in the EOS R images is more attributed to the new architecture of the Canon RF Mount and mirrorless camera system.
The lenses are newer and more advanced, and the back of the lens is much closer to the sensor. With the EF System the lens was 44 mm from the sensor, compared to just 20 mm with the RF System. This must be preventing the light from spreading out as much before it is recorded by the sensor. I haven’t found anything from Canon to support this, but I did find a white paper on the RF system that attributes the shorter distance to improved image quality.
Now, positioning of large diameter lens elements much closer to the image sensor (especially the full frame sensor) would support an important enhancement of image quality.
I started these tests hoping to be impressed, and I was frankly blown away by what I found. Every time I went to the printer to cut the last print from the end of the roll, the hair on the back of my head stood up, and I found myself chuckling as I held the prints up to the light to study the details. I have been excited about the EOS R since first shooting with it in earnest during my winter tours this year, but as I’ve mentioned before, it’s the RF mount that sold me on the system. The EOS R is a great camera, but it’s only the start of an exciting and entirely new system that I can’t wait to see develop.
Now that I know that I can print my images from this system at least as big as I have been with my higher resolution 5Ds R cameras, I’m happier than ever with my decision to move to mirrorless, and even more happy that I decided to wait for Canon to make their move in this field of photography. They have done what I had hoped and taken this opportunity to not just jump on the band-wagon, but to innovate and evolve, or maybe I’d go so far as to say reinvent their interchangeable lens camera system in the process.
Today we’re going to pitch the old Canon EF 85mm f/1.2L II USM lens against the new Canon EF 85mm f/1.4L IS USM lens, to see if it’s worth trading in the old lens for some shiny new glass, even with one-third of a stop smaller aperture. I also hope that this will give you an idea of the quality of the new lens, regardless of how it compares to the old lens.
The angle I’ve taken for this review is that an L lens from Canon lens being tack-sharp when stopped down to say f/5.6 or f/8 is pretty much a given these days, and therefore not really worth spending much time on. Conversely, the beauty of Canon’s wide aperture lenses is that you can shoot with them wide open and still expect great results. However, the depth of field that you get with these lenses at f/1.4 is incredibly shallow, and so my test shots are mostly wide open, both to show you how they perform, but also to show you what you can expect in terms of depth of field should you shoot with them wide open.
A Bit of Background
To give you a bit of background on my use of the 85mm Mark II lens, I bought mine back in 2007, one year after it was released, and have used it on and off ever since. I must admit, there have been long periods when I simply didn’t use it, mostly because much of the work I’ve been doing has been overseas, and for many years my kit was so heavy that it really wasn’t possible to take any more lenses, so it stayed at home. I actually have a rule that if I don’t use a piece of gear for more than a year, I sell it, and have over the last eleven years sold many lenses and old bodies as I’ve upgraded to new versions or bought new kit. But, despite not using the old 85mm f/1.2 lens, I could never bring myself to sell it. It was just that good.
It wasn’t perfect mind; for example, the autofocus was incredibly slow. Probably the slowest autofocus on any lens I’ve ever used. Not quite as big a deal, but when I heard that Canon was going to release an updated 85mm lens, I was happy to note that it had Image Stabilization. I can’t recall now if I had any information on autofocus speed improvements, but at this point in time, I couldn’t imagine Canon releasing a lens without really snappy autofocus, so I ordered a copy of the 85mm f/1.4 lens as soon as it was announced last year.
You might wonder why I bought another 85mm lens when I wasn’t using my old one very much, but the main reason is that this is a great focal length for portraiture, and I do most of that these days on my tours overseas, and quite often in dark places, such as down a well or in a dark adobe building in Morocco last year, or of the Himba people in their huts in Namibia, and even though I might stop the lens down a little bit for greater depth of field, you can simply see more through a wide aperture lens, and of course, it is nice to be able to open the aperture up a little when necessary too.
The other big factor is that the weight of the gear that I’m traveling with is less now than it has ever been. I’ve replaced a number of lenses with fewer lenses, and they are pretty much all smaller and lighter, meaning that I can do most of my overseas travel now with an 18L backpack, and even that has plenty of room for an extra lens, which is why and how I was able to pack the new 85mm lens for my recent Namibia tour.
Another very important factor for me is that the 85mm f/1.2 lens is not weatherproof. Many people think that all L lenses are weatherproof, but that’s not the case. The new 85mm f/1.4 lens, however, is weatherproof. I wouldn’t like to take a lens like this to Africa, be it Namibia or Morocco, without weatherproofing, as the dust and sand get just about everywhere. I’m careful with my gear, but when it comes to getting a shot, I don’t want to be worrying that I might get dust in my lens, or moisture if it’s raining. This, in fact, is the only thing that I dislike about my 5Ds R bodies; they aren’t weatherproof. They have some weather sealing, but it is not like the Canon 1 series bodies that you can literally hose down if necessary.
I am also pleased to report that the Autofocus on the new 85mm f/1.4 lens is incredibly fast. It’s what you’d expect for a prime lens. You don’t even notice any lag when you press the AF button; it just snaps in immediately. This was really nice to see and removes the only frustration that I had with the old 85mm lens. I’m also happy to report that I have not noticed any focusing errors, such as the dynamically shifting back-focus tendencies of the Canon 50mm f/1.2 L lens that I wanted to love, but not with the problems caused by its by-design crappy focusing. There’s none of that happening with the 85mm lenses.
Before we jump in and look at the results of some of my tests, let’s compare a few of the key specifications of the old and new 85mm lenses. Firstly, the original 85mm f/1.2 lens was released in 2006, so it’s twelve years old at the time of writing (July 2018) compared to the new 85mm f/1.4 lens which was released at the end of 2017, so I’ve now owned the new lens for six months. I would have liked to have done this review sooner, but my winter tours had a hold of me, and I didn’t really get a chance to use the new lens until my Namibia tour last month.
