We’ve seen that the a7R is affected by shutter shock. We’ve seen ways to ameliorate those effects. We’ve seen that the a7 is essentially free of shutter shock effects by virtue of its electronic first curtain shutter. What if we put all that together and consider the question, “At its worst shutter speed, is the a7R’s sharpness better than the a7’s?” I thought I’d find out. I set up an a7R with the 135mm f/2 Zeiss APO Sonnar ZF.2 on a Novoflex adapter. Using the tricks that we’ve learned so far, I mounted it like this:
Support for the lens barrel? Check. Shutter over the tripod head? Check. A little extra mass? Check.
Taking advantage of the fact that the only a7 shutter motion that can cause blur is that associated with the closing of the second curtain, and that is hardly ever problematic, I mounted the lens to that camera in a much more convenient way:
You’d never mount the lens this way with the a7R. It’s an invitation for the shutter to make the camera/adapter/lens assembly act like a teeter-totter with the fulcrum at the adapter foot, and you’ve got no protection against adapter slop, either. However, you can get away with it with the a7. Using the Novoflex ASTAT-NEX foot is also a good way to spare the a7 lens mount from the task of supporting the big Zeiss lens, as long as you’re careful about how you handle the camera.
I supported the above setups with an Arca Swiss C1 Cube and RRS TVC-44 legs. I put a Heliopan 77mm variable neutral density filter on the lens. Release was with the 2 second self-timer. ISO 400, lens focused wide open and then stopped down to f/5.6. ISO 12233 target with the camera at a distance to yield an active area 647 pixels high for the a7R images, and 527 for a7 ones. Thus, the lines labeled “6” represent slightly more than 1 line pair per pixel pair in the a7R images, and the lines labeled “5” represent slightly more than 1 line pair per pixel pair in the a7R ones.
Target illumination for the continuous lighting images was provided by a single Fotodiox LED-200WA-56 lamp set to full output, using the supplied reflector. I set the camera to aperture priority, and, with the ND filter set to minimum attenuation, adjusted the exposure compensation to give a shutter speed of 1/640 second. I made an exposure. Then I added sufficient light attenuation to get the shutter to 1/3 stop slower speed, and made another exposure. I continued all the way to a shutter speed of ¼ second. Then I did the whole thing again with the other camera.
I brought the images into Lightroom, tweaked exposure and set white balance by eyedropper on the paper white of the target. I left all the other settings at default. I found the sharpest a7R image (1/640 second), and opened it in Photoshop. Then I found the fuzziest (1/60 second) and did the same. I enlarged both images to 300% using bilinear interpolation. This isn’t the sharpest way to scale up an image, but it’s the about the least likely to introduce artifacts. I checked the a7 images, and saw that they were all about the same sharpness. I opened the 1/640 second one in Photoshop, and scaled it up to 368% using bilinear interpolation, so that the target features were the same size in pixels as in the a7R images.
Here’s the sharpest a7R image cropped to the upper right cross:
Here’s the worst a7R image cropped to the upper right cross:
Here’s the a7 image cropped to the upper right cross:
My conclusion is that even at the camera’s worst shutter speed, the a7R images have higher resolution and micro-contrast than the a7 images. Therefore if you’re shooting in landscape orientation at focal lengths of 135 mm and shorter, and are willing to go to a little trouble to mount your gear carefully, then you’ll get better IQ with the a7R than with the a7 at all shutter speeds. The difference is great enough that it probably extends out to focal lengths significantly longer than 135mm. I leave the testing of that to anyone who has access to both cameras and an ISO 12233 chart.
Lloyd Chambers has stopped using the a7R for lens testing because of shutter shock. I can see why he did it; it makes it hard to sort out what’s the shutter and what’s the lens. However, his needs are different than photographers who don’t test gear for a living, and you shouldn’t interpret that to mean that you’ll get sharper images with the a7.