Yes I know I’m doing these lens tests “all wrong”. I’m an engineer; we know about these things. It should be done in a studio with controlled lighting and the camera on a tripod aimed at a lens testing chart which would enable me to carefully count all the lines of resolution from the center to the edges, et cetera.
The only thing is that would be even more boring than what I am doing, both for me and you. Besides, the idea here is to see which lenses of the bunch are capable of producing good pictures under real-world conditions. Anyway, it’s more fun this way.
So we’re up to the Prakticar 70-210mm f4.5 zoom.
This is another ‘automatic only’ lens so it won’t work on the Canon, but it’s no trouble on the Sony. In fact the maximum focal length of 210mm is the same as that of the least expensive Sony zoom I’ve come across, which gives me the opportunity to see if that length is really good enough for my picture purposes – before buying what is a fairly expensive lens (the Sony 55-210mm).
The Prakticar lenses, as I understand things, came about in the late 1970s as Praktica switched to their bayonet ‘PB’ mount but retained the 42mm screw thread mount for lower-end equipment. These were made by more than one company, and the specifications on this example don’t match any that I found in research (all of which say it should be an f4). I suspect it was built by Samyang, however. This lens radiates “lower-end equipment” as it has an over-all too-light and too-cheap feel to it. Its optical performance only reinforces this impression.
That’s about as good as it gets. When you start pushing the limits you find they were already a lot nearer to you than you thought, and certainly nearer than they ought to be.
Okay this lens has low contrast, washed-out colours, poor resolution, and a tendency to exhibit chromatic aberration almost always. Not good. Not good at all. Many, many disappointing pictures. Let’s try harder and see if we can get a decent shot out of it.
We’ve got to try harder!
Much better, but still rather fuzzy even without zooming in. It’s a good thing digital images don’t cost like film!
That is at least not awful. Some post-processing was involved, and no small amount of luck. Considering the build quality, the operation (sloppy focus/zoom ring and difficulty seeing to focus at only f4.5), and the end results this lens gets a rating of “poor”. It’s hard to get even an artistic sort of image from it.
Oh and what about evaluating the zoom length? A bit of a poser considering the low sharpness, but here is what I know to be a downy woodpecker in an aspen tree at about 80 feet away:
The Sony’s 24MP sensor lets us zoom in digitally, which really betrays the lens’s poor resolution:
C’est la vie photographique, non?
Today’s episode: the Sun Actinon 28-80mm f2.8 zoom.
A little background info: Sun optical was a Japanese company that started back in the 1930s and made some pretty good lenses, sometimes sold under ‘house brands’ of retailers. From what I have read the Sun Actinon is not associated with this company, but rather is a ‘house brand’ itself used by a British photographic retailer called Image. Who actually made it is anybody’s guess. Amazingly, for all the information available on the Internet these days there are still vast quantities missing – and at least an equal amount that’s inaccurate.
Oh how I would like to love this lens! It has so much going for it. First of all it has a good focal length range for a 35mm camera: from wide angle to short telephoto. It has a fairly fast maximum aperture of 2.8 (4.5 at tele) as well. Also a built-in macro focus. It’s compact, fairly light yet with a sturdy feel, and the controls work nicely although the focus ring has a slight slop to it and the zoom ring needs the grip re-glued, but both those problems are from use not design.
This is an automatic-only lens, meaning it won’t fully work on the Canon because the adaptor I have for that camera doesn’t push in the aperture pin to stop it down. Works fine on the Sony though. So let’s see what we get:
Not bad. Good colour and contrast, no sign of chromatic aberration, and you can see the recurring dirt on the sensor quite clearly. Let’s look again:
What the hell happened? If you can force your eye to look at the center you’ll see it’s in focus. The difference between the two photos is that the first is at f16 and the second at f2.8. It is normal for sharpness to fall off towards the edges. All lenses have this issue: it’s a function of the physics of focusing light. Usually a lens will have an aperture where it is at its peak sharpness edge to edge. (If you look at the first image again you will see it too has a slight blurriness on the edges.) But this is the absolute worst example of edge fall-off I have ever seen on a modern lens! It made me check to see if I’d somehow missed a pound of lard smeared on it or some other blatantly obvious problem that could cause the effect. The failure continued in every photo I took, no matter what aperture, focal length, or distance was used: obvious low-resolution around the edges persisted.