Closest Focusing Distance
Another problem that I had with the old 85mm lens was that it’s closest focusing distance was pretty long at 95 cm or 3.2 feet. I was happy to see that the new lens focuses 10 cm closer at 85 cm, or 2.79 feet, but I must admit I’d have really liked to see this brought in a little more. The 10 cm is better than nothing, but when we consider that lenses like my 24-105mm f/4 lens can focus as close as 45 cm or 1.7 feet, almost half, that’s the sort of distances that you get used to. Of course, I’m sure Canon would love to make the minimum focus distance closer too, but I’ll bet it’s the super large apertures that make it physically impossible for them to improve on the current design.
In practical terms, the difference between the minimum focus distance of the old and new 85 mm lenses doesn’t give us the ability to magnify our subject by very much more. Here, in fact (below) is a photo of my old Canon AE-1 camera shot with both lenses at their closest focus distance. The smaller camera is from the old 85mm I measured it as taking up 38% of the frame, compared to the larger semi-transparent camera, from the new 85mm at it’s closest focus distance, which takes up 45% of the frame, so we’re talking about a 7% increase in magnification between the two lenses closest focus distances.
In fact, as we’ll see shortly, the new 85mm actually magnifies the subject slightly more than the 85 mm f/1.2 Mark II did, so not all of this 7% is coming from being able to focus at a closer distance, but let’s look at some more specs first.
As you can see in the next image (below) the new 85mm lens, at 4.1 inches or 105.4 mm, is slightly taller than the old one, at 3.6 inches or 91 mm. But the new lens weighs 75 grams less, at 950 grams, compared to 1,025 grams, or 33.5 ounces compared to 36.2. That’s not a huge difference, and the new lens is still a pretty hefty chunk of metal and glass, but any reduction in weight is welcome as it becomes more and more difficult to fly overseas with our gear.
Another nice advantage of the new lens while we’re comparing their looks, is that the new lens has a much shallower hood, making them in fact almost the same length with the hoods attached, as you can see in this next image (below). The old hood had two buttons to release it and just clipped into place without any twisting, while the new hood has one locking button, but is a twist action.
The lenses have completed redesigned optics of course, with the old 85mm f/1.2 II lens made up of 8 elements in 7 groups, compared to the new lens with 14 elements in 10 groups. Another nice bonus for me at least is that the new lens has a 77mm filter thread, which is the same as my other lenses, and therefore allows me to carry fewer filter variations. The old lens has a 72mm filter thread and is now the only lens that I own with this thread size.
In case you are wondering just how much bigger the lens elements of f/1.2 aperture lens is compared to the f/1.4 model, here is a photo showing them both from the end, and it’s easy to see that the old model is a fair bit wider. If we do the math, literally dividing 85 by 1.2 we find that the old lens had to have at least a 70.8 mm opening for the light to travel through, and for the new one, dividing 85 by 1.4, we get 60.7 mm, so that third of a stop reduction in the aperture saved canon 10.1 mm in the diameter of the lens elements.
According to my tests, I’ve found that the new 85mm is actually a slightly longer focal length than the old f/1.2 Mark II model. I measured the distance between the two sharpest lines of text in this photo of an open book, and found that it was 69.5% the width of the frame in the new 85mm f/1.4 lens, compared to 66.8% of the frame with the old 85mm f/1.2L lens, and both images were shot with the camera on a tripod, not moving at all between the two photographs. That’s a 2.7% magnification in the new lens over the old one.
To see the difference for yourself, click on the images to open them in the lightbox, and then navigate back and forth with your mouse or swiping on a tablet. If the fades transition makes it difficult to tell the difference, feel free to save the images to your desktop and flick back and forth between them on your own computer. Now, although I know that this kind of variance in the spec of our gear annoys some people, personally, I don’t really care about such small variances, but I wanted to point it out so that you know what you are getting if this is important to you.
0.4 Stops Darker when Wide Open
As you flick back and forth between these two images, you’ll also notice that the new 85mm lens is also approximately 0.4 stops darker and has a stronger vignette than the old lens at the same aperture. I should mention though, that if this vignette bothers you at all, it’s easily removed in post-processing. In Capture One Pro, both lenses have lens profiles available, and if I add 100% Light Falloff correction, the vignette disappears and in fact, the entire image looks very similar to the old 85mm lens images in terms of overall brightness.
Evens Out Stopped Down
I should also mention that the darkness that we see in the f/1.4 lens over the f/1.2 lens is only really noticeable from around f/2 and wider apertures. From f/4 there’s hardly any difference, and from f/5.6 it’s hardly noticeable at all, as you can see in these two images. Grab the vertical bar in the middle of the image and slide if left and right to compare the two images.
I actually shot the photos of the book so that we could take a look at a 100% crop of an image from each of these cameras to see how sharp they are wide open. Because the old lens goes a third of a stop wider than the new 85mm f/1.4 lens, I shot both images at f/1.4 so that we’re comparing apples to apples. You should see a handle over the image that you can slide from side to side to directly compare the images from each lens. Again, the camera was not moved, I just switched the lenses out and focused on the same word.
I can see a slight cyan tint in the text above the sharp area, and a slight magenta tint in the foreground text, on both images. The line that is sharp is definitely sharper in the new 85mm f/1.4 lens image though, despite this being shot completely wide open at f/1.4, and that’s pretty impressive. Also, just to reiterate what I said at the start, you can easily see from this image just how shallow the depth of field is when shooting at f/1.4.
