[normhead]

#5 appears to be pulling away.

[ScooterMaxi Jim]

Just want to expand thoughts on my earlier theme. When I got my 40mm xs I was excited that it seemed much richer and with better bokeh than the DA 50. I probably was also influenced by its far-more confident and accurate AF. Image-wise it seemed almost too stark to believe. Then I did a couple similar tests to what Norm has done (except I wanted to check for more-distant bokeh, as well). My K-3 tends to overexpose by about 1/3 stop, and the DA 50 also adds a bit of overexposure. The 40xs underexposes by about one-third of a stop. If anything, it turns out the DA 50 has slightly better bokeh when these other factors are equalized. While I still like the 40xs more overall, it is pretty clear that the DA 50 has a more modern look, and is sharper in the center. The "richness" I perceived initially is almost entirely due to the exposure difference.

That's a very long explanation of why I prefer #5 - because I like the richness that comes with slight underexposure. It almost certainly isn't the sharpest lens, but the warmth and richness might be mostly due to slight underexposure now that I have heard the comments and further studied the images.

These comparisons are a real challenge - and I just want Norm to understand that I'm glad he did it - not intending to nitpick about the setup.

[ffking]

Since the title of the thread is "Your favourite image." I'd completely discourage further speculation on focus points etc.

are you saying where an image is focused makes no difference to how you see that image? I doubt it! - Focus is critical to image success in wide-open fast lenses. I'm not being nit-pickingly critical of your technique, just saying that for me those things make a difference in how I perceive a image - it was because I as looking at it as a test of favourite image not of lens-sharpness that I made the point - but just my opinion

[normhead]

just saying that for me those things make a difference in how I perceive a image

If we could control where an image was focussed with an SLR, we wouldn't need testing benches to test lenses would we? One of the things I learned, was how difficult it is to achieve perfect focus . It can't be done with AF, AF doesn't even focus on the same place every time. Every images is different. People claim they can do better with MF, but I'm not seeing a test that proves it. I can take a burst of a stationary object and have 3 of 4 images in acceptable focus, but not sharp focus. The issues raised by these tests, is how can you achieve sharp focus in every image, without using a testing bench?

It's something I've been mulling over for a few years, when Pop photography, German edition ran a series of tests on AF which of course Pentax won. The problem for me was they were looking at "acceptable focus". I thought "why aren't they achieving the maximum sharpness the lens can produce? " Since doing tests like this, what I've realized is that no AF system does that. They get close enough, not bang on. And though they do get bang on sometimes, in a radom sort of fashion, burst shooting actually greatly increases your chance of getting one.

Most of the images I post on line are bang on, and are the result of 4 or 5 shots bursts, with AF set to focus priority, so the camera is adjusting focus between images, and one or two shots in the burst are perfect focus. Trying to use an MF lens with no chance for a random burst effect I had to resort to bench techniques slowly turning the focus ring moving the focus point very short distances and selecting afterwards the image that was in focus.

So the short answer is, where your focus does affect the image, your mistake is thinking that out in the real world you're going to have a lot of control of that. That's why some kind of real world test of performance is better than lab test. I'm testing the whole system, a bench test just tests the lens. In going through images that are in acceptable focus, you start to realize how rare an image that is in really sharp focus is.

As the recent debacle with the Sigma 18-35 has shown. The fact that a lens can take really sharp images is pretty much irrelevant, if it doesn't focus reliably. There are times when you have time to fiddle until you get it right, but most of the time you don't. I'd suggest that given limited time and opportunities most of us are better off with a lens that is little bit softer, but focuses quickly and reliably than a very sharp lens that is finicky.

I know people are going to throw up there arms and insist they get maximum sharpness on every image, and that they have to have that. And I can't argue with them, I don't know what everyone does or how effective they are. But to me that's just background noise. For me the thing is, i don't know what they did, and i know what results I got. And I know exactly what it took to coax a decent image out of each of the lenses I used. They went from 2 of 5 for the Tamron 17-50 in sharp to 20 images for the Super Tak. even if the Super Tak wins on IQ, I'd have a hard tim recommending it over the Tammy for the everyday shooter.