Which is quite a shame because, well look what it can do in the center:
That’s pretty darn good. Why then does it blur to oblivion around the edges? I don’t know specifically, but a lens so bad should never have made it off the assembly line.
Distant, full focal length, middle aperture – and still the edges are badly blurred.
Hey, see all those spots? You can go crazy trying to keep a mirrorless camera’s sensor clean. Especially when you keep changing lenses. Or take one off repeatedly to look for dirt or grease on it. And no it was not the sensor dirt causing the problem: I switched to other lenses while doing this just to be sure the camera wasn’t in some way the source of the trouble.
Over and over I tried and over and over I cried. This lens had so much potential, and all hopes were dashed because it could not deliver a decent picture under any circumstances. I prefer the consistent low resolution of the Opticam and Cunor lenses to this center-only-sharp, edge-all-blurred disaster.
Lens rating: poor. I can’t even give it a “fair” label because I could not find any circumstances under which it would render an acceptable image. (Cropping out a small center section of the full frame of every image is not reasonable.)
Or more accurately the “post-shoot digital zooming thing”. Or the “making use of 24MP” thing.
The premise is that with a high resolution sensor you can crop quite a small portion out of the full frame and still have a reasonably sharp image to present. This is the opposite of demonstrating why high MP sensors aren’t any advantage once the picture is reduced to ‘normal’ size. Let’s see how it works.
The camera: Sony a6000 with 24 MP sensor. The lens: Pentax Super Takumar 50mm f1.4 (the sharpest lens I own). The subject: a dollar store decoration cardinal standing in for a real bird because it won’t fly away while I’m photographing it. The distance: approximately 50 feet.
This is one picture cropped four ways and ending up at 1500 x 1000 each time.
As you can see, 1/16 of the full frame is about the maximum limit. It is already “soft” and it still doesn’t reveal the bird very well. Of course 50mm isn’t much focal length (about 75mm equivalent) so for actual birding a longer lens would do better. What we’re studying here is the cropping effect on resolution. This size is slightly larger than my ‘normal’ image presentation, and it is noticeably soft even to my eyesight. Applying the “unsharp mask” doesn’t help.
So what do we learn? We learn there is no substitute for being close to the subject in the first place, whether in actual distance or by use of telephoto lens (which adds its own problems due to looking through increased atmosphere). You can make use of some digital zooming this way, but don’t expect it to be a substitute for multiples of focal length.
By the way, this is the third experiment I did like this. I’m using the examples from this one because they present the best demonstration with the best lens. I have also done it with the 135mm Vivitar, with similar results. If anything the longer focal length adds to the drop-off in sharpness because longer lenses tend not to be as sharp. In other words the digital cropping will not only magnify the image, but any flaws in it as well.
The third experiment utilized the Soligor 85-205mm zoom and a Pentax 2X extender as I was trying variations to see how having an actual long lens would affect the field of view. Or to put it simply, to see if buying the 50-210mm Sony zoom would give enough focal length to make birding pictures possible. The result was “no”. Disregarding the softness of the lens, the field of view was not magnified sufficiently to make for good identification of the bird.
As you can see, for birding a long focal length is desirable because you probably won’t be able to get close to them physically. In fact having the wide zoom range is valuable because you never know how far away you will be when you spot something. This is why I like the Nikon P610, although it needs a faster zoom control and better viewfinder. Hence my recent reference to the desirability of an actual DSLR with a tiny 1/2.3 sensor. But that’s not going to happen. Nor will any manufacturer come up with a digital “sport finder”: a wider frame of view with a centered square indicating the actual image area so small objects can be spotted more easily and brought in close. Perhaps some camera has this, but I haven’t seen it.
I’m still at sixes and sevens about the Sony a6000. It is a very good camera and does produce excellent results. For the most part it’s easy to use and has great lens adaptability. There are only a few flaws, and I can’t quite decide if they are too much to put up with. I think I need to use it more before determining whether it’s worth sinking additional money into or selling it off for whatever I can get.
Or: another boring lens test post.
A few moments of sunshine have been granted here, so I took advantage of it. The main purpose was to compare the Canon EF 75-300mm zoom to the Canon EF-S 55-250mm zoom. The results were … interesting.