I shot the X-Rite Digital ColorChecker SG card with both lenses as well, and so that we can continue to get an idea of the difference in the lenses wide open, I set both lenses to f/1.4, and as with the previous photos of the open book, I was using a studio strobe to light the target, so the light source was the same for both images. You should be able to see a handle in the middle of the image (on the blog) that you can drag from side to side to check the difference between the two images. I have turned on the Light Falloff correction for these two images too, so that you can see how much difference that makes. You’ll probably be able to tell that the f/1.4 lens is still very slightly darker, but there is no real difference between the color with either lens.
I have cropped the f/1.2 lens shot slightly so that they are approximately the same size, but you will also probably be able to see that there is a bit of barrel distortion in the old lens, and there is actually just a tiny bit of pincushion distortion in the new 85mm f/1.4 lens. So neither lens is perfect, but there’s definitely less pincushion distortion in the new lens than there was barrel distortion in the old one.
Scary Dislodged Lens Elements
One thing that also I’d like to mention before we start to wrap this up, is that when I got home from Namibia, as I unpacked my gear to clean it and put it away, the new 85mm f/1.4 lens was rattling as though it had been given a nasty knock and something had broken inside. I took both the front and back lens caps off and had a look through the lens, and sure enough, it looked as though one of the lens elements had broken free of its housing, and this scared the heck out of me at first.
I take out overseas insurance for my gear before I travel so it wasn’t such a big deal, but it doesn’t feel great to have a piece of gear break. To ensure that it was actually broken, I put the lens onto a camera body, and as I looked through the viewfinder and half-pressed the shutter button, the Image Stabilization kicked in, and I literally watched the dislodged lens element slide back into place. I gave the lens another shake, and it had stopped rattling!
I went online and found that it’s actually not uncommon for Image Stabilization lenses to come l0ose like this, especially after being transported around. And, when I consider how much my camera bag had been bounced around on some of the dirt roads in Namibia, it’s hardly surprising that the IS lens elements had come out of their locked position. I have not sent the lens in for repairs, as I don’t think it needs anything doing now, and in case you were wondering, yes, the test shots that I’ve shared today were all made after this incident, so if there is anything wrong with my lens, I can’t tell.
All in all, I’m very happy with the new 85mm f/1.4L IS USM Lens from Canon. It’s not one of my workhorse lenses, so it’s certainly a bit of a luxury for me to own, but I can tell you, it was really nice to be able to see through the lens so well inside the dark huts in the Himba Village last month, even though I stopped the lens down to f/4 for this portrait (and f/5 for the one above). The image quality is everything that I’d expect from a modern Canon lens and more. I hope you’ve found this review and comparison useful. If you decide to buy a copy yourself, please use our affiliate links if you buy from our friends at B&H, and that will help to support the Podcast and blog without costing you any more money.
It Worth Upgrading?
If you own the original f/1.2 Mark II lens, and you are wondering whether or not to upgrade, personally, I would. In fact, I did. As usual, this post is not sponsored in any way, and I received nothing from Canon or B&H or anyone else to enable me to create this review or otherwise compensate me for it. I bought the lens with my own money, at the full going price, so I’ve already voted with my dollars (or yen in my case).
I do honestly believe that it’s worth the upgrade, especially if you can get a reasonable part-exchange deal on your old lens. The image quality increase alone makes it a no-brainer for me, but the lighter weight, slight decrease in bulkiness, the weatherproofing and the addition of Image Stabilization are all very nice added bonuses. These things give me the confidence to take this lens anywhere, and hopefully continue to make some beautiful portraits with it.
I recently switched newsletter services from Campaign Monitor to MailChimp, and today take a look at some of the key differences between the two services. Now having experience with both, this isn’t going to be a MailChimp vs Campaign Monitor death-match, because they are both great systems. My intention today is to point out the differences so that you are in a better position to decide which one is for you, or at least which one to try first.
I know this isn’t directly photography related, but it very much is business related. I believe it’s very important that any business tries to build a mailing list to keep in touch with your audience and customers. The Internet is becoming a very crowded space, and we are all fighting for eye time, so it’s important to develop ways to stay in contact and interact with people that are interested in what we have to say.
People digest information in many ways. For example, although I release the audio recording these Podcast episodes each week, I know that some people still prefer to read. Some people come to the Web site and read rather than listen, and others subscribe with our RSS feed. Some people obviously subscribe to the Podcast directly in iTunes, and others listen via the various players and syndication applications that pick up my Podcast.
Having put in pretty much full day each week to make these Podcasts, I of course want to reach as many people as I can with them, and one of the things that we’ll look at, is the automatic delivery of my blog posts each week now, via MailChimp. Although not everyone is going to open the mail, I find that proactively pushing information to people’s mail inbox is a great way to augment what I’m doing with the Podcast and blog.
I also have people that sign up for Tour and Workshop information, and that gives me a way to directly alert people that are interested to new tours as information is made available. Before I started using Campaign Monitor a few years ago, I maintained groups of email addresses in the address book on my computer, but that can get messy and hard to maintain, plus the emails that I sent out were mainly just text and links, and didn’t look great. I also had to BCC people to protect their privacy, but that is a surefire way to get your email sent straight to the spam folder.
The services we’ll look at today enable us to create great looking email, that are less likely to be automatically marked as spam, so they are more likely to get read than a simple text email. I know a number of very successful pros that send out plain text email, and they really don’t do their content justice in doing so.
OK, so to start our comparison of MailChimp and Campaign Monitor, let’s take a look at the pricing. Both companies enable you to set up an account for free to get a feel for the system. MailChimp actually allows you to have a completely free account for up to 2,000 subscribers, and send up to 12,000 email per month without paying a penny, which is awesome, and a great way to get started. You only start paying for your account when you go over 2,000 subscribers, or if you want some of the extra features only available with a paid account that we’ll look at shortly.
Campaign Monitor pricing starts at $9 for a monthly account, with up to 500 subscribers, or you can use a free account, and pay $5 per campaign plus 1¢ for each recipient, so if you were to send a campaign out to 1,000 subscribers, you’d pay $15 for that campaign but there are no other charges. This is great if you just need to manage a list of subscribers, and don’t send out emails that often. If you will send out an email more than one a month though, a monthly plan probably makes more sense.