"Which is going to give you the absolute best image", is a different question than "which gives you the best chance of capturing the best image in a given shooting situation." In low light shooting at ƒ5.6 I discovered the Super Tak gives you almost no chance. Even focus assist doesn't work. And unless you have a testing bench, focusing the Super Tak wide open then stopping down isn't really an option, On a tripod there's chance you will inadvertently move the tripod while stopping down the lens.

---

Last edited by normhead; 02-01-2016 at 01:05 PM.

[Nicolas06]

For me, this depend heavily of the subject and settings. When I shot landscapes, except if the front plane is really near me, usually I would get perfectly in focus pictures. Some picture might not be, but the failure rate is low.

Now if I shot portraiture, usually with the focus point on the eye of the subject, it will depend a lot of the apperture I choose. f/4-f/f5.6 or more closed down would do the job. f/1.8 is another issue.

This also depend quite heavily of the lens, the body and the lighting conditions. My FA50 was almost unusable near the max apperture with the K5, it got better with the K3, but the FA77 with K3 tend to perform quite well.

[UncleVanya]

Focus issues are tricky when comparisons are made. In a vacuum the point made about this is what autofocus does rings true. But once you slice out comparison bits of the photo to look at (pixel peeping) the ability to separate good from bad focus from good from bad sharpness is compromised. Since each lens had many trials and we have no idea if this was a focus miss or a focus hit some may have difficulty separating the lens performance from a poorly representative shot.

That said, it was a lot of work to set this up and it is well within reason to ask what shot we prefer. Making a buying decision based on this thread however might be a poor decision.

[normhead]

Focus issues are tricky when comparisons are made. In a vacuum the point made about this is what autofocus does rings true. But once you slice out comparison bits of the photo to look at (pixel peeping) the ability to separate good from bad focus from good from bad sharpness is compromised. Since each lens had many trials and we have no idea if this was a focus miss or a focus hit some may have difficulty separating the lens performance from a poorly representative shot.

That said, it was a lot of work to set this up and it is well within reason to ask what shot we prefer. Making a buying decision based on this thread however might be a poor decision.

As stated before, images were selected based on the clarity of the feather tips in the centre pixel peeper, each final image was selected from multiple images. So, even though it's been stated over and over, the same focal reference was used for each photo. I'm having in little trouble understanding why this is so hard to comprehend. The only way to improve this would be to include a DoF scale and perhaps increase the number of image. Some lenses were very close to their minimum focusing distances, DoF for most lenses was very thin.

[monochrome]

This the the Consumer Reports age, in which everything is judged and ranked so that a 'Best' can be declared. Then we argue about the testing methodology, the ranking protocol and sample variation. And whether a fault was lens softness or focus error.

FWIW, I selected the brighter, more vibrant images rather than the sharpest - because I like them better. I think I voted 6 base on set #3.

[UncleVanya]

As stated before, images were selected based on the clarity of the feather tips in the centre pixel peeper, each final image was selected from multiple images. So, even though it's been stated over and over, the same focal reference was used for each photo. I'm having in little trouble understanding why this is so hard to comprehend. The only way to improve this would be to include a DoF scale and perhaps increase the number of image. Some lenses were very close to their minimum focusing distances, DoF for most lenses was very thin.

Actually I missed that detail - that helps. Thanks. I think that's fine. I also think there may be minor variations that may be hard to control for and this is a nice effort to show these lenses. If that isn't clear I'm saying it this way to clarify. I don't think anyone should base purchase decisions on a single bit of info in any case - not just this thread but any single review etc.

[BrianR]

Does this thing move for you guys? I didn't ask them to, but picasa turned all my tests from yesterday into an animated gif... how strange is that?

Ok, the animated gif is wacky. We'd noted the differences in magnification for the red ball in the dreamcatcher but I didn't clue in that the background looks to be the same size in all the photos.