I’m going to spare you the innumerable shots made under various experimental conditions and get down to a few that demonstrate the differences. Keep in mind the EF-S lens has image stabilization, thus to make things “fair” I had the ISO at 400 so the shutter speed could be quick (1/500 and 1/750) and the aperture stopped down a bit (f9.5 and f8.0) to give maximum advantage to both lenses. As usual these are 640×427 crops of the full-size images.
The first two are taken at 75mm and approximately 75mm as this is the minimum focal length of the longer zoom.
This is a very difficult one to call! I think I see a slight edge in the 75-300mm in fact.
The next pair are at 250mm and approximately 250mm – the maximum of the shorter range zoom.
Here the prize goes to the 55-250mm EF-S as there is noticeably better detail in the bark. It is still a very close thing, and in normal size presentation the difference isn’t noticeable. What is noticeable in each photo is that the colour rendition and contrast is the same between both lenses.
Now, would a brand-new edition of the EF 75-300mm lens be better? Maybe, but I doubt it as I don’t see any defects in the glass itself. Certainly a new one would operate more smoothly, but it would also cost 5X as much. I suppose the $600+ version with IS would be better, but … I think I’ll save that money towards a replacement to the super-zoom Nikon. I really only bought this lens so I’d have a long zoom for the Canon 5D if/when I manage to obtain one (the EF-S lenses don’t work right on full-size sensors).
The bottom line with any lens is: can you take a good picture with it? Let’s see:
The answer appears to be “yes”.
Otherwise I have to report no progress on The Plan as of yet.
I bought a used Canon EF 75-300mm lens. New this lens retails at $329 plus tax. I paid $50 plus shipping. About 1/5 the cost, which makes it a bargain. But is it any good? I got to try it out briefly on Tuesday, before the weather turned against me. Unfortunately the day became excessively windy as the front moved in, which together with everything else I had to do curtailed my testing. Anyway here are the initial results.
First of all, it’s used. The barrel is a bit sloppy from wear and the zoom action isn’t smooth. Neither is bad enough to be objectionable to me, especially not at the price. The focus is fast enough and for the most part accurate. I managed to confuse it a few times as it couldn’t sort out what to fix on if the scene was too complex (like lots of branches) or if the subject was not a major portion of the picture (as with a raven in a large amount of sky). So how is the sharpness?
The dog’s head is shot at 75mm (120mm equivalent due to the crop factor of the sensor). The 100% section of the full image (second photo) is pretty sharp. I’d rate it very good, in fact.
The airplane is shot at 300mm (480mm effectively). The 100% segment of the full image is pretty fuzzy. There are contributing factors like a lot of atmosphere to look through and the plane being in motion (although at that distance it didn’t appear to be). There is no image stabilization on this lens, which might have helped as the shot was hand-held. Nevertheless I can’t rate the long focal length results high. Good, but not very good. The Nikon would have done better using its super-zoom capacity and IS (I know, because I’ve shot the same sort of picture with it). In theory the extra MP of the Canon should allow better ‘digital zooming’, but it just isn’t there. Physics gets in the way of the hypothetical possibilities.
The colour rendition is very good. The contrast is slightly low. These factors hold true over the whole focal range, from what I’ve seen so far.
Needless to say I have not finished testing this lens. I’ve hardly begun, in fact.
Well it looks like the sun is never going to shine again around here. At least not for longer than a few seconds at a time. So in the slightly less darkness of the past weekend I did some more testing of the Canon 40mm EF prime lens. Let’s see how it went.
This sky picture shows that for ‘general views’ the 40mm is more than acceptable, even with the slightly telephoto focal length (due to the 1.6X crop factor on the APS-C camera). It has good sharpness, contrast, and colour rendition even under less-than-ideal lighting conditions. Now let’s see how it does at 100% of image size.
The first is the full picture shrunk down to 640×427 (note this is about 53% reduction). The second is a 640×427 crop of the full-size image. Very good results here: the blurring of line detail is due to sensor resolution limits not optical fall-off. In film terms you’d be looking at the difference of the grain versus the glass.
Now for the ‘acid test’. How does it compare to the Canon 18-55mm EF-S zoom shot at approximately the same focal length? I’ll spare you the full size images and just go with the cropped segments of the 100% size.
Not bad at all. In fact we can see the 40mm comes out ahead in resolution, colour rendition, and contrast. The latter two by an almost imperceptible margin, but the sharpness difference is definitely noticeable.