Both companies have a sliding scale pricing model, where you pay more as the number of subscribers to your newsletters grows. Campaign Monitor charges $9 for that first 500 subscribers and with their basic plan, you can send up to 2,500 email per month, basically sending to your entire list once every week. Once you go above 500 subscribers, you’ll jump to the $29 plan, which is good up to 2,500 subscribers and 12,500 email per month. If you need to send more email that the limit of these basic plans, you can select the unlimited plans, but these are quite a lot more money. The $29 basic plan for example jumps to $59 for unlimited emails. Check out Campaign Monitor’s pricing page for more information on larger volume lists.
MailChimp’s pricing starts at $15 per month for 501-1,000 subscribers but that is for unlimited sends out of the gate. The MailChimp scale is more granular than Campaign Monitor, increasing with every 500 subscribers. I actually switched from the $29 a month plan on Campaign Monitor, to the $30 per month plan on MailChimp, but because that now includes unlimited sends, it will actually work out cheaper for me, especially as I’ve now enabled automatic emails when I release new blog posts, as we’ll see later.
Another major difference between the two services is how they manage lists and subscribers. Campaign Monitor allows you to have the same person subscribe to multiple lists, but then when you send out an email, you can select multiple lists, and if the same person is in more than one list, they will only receive one copy.
I found this easy to manage, and it enabled me to post a subscribe button for my Tours & Workshops newsletter on my tour pages, and I could post another button for my general information newsletter elsewhere, and I didn’t have to worry about which list people were in.
MailChimp however has no linkage between mailing lists, and although you can put people in multiple lists, you can’t select multiple lists to send out an email to. Now that I have it set up, it’s quite intuitive, but there was quite a learning curve initially, especially having come from what I considered to be more intuitive functionality on Campaign Monitor, although ultimately MailChimp’s lists enable you to do a lot with their Segments and other add-ons that we’ll get to shortly.
Because there is no way to send to multiple lists in MailChimp, if I had kept my old structure, with one list for Tour information, and another list for General Information, if I wanted to send the same newsletter to both lists, I would literally have to send it to both lists, and people that are on both lists would receive two copies, and in doing that, I’d run the risk of people marking my newsletters as spam, which is obviously best avoided. At the very least, I’d probably see more people unsubscribing from at least one of the two email, if not both.
MailChimp uses Groups and Segments to achieve the same thing, so if you want to avoid sending multiple emails to the same recipient, you need to put them into a single list, separated into groups. Here (below) you can see how my main MBP Newsletter list is divided into General Information, Tours and Workshops, Pixels 2 Pigment and a few other groups.
MBP Newsletter List Groups
Then, to send email you create Segments. The cool thing about Segments is that you can create Segments based on various rules, so for example as you see here (below) I can create a segment just for General Information, and I can create a segment to send to both the General Information group and Tours & Workshops group, or any combinations of group in my list.
This way I can easily create larger groups to email when the information I’m sending out is likely to be of interest to more than one group, and of course, if the recipient is subscribed to more than one group they still only receive one copy of the email. I also have some other lists that people are added to, and I can merge lists easily as necessary.
Another benefit of the MailChimp system is that you can keep your monthly costs down by only having recipients in a single list, assigned to various groups, because if someone is subscribed to multiple lists, they are counted as separate subscribers, so you will jump up to the higher price bracket sooner.
Sign-up Buttons and Forms
To get people signed up, both companies provide various forms to use on your Web site, and both have an iPad app that you could use at events or exhibitions, and they both have a Facebook signup form.
In general, I think the Campaign Monitor forms are better looking and easier to embed in your web site. I found it took more time to get the MailChimp forms looking how I wanted them too, although the forms themselves do have to be more complicated to handle things like signing up for specific newsletter groups.
Once you have them working, both companies do the job at hand, but one area that Campaign Monitor has the edge is their simple popup button, which I could easily embed into a page, and when clicked would just show a name and email address field, to get the user subscribed quickly and easily.
MailChimp don’t have this, and it’s one thing that I really wish they’d work on. They have a popup style form, but they even call it the ‘evil popup mode’, probably because it’s is timed to display after so many seconds, which I find really annoying too.
I may well change this, but I’ve started to use a WordPress plugin called SideOffer with a custom background graphic to add a simple tab to the top right of the browser window, which when clicked opens a small window containing the form. I experimented with a number of ways to actually build the form, but the native MailChimp forms work best because they tap straight into the system, which enables them to check if people are already signed up, and if they are, an appropriate message is displayed and an email sent with a link to change existing subscription options.
You can of course also just provide a link to a simple page with a sign-up form, which works fine and it’s easy to customize the forms with your logo and group options etc. Again, these aren’t quite as pretty as Campaign Monitor, but they do the job.
Both companies enable you to set up autoresponders, which are emails sent automatically to new subscribers as they sign-up. Both companies enable you to fully customize the autoresponder template, so that the new subscriber receives an email with a similar look and feel to your newsletter and you can include your own welcome message and links to things that you’d like to draw to your new subscriber’s attention. Neither company offers an autoresponder with their free plan, which in the case of Campaign Monitor would be a pay per campaign plan, so if you want to set up an autoresponder, you have to pay for a monthly plan with either company.
Both systems have a huge array of templates for you to use and customize. Compared to the templates that I’ve been using until recently on Campaign Monitor, I’d say that MailChimp had the edge on customizability, but with the recent addition of Canvas templates in Campaign Monitor, there is probably now very little in it with regards to what you can create and the ease of use.
