|
Forum: Pentax SLR Lens Discussion
04-30-2016, 03:08 PM
|
| |
I don't have the expertise to say what the cost difference would have to be. We don't know if that pricier Nikon lens in your example costs $50 more to make than the less expensive version, or $500 more. There's a lot to marketing and at least part of the cost difference could be that, combined maybe with sales volume. I'm guessing there are mechanical differences too that contribute to the cost and price difference, and possibly those mechanical difference contribute to the more consistent performance of the higher priced lens.
I would probably pay about 20-30% more for uniformity according to my ability to measure, which is at least a couple of generations behind current sensors. I'm sorry that I don't have any access to other brands for comparison so I don't know what Nikon and Canon and Sony and Olympus are doing now in terms of consistency of comparable lenses. As I've explained I'm not looking for theoretical optical perfection, which of course isn't possible, just close-to-perfection in meeting the design goals to the limit of a consumer's measuring ability. Like with these last two copies, I definitely like the last (4th) one better, but even it's a little off on the right side at 85mm. Otherwise it's' much more consistent than the other 3 copies I've tried, and by f8 or f9 it's so close at 85mm on that side that I'm not going to reject the lens because of it. For a lot of what I do I'm going to need more dof at 85mm than 5.6 will provide anyway, and I've got another lens that overlaps that higher part of the range and performs better (and probably would even against a perfect copy.) It's still annoying that I should have to put up with that, but the lens has a very useful range, and the AF works (unlike my Tamron 17-50, currently), and it has WR.
Here on the forum people place a lot of importance on lens performance of one model vs. another, but it's somewhat disingenuous to do that and then discount these situations where so many copies don't live up to the performance their designs are clearly capable of.
|
|
Forum: Pentax SLR Lens Discussion
04-29-2016, 08:53 PM
|
| |
It could also be that the lens is just internally inconsistent, although I haven't seen evidence of that in my results. As an example, with every copy I've tried, just nudging the manual focus ring has caused the image to obviously shift in the viewfinder, so apparently components are moving around inside the lens that probably shouldn't be. I'm somewhat surprised that so far, I haven't seen a problem show up related to this, particularly with vertical images. My 16-45 would produce soft edges on the top or bottom on vertical exposures, but that wasn't really a surprise with way the front of the barrel flopped around.
|
|
Forum: Pentax SLR Lens Discussion
04-29-2016, 08:02 PM
|
| |
I tried to make it obvious from the filename which was the third copy, but your criticism is valid. However, regarding effort in general, my argument is that none of us should be having to put any but the most trivial effort into testing at all. Not you in looking at the images, and not me in creating and posting them. Pentax should be doing the testing and examination for us. We should just be able to read the thread and look at pretty web-sized pictures and have the confidence that if we were to download a full-sized image, we aren't going to be terribly disappointed when we pixel-peep at any edge or corner. I agree about the whole parallel to the wall thing, etc. Almost of us are limited in the precision with which we can test, and probably we get some false-positive failures because of that. Again, that's why Pentax needs to be the one doing the testing, using precision test equipment designed for the purpose, in tightly controlled conditions.
Having said that, I decided the third copy was just not good enough. So I had to appeal to amazon to send yet another copy, because they'd essentially cut me off, and... wow. Perfect? No, but very good - at least at 16mp and with an AA filter, which is all I have to test with. This new copy (which incidentally looks kind of not-new) is good not just with brick walls or resolution charts, but with practical applications. So I'm keeping this new(ish) copy, and now I understand why others are so happy with these lenses. Well, not really. I'm betting the vast majority of these lenses are significantly defective. But all the copies I tried were pretty good across most of the frame at most focal lengths, and that's probably the standard most people are applying.
|
|
Forum: Pentax SLR Lens Discussion
04-24-2016, 09:31 AM
|
| |
Ok, no words this time. In the interest of not wearing out my K5 bodies, I'm now testing on a K200, so you've that that AA filter in the way, and only 10mp. As with the other examples you need to view this image at maximum magnification. This happens to be 85mm. I'm only posting the one 85mm example for this copy because it's relatively good elsewhere.
https://drive.google.com/folderview?id=0B6EBzgJXUw28ZGQxRXlmZXN4bEk&usp=sharing
Would you keep or continue on to more copies?
|
|
Forum: Pentax SLR Lens Discussion
04-16-2016, 05:54 AM
|
| |
This is probably a case where, if you don't have any bad experience with a lens, you think the cases of bad experiences are the exception not the rule, so you tend to trivialize them. But it's important for people who are looking at the thread to actually see what the lens can do, both good and bad. And only the text will tell them if the good results are from the 138th copy of lens after 137 had been rejected. If you're going to burn through a k3 just testing these lenses before you get a good copy, that's a consideration for some people.
I liked the lens enough that I have yet another copy on its way. Will it be good? Probably not. But then I'll have just another 134 to go.
|
|
Forum: Pentax SLR Lens Discussion
04-15-2016, 06:03 PM
|
| |
I like the subject/composition, etc. but 8708 seems to show a horrible case of decentering on the left - and in some of the other photos the left side looks a little suspect too. Anybody else notice that?
|
|
Forum: Pentax SLR Lens Discussion
04-14-2016, 05:36 AM
|
| |
I asked B&H and they said they return lenses returned to them as defective to the manufacturer. I'm not sure if "not good enough for me" qualifies as defective.