Could you make a guess at the distances from camera to foreground and background walls? It will be interesting to see which lenses match up with which magnification characteristics. The suspense!

[normhead]

Actually I missed that detail - that helps. Thanks. I think that's fine. I also think there may be minor variations that may be hard to control for and this is a nice effort to show these lenses. If that isn't clear I'm saying it this way to clarify. I don't think anyone should base purchase decisions on a single bit of info in any case - not just this thread but any single review etc.

The results of my last test were so different, if you combine the two, you know nothing. So, I'm trying to come up with a way to do some kind of definitive testing that won't become too dense to be understandable. And, I'd really like to have a "face off poll" with the top three finishers in this poll. which looks like it's going to be the Super Tak, the DA 35 ƒ2.4 and the Tamron 17-50.

But that would mean I have to try and focus the Super Tak again, oh woe is me.

[normhead]

OK, unlike my previous test. this test came out in line with popular wisdom with the exception of the Tammy 17-50.

Lens 1 FA 35-80
Lens 2 FA-J 18-35
Lens 3 DA 35 ƒ2.4
Lens 4 DA 18-55
Lens 5 35mm Super Tak
Lens 6 Tamron 17-50
Lens 7 DA 18-135

The forum has spoken.

Anyone interested in a show down among the top 3, any one of which could have been the top vote getter is all the people who didn't vote for one of the top 3 had voted for it?

Lens #1 by a significant margin was taken with the Super Tak, although it did take 20 exposures to get the winning image.

Best of the AF images is the Tamron 17-50 followed closely by the DA 35 ƒ2.4

[bertwert]

I guess I like the Tamron followed by the DA 35 then Tak 35.

[normhead]

I guess I like the Tamron followed by the DA 35 then Tak 35.

I'm guessing those three lenses are separated by issues other than sharpness, given that they all got roughly 30% of the vote. At that point people are probably looking at how the lens renders an image and their personal preference in that regard.

[RGlasel]

I don't think anyone should base purchase decisions on a single bit of info in any case - not just this thread but any single review etc.

While I'm waiting for Norm to reveal which lens is behind which door, I'll go off topic on this point.

The problem with buying decisions isn't a lack of information, it's the interference between all the bits and pieces of information beamed at us that makes it so difficult to clearly see the best choice. Heck, buying raisin bran can be stressful, did you know there is a significant difference between Kelloggs and Post Raisin Bran? Not from the outside, both use a similar colour scheme for the box and it is easy to pick up the brand you didn't want while driving your cart down the aisle, but once you pour the flakes into your bowl, you will notice a big difference in the size, shape and texture of the flakes. Post goes for brand name raisins, Kelloggs produces a lighter flake that looks better in the bowl (in my opinion, of course). Different courses for horses, you might say.

You are quite right when you say nobody should make purchase decisions based on a single bit of information, but not because we make poorer decisions, but because we worry more about making bad decisions. The #2 reason to buy anything is to relieve stress, once our basic needs are met, the next step is to deal with the anxiety that comes from worrying about poorly allocating the limited resources we have. Carefully weighing information about our purchases isn't much different than trying to figure out where the best spot to catch a fish or hunt an animal is. You can always gather information after making the purchase, but then you have to filter out the negatives so as not to re-introduce the anxiety relieved by making the purchase.

Even at the corporate level, there is no "right" way to determine the best buying decisions. When it comes to judging the truth in tests and reviews, it really comes down to which testers and reviewers we most closely identify with. To my thinking, that's the point behind Norm's test, in the absence of all the background information that we would like to use in order to judge his testing methods, we have to find something in the images themselves that we can tie our cans to, in order to come to a decision we can commit to. It could very well be that the image we select came from a lens we would never buy for ourselves, once we know what it is.

-------------

In the time it took me to type the above, Norm lifted the curtain. I guess if I ever want to take a picture of a dreamcatcher hanging on a wall with a tripod and lots of time, I should buy a used Super Tak.