Campaign Monitor Canvas Template Design
I tend to find that MailChimp’s templates lend themselves a little more towards reuse. For example, in Campaign Monitor I found myself copying and modifying old templates or creating new templates quite a lot, but with MailChimp I’m finding that the templates are more flexible, so I can create what feels like a more higher level template, and use it more often for various types of email.
Both systems have code blocks that you add to your template or email as you write it, and these can be used to create great looking emails without a lot of effort. If it’s a pain to actually create an email, you’ll do it less, so ease of use here in paramount, and both do a great job.
Both companies also offer both a regular mail client preview and iPhone preview of your email as you complete your design, but MailChimp don’t do themselves any favors by messing this up. As you can see here, the Campaign Monitor preview has the iPhone version correctly scaled and responsive.
Campaign Monitor Email Preview
MailChimp’s preview however shows the images too large in the iPhone view, as we see here (below). In reality, the MailChimp email are also correctly scaled when viewed on the iPhone, but that isn’t correctly displayed in the preview making it feel more like a gimmick than a useful feature.
MailChimp Email Preview
Blog/RSS to Email
So far, with regards to most of the stuff we’ve touched on, Campaign Monitor is either neutral, or perhaps even does a better job than MailChimp in some ways, but one of the things that MailChimp does do better, is the RSS to email feature. Most blogs have what’s called an RSS feed, which contains either a summary of posts, or the full posts in a simplified format to be read by syndication readers such as Feedly, Reeder and Leaf etc.
This provides a way to create automatic updates, such as the RSS to Email functionality on both Campaign Monitor and MailChimp. Basically the content of your RSS feed is used as the body of an automatically generated email that is sent to subscribers of a specific list.
The reason I was never able to turn this on when using Campaign Monitor, was because it insisted on including the last four posts in my feed, and because I have the full post in my RSS feeds, this would result in a huge email going out to subscribers. All I wanted to send was the most recent post, and that’s exactly what MailChimp does. A very subtle but important difference, and one of the factors in my decision to switch systems.
This has enabled me to create an automatic campaign via MailChimp that looks at my blog’s RSS feed every day at 8pm Japan time, and if there is a new post, it creates an email and sends it out to everyone that has subscribed to receive Blog Posts Delivered by Email.
A Stronger Push
You might think that’s a stupid thing to do, because I want people to visit my site, right? Well, that’s true, but I also want people to be able to receive and consume my content in any way that suits them. Some people only subscribe to the RSS feed, read my content and never visit the site, which is fine. Likely some people subscribe in iTunes, listen to the Podcast during their commute, and never visit the site, which is fine too.
I spend a lot of time each week putting this content together, and the more consumption options I can provide people with, the better. Of course, too many choices can lead to none being selected, but in this case, the visitor is already looking at my newsletter sign-up form, or subscriptions page, because, hopefully, they’ve liked what they’ve found, and want to receive regular updates when new content is released.
The other benefit of an automated email is that it’s what I’d consider a stronger push than iTunes or an RSS feed. I use RSS in a program called Leaf, which syndicates all of the feeds that I’m subscribed to, and I can flick through hundreds of blogs, only looking at newly published content, and then have everything that I’ve read automatically disappear from view. It’s great! But I have to be in the mood to do a bit of browsing, and start Leaf up, and I only do that a couple of times a week.
With email though, I open it first every day. I’d hazard a guess that most people do. So when someone gets up and checks their mail before running out of the house to go to work, they’ve picked up my new post. That means that they can now read it on the train, or even listen while driving if they have a data plan, because the audio player works fine in some mail clients. With very little effort on my part to set this up, I have enabled a new method of consuming my content, and that has to be a good thing.
A/B Split Campaigns
Both MailChimp and Campaign Monitor can send what are called A/B Split campaigns. This is basically a way of testing what works better in your email, based on a different Subject line, from name or in the case of Campaign Monitor, different content in the email, or with MailChimp, different delivery times.
Campaign Monitor splits the campaign 50/50 which enables you to check the results of the entire campaign for future reference. MailChimp on the other hand has the ability to start with an adjustable percentage of the entire list, and then send the rest of the list recipients based on the winning mail style after a preset period of time. This is great, as you learn from the test, but then act on the results during the same campaign rather than having to wait until you send your next newsletter out.
MailChimp A/B Campaign Settings
Once you are ready to send your campaign, both companies allow you to send straight away, or schedule the newsletter to be sent at a set time. Another couple of nice features in MailChimp are the ability to send with Timewarp, that will send at a set time, say 9am, in your recipients’ timezones, and you can also allow MailChimp to optimize the send time based on click activity from previous campaigns, so that you have the best chance of people actually opening and clicking on links in your email.
MailChimp Send Scheduling
Once your mail campaign has gone out into the world, both systems provide comprehensive reports on how many people actually opened the email, and how many people clicked on a link in your email. I remember a conversation with the folks at MailChimp when I was at the Photoshelter Luminance conference in New York in 2012, and they seemed quite impressed when I told them that between 60 and 70% of my recipients open my email, but I had no way of knowing in Campaign Monitor whether that was good or bad.
MailChimp Reports with Industry Averages
In MailChimp, they actually provide industry standard statistics in your reports so that you can see how you’re doing without having to go searching for this information online, which is a big advantage to this system in my opinion. There are lots of other reporting features as well of course, breaking out your subscribers by country, age and gender etc.
MailChimp also has a cool add-on option called Social Pro, that collects data on your subscribers based on the social networks that they use, how active they are, and how many people follow them etc. This is useful because you can create Segments based on the information gathered, so I could for example send an email regarding something Facebook related just to people that I know are active on Facebook.
Social Pro costs an extra $1 per 500 subscribers per list, so it’s not a big deal to add, and depending on how you manage your lists you may not need to add this to all of your lists. I have a number of different lists that I actively use, but I only have Social Pro activated on one of them.
