Staff note: This post may contain affiliate links, which means Pentax Forums may earn a small commission if a visitor clicks through and makes a purchase. If you would like to support the forum directly, you may also make a donation here.
Series Contents
Those of you who read the prior test of the DA645 28-45 will know the story. I purchased a pre-owned lens from Adorama at a good price, and the lens turned out to be a problem performer. It was fine at some angles, but when level on the tripod, aimed at my test scene, it did not match the performance of either the 35mm prime or the 45mm end of the 45-85 zoom, and it missed by a mile. My suspicion is that something that is supposed to not be loose was, and fell out of alignment under certain conditions and into alignment at other times. Adorama took it back without a hiccup, and I sent that money to Australia to a forum member who offered me his lens instead. The beast was proving physically unmanageable for him, and I can understand that. It is indeed a beast, as all who have reviewed it will confirm. But compared to my Sinar P, it's light as a feather. And compared to a Zeiss Jena 180mm f/2.8 Sonnar (or, Heaven forbit, the 300), it's not particularly beastly. Put that Sonnar on a Kiev 60, and carry that around for a bit. Or a Pentax 67 with a 300mm lens and TTL finder. You'll never complain about the 645z again.
But I do consider carrying it around as part of my workout regimen. I warm up using my dumbbells, usually in the range of 35-50 pounds each, as a warm up for the big lift of carrying this puppy. Okay, it's not that bad, but it will still be favored by those who work out in the gym.
Today I made new test photos with the replacement lens, and conditions were a bit more like my original test in terms of lighting. It was a gray day just as it was in the late spring when I tested the other lenses in this series. But unlike that prior occasion, it never made it above freezing here so the camera got a bit of a cold-weather workout. The winter conditions also mean that the bushes and trees at Chez Denney are more leafless than in prior photos.
Here's a general view at 28mm:
As before, these are straight out of the camera, processed in DxO Photolab with the driest possible settings.
28mm
Here's a 100% crop from the middle of the image, at f/4.5. Remember that these 1:1 crops, when displayed on a typical monitor at 100 pixel/inch, are part of an image seven feet wide.
One of the tests of a zoom is whether it's better to carry a range of primes. The only good prime that competes at this end of the zoom range is the DA or DFA 25mm lens, which is of course no longer in production. The only lens I have in my collection that hopes to compete at this focal length is an Arsat 30mm Fisheye. So, what the heck, let's show the center crop from that lens (which is itself also a beast). Here is the image made at f/11, so we are comparing the best aperture of that lens with the wide-open aperture of the 28-45.
The Arsat doesn't stand a chance. But if you need a fisheye, and only a fisheye will do, then the Arsat is one.
But the real test of a wide-angle lens is in the corners. Is there field curvature? Does the sharpness hold up in the corners? Let's look. Here's the lower right corner:
This isn't curvature--it's just loss of sharpness in the extreme corner at f/4.5. But this IS just the very corner--compare this with the full image above. We're talk the four or five inches from the corner on that seven-foot print. This loss of sharpness in the corner nearly completely cleans up at f/5.6, and we'll look at f/8 further down and see that the effect is completely resolved once we've stopped down a bit.
What about curvature: Here's the upper left corner:
The branches you see are about in the focus plane, and they are sharp. The loss of sharpness that would correspond to the lower right corner is suffused into the bit of house that is quite out of focus. I don't see evidence of curvature here.
And here's the lower left corner, which was generally sharp with the FA35 (which is known for curvature). The field looks acceptably flat at 28mm.
While we are at 28mm, let's look at several apertures and see if we can determine the ideal aperture. Here's the center again at f/5.6:
f/8:
f/11:
f/16:
And f/22:
There is no practical difference between f/4.5, 5.6, 8, and 11 in the center of the image and in the focus plane. F/16 starts to degrade slightly, though I may be imagining it, but not enough to keep me from using it even for a large print. F/22, though, would allow only a more subdued print size. But insufficient depth of field trumps diffraction.
Before moving away from 28mm, here's the lower right corner at f/8:
At f/8, it's sharp right to the corner.
35mm
I'll spend less time at longer focal lengths now that we know the basic behavior of the lens. Questions for 35mm: Does it out-perform the FA35mm prime lens? Are the corners better than at 28?
Here's the center at f/4.5:
F/5.6:
And at f/8:
Here is the FA 35/3.5 at f/11:
This makes me want to check that f/8 exposure, which is very slightly softer than the f/4.5 image. F/5.6 is the clear winner of all of these, but the 35 may be diffracting a touch at f/11. The 35mm prime is very good in the center, but it's not better than the zoom. It is a LOT cheaper, though.
Here is the upper left corner at f/8:
Again, no sign of field curvature. The branches and dead leaves are close to the focal plane. The wood at left is only five feet from the lens, and the background pines are three times as far as the focus plane. There is a touch of movement blur here--shutter speed was 1/13 second. The camera was, of course, mounted on a tripod, but that river birch tree will move in not much breeze.
45mm
Center of the image at 1:1, F/4.5:
Here's the 45-85 zoom at 45 and f/4.5:
The 45-85 is stunningly good, especially at the short end. The 28-45 is a bit better.
Here's the same comparison at f/8:
Again, the 45-85 is outstanding, and the 28-45 is a bit better.
Finally, a word about focus accuracy. This lens front-focused very slightly, and I set the focus adjustmentment for this one to -3.
In conclusion, a good example of the 28-45 is an amazing performer. Please see my previous test of the failed lens for details about shake reduction, which I did not test again here. I find that shake reduction is good for about a stop, but I did not use it in the above images that were made on a tripod using mirror pre-release.
But these lenses are not bullet-proof. If you buy one pre-owned,
test it.
Rick "just about thawed out" Denney
Last edited by rdenney; 12-10-2018 at 12:33 AM.