So how is the image quality when you really put some effort into making a picture?
Very good indeed. Perhaps as sharp as the Nikon? Well that is something I’ll test next:
It is difficult to get the Nikon to exactly 40mm focal length, and it does have a slightly less MP (16 instead of the Canon’s 18) sensor. Also we see the colour difference and the lower contrast the Nikon has developed over time (it used to be crisper). In terms of outright sharpness, the Canon has a small but visible advantage. I tried to take a close-up image to compare the two, and the Nikon did its now usual false focus lock. *sigh* One day it is going to fail me when I really need the picture, and that will be that.
One more shot from the Canon:
All these images were shot at ISO 200 to use a slightly smaller than wide-open aperture to optimize sharpness (shutter speed is not much of an issue at 40mm). Even under the shooting conditions available the 40mm EF lens performed remarkably well. In fact I’m so encouraged by the results from it that I may just spring for a Canon 75-300mm EF lens to see if that is better than the 55-250mm EF-S zoom I’ve got.
Plus a few random photography remarks.
I missed a few shots this past week (including a marmot) as it is impossible to grab a camera and turn it on and frame up the subject and take the picture while you’re driving down the road. Even if the road is a gravel logging path and you’re only doing 10 KPH at best. I did manage to stop upon hearing a rat-a-tat-tat noise in some dead birch trees, thinking it was a woodpecker. I spotted something red moving in the viewfinder and clicked, waiting to figure out what it was until later. It was this:
This image was shot with the Canon T100 using the 55-250mm zoom at full extension. (This is a 640×480 view cropped from the full frame, not reduced.) With the crop factor this works out to about 400mm on a 135 camera. Why the Canon? Why not the famous Nikon P610? A couple of reasons. First, the Canon ‘fires up’ quicker; turn the dial and it’s ready (the Nikon works its way through some motor gymnastics before it’s ready). Second, the Canon’s optical finder is easier for me to see through. This is getting to be a problem, especially when trying to spot small subjects like birds in the distance.
Which brings us to point number three: optical and digital zooming. The Nikon can outdo the Canon optically by a factor of over 3X (1440mm vs. 400mm). This is because it has a smaller “2.3” sensor (which also reduces its effectiveness in low light). But the Canon has a slight edge in MP of about 12% so it’s better at post-shoot digital zooming. This has lead me to the decision that a T7 with its 24MP sensor would be even better for me – 50% better you might say. So in a weird way post-shoot digital zooming helps make up for my failing eyesight. Something to think about as the Nikon keeps producing out-of-focus pictures due to the loose lens (no, I can not tell if the image is in focus in the finder of any camer: I am dependent on autofocus and depth-of-field).
Now here’s a tiny butterfly. Not being a lepidopterist I don’t know what kind.
These were taken with the Fuji F80 EXR, of all things. Not really the camera for the job but it did it fairly well. The first shot is full frame, the second digitally zoomed (as is the third). You may notice a colour shift between the two shots as the camera tried to come to grips with the scene. The second image is somewhat washed out in the dried grass but the colour is better on the butterfly.
Now here’s what the Nikon did with a much larger version of what looks like the same butterfly (the first one was perhaps an inch long, the second closer to three inches):
The Nikon has utterly failed to focus a couple of times, usually when the lens is pointing down (which is telling). Here’s a shot that shows the motor-driven lens isn’t as quick as it should be. Look for the bird.
Of course if the subject will sit still, it’s fine:
Or if the conditions are right and you don’t need maximum magnification on one small object far off in the distance:
And back to the Canon for a combination of maximum optical zoom with a bit of digital as well:
A discussion elsewhere about zoom lenses and whether or not you need them reminded me of an old movie camera I used to have: the Kodak Medallion 8. It took 8mm cartridge film, and had a 3 lens turret that allowed you to switch between wide-angle, normal, and telephoto just by pivoting the lens elements around (not while shooting of course). A cheap version of a zoom, and easier than changing the whole lens!
The difficulty with spotting a subject at the ‘normal’ focal length of your eye and then getting a camera fixed on it with a telephoto lens is sometimes aggravating. Being able to spot it with the zoom at wide/normal and then twisting a ring to close in on it is much better. Motor zooms are slow for moving subjects like birds. Sometimes even the fastest autofocus and shutter release is slow; birds can be really, really quick!