One of the main reasons I decided to give MailChimp a try recently, was the amount of community support I see for MailChimp compared to Campaign Monitor. Very often for example, a WordPress plugin will support adding people to a MailChimp list, but not Campaign Monitor. I’m not sure why this is, but it’s a trend that I’ve come across a lot over the last few years, and integration with my Web site is obviously an important consideration.
Both systems offer everything you need to manage mailing lists, easily create beautiful templates and emails to send out, and report on the results of your campaigns. I think MailChimp has the edge on the kind of information available in your reports, and I get the feeling that MailChimp are more interested in helping their customers to understand what they are seeing in relation to the bigger picture, and not just providing the necessary information to do your own research and analysis.
Having already been comfortable with Campaign Monitor, then spending a lot of time setting up MailChimp and getting used to the new system over this last month, my overall impression is that MailChimp takes more time to get up and running, but once that time has been invested, the system is a little more powerful. If you want to get up and running quickly, with a very capable system, Campaign Monitor may be the preferred choice.
Note though that whatever you decide, it doesn’t really have to be a final decision. You can export your subscribers from both systems, and import them to the other very easily, and both systems can be set up and seriously played with without paying a penny.
Anyway, I’ve voted with my pennies, and will be cancelling my monthly subscription with Campaign Monitor for now, but really, they are both great systems, and my hope with this comparison review is that it helps you to decide which one to go with if you need something like this yourself. I know it’s not directly photography related, so forgive me if you are totally uninterested, but if you are building a business, and don’t have a similar system in place yet, I hope this was useful.
If you should decide to use MailChimp, you can support this Podcast by using our referrer link: http://eepurl.com/ZyXaT
On March 19th, 2010, Canon released the updated version of their famed workhorse, the EF 70-200mm F2.8L IS II USM lens. I have spent the last three days taking mine through its paces, and today I’m going to share my findings.
I’ve been waiting for this update to the 70-200mm F2.8L IS lens for over two years. It was my workhorse lens since I bought my version one copy in 2006, and I was very happy with my results from it when I was using a 5D as my main camera. As soon as I upgraded to the 1Ds Mark III and then also shooting with the 5D Mark II, it became obvious that this camera was not resolving images quite enough to produce sharp shots when used wide open at F2.8 on these 21 megapixel cameras.
To compound the problem, at about the same time as I bought the 1Ds Mark III, I also bought a 300mm F2.8 L lens, which is as sharp as tacks, even when used wide open at F2.8. This is not an apples to apples comparison of course, as it is rare that a zoom lens will be as sharp as a prime lens, and the 300mm F2.8 is an exceptionally sharp prime lens. Still though, the timing of the purchase certainly affected how I felt about my old 70-200mm, and I found myself reaching for this lens less and less over the two years or so that followed.
Although I would usually part exchange my old lens when I buy a new one like this, I decided to keep my old 70-200 until I’d run some tests, so that’s exactly what I’ve been doing for the last three days. I thought it would be great if I could not only review the new version, but also I really wanted to compare it with the old one, to see if there really was a difference. This will also help those of you that current own a version one 70-200, and want to know if it’s worth upgrading or not. The definitive answer of course is, “it depends”. What I will say at this point is if you are already happy with the results you are getting with your current lens then don’t sweat it. There’s no need to upgrade if you are getting the results you want. The following review will just give you some additional information to help you make a comparison and weigh up the benefits against the cost if you are thinking about an upgrade.
What’s in the box?
The lens comes with a nice strong case, not the usual gray pouches that most lenses come with, which is nice for when you just want to throw it over your shoulder for more casual shooting about town for example. One word of advice about the case though, is attach an old camera strap and not the centimeter wide strap that comes with it, as that thing will cut your arm off from the shoulder if you walk around with it all day.
What’s in the box?
You’ll also find the usual lens cap, rear dust cap, instructions manual, warranty card and a nice newly designed lens hood, as well as the lens itself of course. The box itself is considerable bigger than the original 70-200s box, as you can see below.
New Box is Much Larger
Newly Designed Lens Hood
As I said, the lens hood design has been changed, and there’s now a little button that you have to press before you can turn the lens to take it off the lens, both when it’s attached backwards for storage, or attached in the normal shooting position. This is actually a great improvement in my opinion. I found that the old lens hood turned too easily, especially after a little wear, so it could either drop off, or turn slightly while you are shooting, and because of the shape of the hood, if it turned, it could theoretically give rise to vignetting. Although this never happened to me with the old hood, I have noticed it in a rotated position in the past after walking with the lens dangling by my side, and had to straighten it up before shooting.
New Hood Release Button
The new hood is also a matte finish, rather than the glossy plastic that the original hood had. The inside of the hood has a felt-like material on it, the same as the original one, but the matte plastic on the outside will make handling the hood better, compared to the glossy plastic old hood, which could be a little slippy to handle, especially with dry hands.
Comparison with Hoods
Wider Focusing Ring
Minor changes are that the new 70-200mm F2.8L IS II USM lens is about 2mm longer than the original, and has some smoother curves in places, and a lower profile switch panel. Externally the other obvious change and another improvement is the larger focusing ring. The new versions focusing ring is about 1.5 times wider than the original, which means you won’t have to search for the ring to manually focus while shooting. This is another one of those things that I didn’t realize was a problem until they fixed it, but I do seem to recall fumbling for this ring in the past. I manually tweak focus quite a lot, especially when shooting with LiveView, and I found it was just there for my hand when I reached for it, but I didn’t notice that the ring was bigger until I lined the two lenses up side by side to photograph them to illustrate this Podcast episode.
Comparison of Old and New
Internally, the new version now has 1 fluorite and 5 UD elements, compared to 4 UD elements and no fluorite lenses in the old version. The 70-200mm F4 lens also has a fluorite element, and has historically been much sharper than the F2.8 lens, even at F4, so the fluorite element is a welcome addition to the new 70-200mm F2.8 lens.