I hate to return a lens like these 16-85s as defective, because there are people who would be happy with them. So they aren't really defective, yet clearly not what they were designed to be capable of, either.
I never knew there was such a thing as decentering in the film era, but made medium-large prints where there was no discernible difference in performance in the edges/corners. Of course the centers were a little sharper than the edges, but everyone expected that. We didn't have 16-85mm lenses then, but I did have a couple of more modest zooms.
|
|
Forum: Pentax SLR Lens Discussion
04-13-2016, 03:35 PM
|
| |
I added another image to the google drive folder, above: 2nd-16mm.jpg
https://drive.google.com/a/paultibbitts.com/folderview?id=0B6EBzgJXUw28ZGQxR...k&usp=sharing#
Compare the printing at the top and bottom of the frame, particularly the top left. 26mm.
I realize that my testing isn't scientifically controlled etc., but I'm only posting this as one example of what every other test showed. Brick walls, etc. The lower focal lengths, particularly in the mid-20s, are just weak on that one edge and that corner in particular. If it was just one test I'd redo it using some other subject, etc. The center and the other edge is sharp, meaning focus has to be close. If this was my Tamron, the issue would be that the edges don't focus at the same distance as the edges, so you have to stop it down. And it would be just as bad as the worst corner here at, say, f4. But both sides would have the same blurry, fuzzy image.
I asked Pentax and the official word is that if you can tell a difference between any edge and corner and another... send it back. I wouldn't hold a lens to that high a standard, but really, if the lens can do what it can do at the bottom of this example, I want that all across the frame. Honestly, I'd settle for a little less resolution than that at the bottom; I just think it should be consistent. Maybe not if I compare using some future-generation 200mp body, but I've got 16mp and and AA filter and I can see this.
This really is a very, very good lens by my standards, on one side or the other at every focal length, and in the middle 2/3rds of the frame at every focal length. But this copy is going back too. I'd really like Pentax to address this seriously and tell us what we should expect. I wanted to send them examples but they weren't interested. I wouldn't even mind buying one and sending it in to be fixed if they could explain to me exactly how they're going to fix it and how they're going to test it and to what standards. With my Sigma I got nowhere with centering issues - what was good enough to them just wasn't even close to me (it was twice as bad or more than either of these copies.)
Incidentally I have bought Pentax lenses - and, well, if you count the Tokina 10-17 as Pentax or whatever - and been happy with the first copy I've gotten. In one case I tested a new Pentax lens I'd bought against another copy and my new one was worse, but I kept it anyway, because it was so close (and swapping for the other one would have had other issues.) So it's not like I won't accept some slight variations. I just don't think this is a slight variation.
|
|
Forum: Pentax SLR Lens Discussion
04-12-2016, 06:33 PM
|
| |
Well the results are in on copy #2, and... I'm not sure what to do. It's mostly better than copy #1 in "real life" use, except on the left side at 24mm (I tested 16,24,50, and 85.) This new copy holds up pretty well at 24mm until it gets to about the last 5% of the left side of the frame, where resolution totally nosedives. Usually lenses seem to gradually lose resolution as you near a corner or edge, but this one hold up well... until it doesn't. On the resolution test target, where the original was soft on the right side at longer focal lengths, this one is soft in the lower right and upper left. So really the performance seems to vary not just with focal length but with focusing distance.
Overall the 16-85 is a very versatile lens and performs a lot better than most of my other zooms. In the center 2/3rd of the field it's really very good. I'm not sure I'll do better than this current copy, unless maybe I try dozens. I really wish Pentax would hire somebody to test these lenses so we didn't have to.
|
|
Forum: Pentax SLR Lens Discussion
04-10-2016, 08:23 AM
|
| |
Here.The centers were so close between all three it didn't matter to me so I won't post those.
|
|
Forum: Pentax SLR Lens Discussion
04-09-2016, 03:27 PM
|
| |
I just got a copy of the lens yesterday. It's going back because it appears slightly - and I admit slightly - decentered at the long end. So I'll try another copy. This seems more of an issue up close, like with an 8.5x11 test chart, but there it's pretty clear. I can flip the camera over and the better edge performance moves from the one side to the other. Performance at the wide end seems pretty good. On the other hand all I have for testing is 16mp with an AA filter, so I'm thinking the differences would be more on 24mp without one. I'm partly considering this lens due to my AF problems with the Tamron 17-50 (another thread), but, at least with the copies I've had, the 16-85 just isn't in the same league. It's better than the other wide zooms I've had (16-45, the kit 18-55, and some earlier FF zooms like the 28-105 Tamron/Pentax), but the 17-50 sets a fairly high bar by my standards (unless you count the curved focus plane, which is definitely a problem for some real-world applications.) Using the same test chart both sides of the Tamron appear very similar to me and beat the better side of the 16-85, but not by nearly as much as they beat the worse side.
The decentering thing just seems to be something Pentax can't get under control. Really if they want to keep selling higher and higher mp bodies, they must realize that people are going to notice issues like this more easily. At least this lens was close, unlike some truly horrible copies of the 55-300 I've tried.
|