But I can see where if I continue shooting wildlife I would stay with the medium-to-long manual zoom lens, quick center-spot autofocus, optical finder, and as many pixels as possible to facilitate post-shoot digital zooming. This is not the best combination for everyone, of course; I just like shooting wildlife and that’s mostly best done from a distance. Even butterflies are reluctant to hold still while you move in closer.
Addendum: Since writing this I came across another blog wherein the author made a statement along the lines of “like many people I like the effect of limited depth-of-field”. Am I the last person on Earth who wants sharpness in photos? Yes sometimes it’s nice to blur the background, but not always! And if the whole of the subject isn’t sharp … well to me it just looks like someone did a bad job photographing it.
Shutting up now. Do what you like.
I have an old Soligor M42 mount 80-205mm f3.5 macro zoom from my Pentax equipment. I tried it on the Canon once, but didn’t like it. Mainly because this lens lacks an Auto/Manual aperture switch, meaning that as-is you can only shoot wide open. I probably don’t have to explain that on a lens that long the depth of field at f3.5 is near nil, never mind that full aperture is rarely the sharpest selection for any lens.
Dan James suggested I glue the stop-down pin in, making it a full manual-only lens. It’s a good idea, but before taking that step I decided to give it another shot on the Canon so I could ask myself that all-important question “would I really use it?” before going ahead with the modification. Herein the results.
Something there is probably in focus somewhere, but I don’t see it. What I notice most is the colours are interestingly rich. This is due partly to the characteristics of the lens and partly due to the fact some idiot forgot he had the 2X Pentax extender on as well and didn’t compensate the exposure for it. Well if you’re going to screw up a digital exposure, go under with it. Going over just gives you washed out areas with zero data to work with.
This combination also brings up another issue with the lens: it is big, long, and heavy. 205mm of classical glass is 328mm on the crop sensor Canon, or 656mm with the extender. We’re talking Hubble Telescope neighbourhood here. No Image Stabilizer on this old lens! Almost impossible to handhold steady even for a normal person, never mind some old guy with permanent shakes. Crank up the ISO, turn the shutter speed to max, and you can still get nothing but a blur due to the practically non-existent DOF.
Now here is a picture which brings up another issue, albeit one not the fault of camera or photographer. At this time of year our sun angle is really low in the sky. This means the light is coming through a lot of atmosphere even at noon, making for very warm lighting. Also long shadows. Also lots of glare. We’re talking windshield-laser-of-death kind of thing.
I tried some processing on this shot and kept going back to the out-of-camera original as best for lighting and colour. It’s not a spectacular image by any means, but I like the light. I’d like it more if it were even remotely sharp. Even my eyes can see that it isn’t. Pretty sure I had it focused, but a lot can change when you push the button.
I often take this shot as a lens test. The wire is slightly diagonal to the ‘film plane’, yet we don’t see any particular point of sharp focus. Or at least I don’t. I have to conclude that there was a focal point, but the image simply isn’t sharp anywhere. How much of that is the wide-open aperture and how much is the glass? Good question.
This is a failure. It seemed like it would be a good shot, but in the end it’s pretty dull and yes it’s fuzzy. Since these pictures are done at different distances you’d think one of them would turn out sharp, but it isn’t looking good for the ol’ Soligor.
Ah, who are we kidding? There was no chance of a macro shot coming out sharp. But I had to try it because it’s one of the lens functions. Kind of silly to hand-hold it too.
This shot started out as blah. One aspect of telephotos is how they look through the haze, and pick it all up. That’s one difference between getting close with a lens and getting close with your feet. Really this started out as a mass of bluish-gray dullness, so I decided to ‘cook’ it with brightness and contrast adjustments. Now it’s an artistic rendering of scorching desert heat on the wires. And you hardly notice it isn’t sharp. Heat blur. Or something.
What conclusions do we draw? The Soligor is not a sharp lens. It is heavy and hard to handhold. The fact is I can get similar results with the 55-250mm Canon lens with fewer exposure/focus issues. What’s more, for really long shots the Nikon P610’s 1440mm equivalent blows the Soligor to pieces in every respect.
I may yet go ahead and glue that pin down, just to see if the lens is any better when stopped-down. Of all my old Pentax mount lenses, this one clearly is the worst. I’m not sure I noticed it at the time.