Of course, the other nice changes is an additional stop of Image Stabilization. The original lens had 3 stop IS, and the new one is purported to have 4 stops of IS. This I guess and the other changes mean that the already somewhat heavy 70-200mm F2.8 lens, at 1470g is now 20g heavier in this version II incarnation, at 1490g. Either lens will start to take its toll on your arms and shoulders if used hand-held for any length of time, but once you are used to that, it’s definitely a hand-holdable lens, especially with it’s now 4 stop Image Stabilization.
Shorter Minimum Focus Distance
One other improvement that I should note that I’m very happy to see is that the new version now has a shorter minimum focus distance, which means you can get closer to your subject than you could before. Now, the specifications for this lens differ from the Japan site and the US Canon Web site. On the Japan site, and on the lens barrel itself I should note, it says that the minimum focus distance for the old version is 140cm, but on the US site, it says 1.3m or 130cm. Assuming that what it says on the lens itself if correct, the version II now gives us 20cm shorter minimal focus distance, which is actually very important when shooting flowers for example, or even portraits say, when you really want to fill the frame with the subjects face.
So, I did a number of tests with charts etc. and I did a lot of real-world shooting as well, which we’ll get to later. First, let’s take a look at how the lens fairs under some relatively strict but by no means scientific tests. The first thing I did, and I advice anyone to do when you buy a new lens, is to download and print out a lens test chart and do some tests. I used Tim Jackson’s Focus Test Chart, but you can also buy something like one of the Lens Align Pro Focus Calibration products, which I think I’m going to pick up at some point. Auto-focus accuracy can vary per copy, and although Canon manufacturing and quality assurance standards are very high, occasionally a bad copy gets through the production line, and you usually only have a week or two to check if yours is OK, and still be able to return it to the shop from which you bought it. Because of this, I test all of my new lenses for this as soon as I buy them.
I’ve included a photo of the chart that I shot wide open at F2.8 for your reference, but because the lens performed exactly as I’d expect, with the focus spot on after laying the test chart on the table, then shooting it from a 45° angle, and the using the auto focus to focus on the center line, I’m not going to share the full details of the test. Basically I shot the chart at 70mm, 115mm and 200mm, and worked my way through the apertures from F2.8, F3.2, F4, F5.6, F8, F11, F16, F22 and F32. The lens focused right on the line, and the results were fine.
I did want to see how sharp the lens was, both as a stand-alone test, to see if it lives up to my expectations, and in comparison to the original 70-200mm F2.8 lens, so I also photographed an ISO 12233 Resolution Test Chart, that I downloaded from Cornell University’s Web site.
ISO 12233 Resolution Test Chart
The results were very favorable for the new version of the 70-200mm lens. I created a few animated GIF files that we’ll look at below to illustrate the difference in the resolving power between the two lenses. This first animation shows a comparison of the center of the chart shot with both lenses wide open at F2.8, at 70mm, 115mm and 200mm. We can see from this that the version II lens is just so much sharper than the old version at both extremes of its focal length, and in the middle. I start with the old version of the lens, and then switch to the version II image for each focal length, and you can just see the focus snap in as I switch to the version II image each time. This is very impressive to me.
Wide Open F2.8 @ 70mm, 115mm and 200mm Comparisons
Note that these animations are 100% crops of the original RAW files, and I ran only the default sharpening that Lightroom applies to all RAW files, unless you change the settings. There’s nothing else done to these images
The next animation shows the center of the chart, shot at 115mm focal length, and I rotate through all main apertures from F2.8 right down to the smallest aperture at F32. Although lenses generally get pretty soft due to diffraction when you stop them down through F16 and smaller, I shot F22 and F32 images as well, so that we could see just how much diffraction affects this lens, and again, I was very pleased with the results. As we cycle through the apertures, the first thing that you’ll notice is that the lens is sharp wide open at F2.8, and doesn’t really change through to F11, and then starts to get very slightly softer at F16, but even at F22 is suffers very little from diffraction, and even F32 is usable if you really needed the additional depth-of-field that this smallest aperture will provide, if you can accept a slightly soft image. Note too that at F32, to my eye, the lens is still sharper than the original version of this lens at F2.8, so again, I’m very happy with these results.
100% Crop Full Aperture Range @ 115mm
So far we’ve looked at the center of the chart, which I shot obviously with the center of the lens. As you know though, most lenses are less sharp around the edges than they are in the center, so I’ve also created an animation to show the top right corner of both lenses at F2.8. We can see that the old version of this lens is a mess at the edges at F2.8. (Note here though that the results are slightly skewed, because I was shooting up at the target by a centimeter or so. It was the only place on my wall that I could stick the target!) The results do get slightly better as you stop the lens down of course, but as I like to use lenses wide open, to capture scenes with flowers for example over at the edges, this has been a real pain for me with this lens. Because the chart, even printed out on 13×19″ paper, didn’t fill the frame at 70mm, this animation only shows the corners when shot at 115mm and 200mm. You can see here that even the version II is much softer in the top corner when used wide open that it is in the center, but within acceptable limits in my option. We can also see that it performs better at 200mm than it does at 115mm.
100% Crop of Top Right Corner @ 115mm and 200mm
These tests are all well and good, but I know you also want to see how the lens fairs in the field. Again, I like to shoot the lens wide open, and as that’s usually the weakest aperture for most lenses, especially zoom lenses, I shot many of these examples wide open at F2.8. This first example photo, of a field of oilseed rape flowers, has the main subject along the right third, with the flower head close to the top of the frame.
Oilseed Rape with 70-200mm F2.8 Version II
I’m only going to include the 70-200mm F2.8 version II image here, as at the Web size, you really can’t see the difference between this and the version I image. But here are two 100% crops, first of from the Version I lens, and then from the version II lens.
100% Crop – Oilseed Rape with 70-200mm F2.8 Version I
100% Crop – Oilseed Rape with 70-200mm F2.8 Version II
You can see that the Version II lens has produced a much sharper image, even though the subject that I cropped out here is close to the edge of the lens, albeit not the very corner. Note that if you want to get a very quick comparison without scrolling, you can click the thumbnails at the bottom of this post, and navigate back and forth by clicking on the left and right side of the images.
I should note too that the original lens, although certainly softer, has produced a very nice image. If you are happy with that amount of sharpness, then you certainly won’t need to run out and buy the updated version. Let’s continue to look at some examples though, before you fully make up your mind.
I know that many people also use this lens as a portrait lens, so I paid a visit to my friendly Barber again, and asked him to pose for a few shots, again using both lenses for comparison.
My Barber – Ishioka-san
Below again are two 100% crops of images from both the old and new version of this lens for comparison.
100% Crop with Version I of the 70-200mm F2.8L Lens
Again, very acceptable sharpness, but here’s a 100% crop from the version II lens.
100% Crop with Version II of the 70-200mm F2.8L Lens
So again, although the old version has produced a nice image, even wide open, the new version is sharper. I should also note here that if you think I’m being cruel by showing you a 100% crop of my barbers wrinkles, note that he’s over sixty years old. My eyes are more wrinkled than this and I’m 20 years younger than him!
Next I made a trip to a local temple called Daienji, and tried out the new four stop Image Stabilization. I have to say that I was not totally impressed with this. Not unimpressed with the IS in general, but I did not get great results as often as I’d hoped. I did get some usable shots though, and have in fact uploaded these last few images to my online gallery, so you can jump to them with their number that I’ll call out as I usually do as we look at these last few real-world example images.
First up is image number 2517, of some Jizou statues at the temple. I shot wide open at F2.8, and focused on the nearest eye, as I would a portrait shot. I noticed the lens searched quite a lot here, but it was very dark. The sun had already very low in the sky, and behind buildings. The temple grounds are walled in too, so there was little available light. I found though that the lens search less when I zoomed in, and got rid of some lighter patches in the background. It didn’t feel great, but this only happened with this subject, so I’m not going to panic about this just yet. I’ll update you later though if I see more of this searching. Let’s also bear in mind that this exposure required 1/30th of a second at ISO 200, so you can appreciate that there was not a lot of light.
Daienji Temple Jizou @ F2.8 1/30 ISO 200
In the next image, number 2519, I was shooting at 125mm which means if you use the rule of thumb of using the focal length as the minimum shutter speed, i.e. 1/125th of a second, we can calculate that four stops below that is 1/6th of a second with ISO 100.
Potchari Jizou @ F2.8 1/6 ISO 100
Here I was testing that the new four stop Image Stabilization was as good as I’d hoped. I was kneeling in a similar position to my MBP Kneeling Man logo, but cranked over to one side a little to avoid an obstacle, so it wasn’t the most stable pose to shoot from, but I ended up shooting around 20 frames of this subject, and only about three were sharp. The others ranged from slightly soft to totally blurry.
Kareshi no Shashin F11 1/6400 ISO 100
I’d focused between the eye and the nose, to get some definition in the shape of the nose, and get the inside of the eye sharp, kind of juggling priorities to still get an overall well focused image, without increasing my depth-of-field. I use all sorts of tricks to keep my lenses wide open. 🙂
I actually uploaded six images shot with the new version II 70-200mm at this temple to Flickr and my Web site, but I’ll skip them here for the sake of time. I will put a link into the show-note though that will list all images shot with this lens in my online gallery, if you want to take a look.
As I was walking out of the Showa Memorial Park on Saturday afternoon, the sun was low in the sky, and there were lots of people still on the boating lake, and I shot one last image that I’d like to leave you with today, and that is number 2516 (right).
I shot this with an aperture of F11, and a shutter speed of 1/6400 of a second, because the sun was very bright reflecting off the water. I didn’t mind this as I didn’t particularly need a shallow depth-of-field here, and I didn’t really want to take the time to fit an ND filter, as there were a lot of people on the lake, and chances like this, with just one boat in the frame were not going to come along so often. I really like this image though, with the well-defined silhouette figures of the young couple enjoying a later afternoon row on the lake. There’s even a little bonus duck paddling along the top of the frame.
I just wanted to share this with you as a last example of the quality of this lens though, especially as I haven’t shown you any real-world examples with the lens stopped down below F2.8 yet.
Here though is a 100% crop (below) of the young woman and half of the guy in the boat. I’m sure you’ll agree that the sharpness of these silhouettes is incredible. Also, the way the lens handled the specular highlights is pretty impressive too. There’s nowhere that these highlights are overly bleeding into the silhouette of the couple. At least when I zoomed in to look at this one, I had one of those hair-standing-up-on-the-back-of-my-head moments.
100% Crop – Kareshi no Shashin
So there you have my initial thoughts about the new EF 70-200F2.8L IS II USM Lens from Canon. if you currently own a version one lens and are using it with a high-resolution camera like the 5D Mark II or 1Ds Mark III, and if you are happy with your results, then you have nothing to worry about. I personally think the upgrade is worth it to get images this sharp wide-open. Even stopped down, the version is not as sharp as the new version, because it is simply out-resolved by the 21-megapixel sensors. As I say though, if you are happy with your current results, don’t sweat it.
I though am very pleased that I now have my workhorse 70-200 F2.8 lens back. Wild-horses couldn’t have kept me from digging deep for this one, and now that I’ve tested it out, I’m very pleased that I took the plunge.
NOTE: The week after this Podcast/Blog post, I did a follow up review, having tested the Version II F2.8 lens with the 1.4X and 2X Extenders (teleconverters). You can read and listen to the follow up review here: