[JHD]

I sold my DA15 because it wasn't wide enough. It's easy to get caught up in the whirlwind of popularity and numbers, only to end up with the wrong lens.

[normhead]

After looking at the images and considering the options, I don't see how I am going to get away without both. LBA continues. The 8 mm images are so much better than I expected.

It's quite possible that I will need 12 or 13 in a corrected aspherical lens to match the 15 I wanted, taken with my fisheye.

But, being able to throw the 15 ltd in my bag, on a long hike. Is also very desirable. And Adams review is very favorable.

I'll probably buy the 15 first... almost all of the photos we sell are taken on canoe trips and hikes. If I can get that 15 this summer, it will make a difference right away. But the 8-16 has definitely made it on to the list. The one issue left being whether to sell the 10-17. There are many images where the fisheye distortion makes an otherwise good image unusable, there are other images where the distortion is what makes the shot special.

Thanks for all the input. Feel free to keep the thread alive. At some point, I'll come back and post some comparison images. The 8-16 at 15 mm, the 15 Ltd and the 10-17 @15mm. At some point I'll be taking a few shots to analyze. so I may as well post them if I get something useful.

[Marc Sabatella]

But I plan to use it on my E-PL2 now that I got a DA adapter, so I will be able to see better if the problem is sharpness of just the fact that AF is unreliable on the Pentax bodies (I cannot MF this lens with the OVF).

Again, controlled tests would resolve this quite easily. But my guess is that it's a simple matter of not focusing carefully enough. The well-known fact that the focus sensors are rather large becomes more and more of an issue the wider the lens, because objects are rendered smaller and smaller.

From the samples and reviews that I saw, the Samyang 14/2.8 may be the sharpest wide lens available for Pentax. My DA 15 is not as sharp as the Samyang samples I saw around.

Comparing images of different scenes from different photographers with different cameras and different settings and processed with different software - there is no possible way one can make any conclusions whatsoever about lens differences. That's why controlled tests are so important.

[Marc Sabatella]

If you mean MTF numbers, check 16-45 results at Photozone.
Sharper borders than DA 15mm below f/8

Perhaps, but given that the borders are only about 5% of the picture, and this still is only true at the apertures where one would not actually typically shoot a scene when looking for border-to-border sharpness - it hardly constitutes a case of the zoom being better than the prime, even when limiting consideration to sharpness only.

[Marc Sabatella]

Well, I just used the DA 15 on my E-PL2 today and I now changed my mind about it.

Glad to hear it! Sorry, I posted my other reply before seeing this (I usually try not to do that), but I guess my comments there are probably still valid.

So the DA 15 is indeed a sharp lens and my different impression seems to be caused mainly by focusing issues.

Could be, but without controlled tests, it's going to be impossible to say. As I mentioned in my previous response, focus with ultrawides is always tricky, and not because of any specific fault with any cameras.

[starstuff]

I wish it were easier to get my hands on these lenses for side-by-side comparison. I've been longing for an ultra-wide too, and physical size and weight are critical factors for me, so I've been thinking about the DA 15mm. What I'd really like to do is to see the lens next to the Sigma 15mm fisheye (according to specifications just a tad bigger/heavier), shoot for a few minutes with each to get a sense of how they feel. Sure do wish there were a nearby camera store that sold both. Does anyone here have anything to say about these lenses compared to one another?

[Laurentiu Cristofor]

Glad to hear it! Sorry, I posted my other reply before seeing this (I usually try not to do that), but I guess my comments there are probably still valid.

No problem. I added a redirect post.

Could be, but without controlled tests, it's going to be impossible to say. As I mentioned in my previous response, focus with ultrawides is always tricky, and not because of any specific fault with any cameras.

Focusing is tricky *on DSLRs*. It is not at all tricky with MILCs where you can zoom into the area you are attempting to focus on.

As for testing, it can be misleading too. If we compare two lenses and one scores a max of 40lpmm in one out of ten trial shots, while another scores the same in 8 out of 10 tests, they will both look equivalent in test scores, but the effort to get good results will not be revealed by their equal score.

It does not matter to me how sharp a lens is in lab conditions because my photography is not done in lab conditions - if I cannot focus a lens reliably under regular use, what good is its *potential* sharpness to me?

so there is no real point looking for verification every time someone posts something most people already believe or know to be true. It is when individual results vary from what is generally known to be true that one rightly *should* question those outlying results.

You should be careful with this line of thinking - it leads to herd mentality.

The thing specifically relevant to lens testing, though, is that it is easy to stuff up a test to make a lens look worse than it actually is. Whereas it is next to impossible to stuff up a test to make a lens look *better* than it actually is. If one given controlled test shows great sharpness, then that lens *is* capable of great sharpness, period, and no further tests will cause that to not be true.

See above. Capability is not an indicator of performance. You can dig a trench with a spoon, but it's not going to be efficient. A sharp lens that misfocuses will be worse than a slightly less sharp lens that focuses perfectly. What matters more is the rate of keepers you get from a lens, or otherwise put, its reliability in producing high quality images.

In the specific case of the DA, there may be a combination of focusing issues with the fact that it is a relatively slow lens, so it will have a harder time nailing focus in low light. For example, in one set of tests, I tried focusing on a print and I used the ability to see the printing dots as a resolution bar. With the E-PL2 I could see the dots while focusing and could easily take images that had such detail. With the K-7, I couldn't get those dots to show up - I couldn't see them in the viewfinder and AF seems to have consistently failed to focus on the print surface. This was done late afternoon, so light wasn't great. I moved to a table with a lamp and I repeated the test - the K-7 was more reliable now and scrutinizing a crop, I could see all the detail I could hope to record.

My conclusion at this moment is that the DA15 is as sharp as I need it to be, but it requires accurate focusing to benefit from that sharpness.

[audiobomber]

Perhaps, but given that the borders are only about 5% of the picture, and this still is only true at the apertures where one would not actually typically shoot a scene when looking for border-to-border sharpness - it hardly constitutes a case of the zoom being better than the prime, even when limiting consideration to sharpness only.

The measurement may be taken at 5%, but the actual blur covers a lot more of the frame than 5%, as shown in the test photos I've posted several times. I have also stated several times that the DA15mm beats the 16-45 for contrast, flare, distortion and CA performance. But it is not as sharp as the 16-45 at 16mm, across the frame at any aperture (on my lenses). You have claimed that the prime sharper in the center, but I can't see it. Of course, one would have to print to a very large size for anyone to see a difference in resolution, whereas the other parameters are fairly obvious. The DA 15 is the better lens, there's no question about it.

I would also say that my 16-45 is sharper and has much better CA performance than the one tested by PZ. You can view these at 100% to see for yourself. The difference in CA between my DA 15 and my 16-45 is very subtle.
https://picasaweb.google.com/bonhommed/Comparo?authkey=Gv1sRgCLOD9LjmoOKTlAE...99863119344722
https://picasaweb.google.com/bonhommed/Comparo?authkey=Gv1sRgCLOD9LjmoOKTlAE...18589306741906

I'm not trying to say the 16-45 is the better lens, it is not. But the way people carry on about the 15mm is a little out there IMO, and the 16-45 deserves more respect than it gets.

[jsherman999]

After looking at the images and considering the options, I don't see how I am going to get away without both. LBA continues. The 8 mm images are so much better than I expected.

It's quite possible that I will need 12 or 13 in a corrected aspherical lens to match the 15 I wanted, taken with my fisheye.

But, being able to throw the 15 ltd in my bag, on a long hike. Is also very desirable. And Adams review is very favorable.

I'll probably buy the 15 first... almost all of the photos we sell are taken on canoe trips and hikes. If I can get that 15 this summer, it will make a difference right away. But the 8-16 has definitely made it on to the list. The one issue left being whether to sell the 10-17. There are many images where the fisheye distortion makes an otherwise good image unusable, there are other images where the distortion is what makes the shot special.

Thanks for all the input. Feel free to keep the thread alive. At some point, I'll come back and post some comparison images. The 8-16 at 15 mm, the 15 Ltd and the 10-17 @15mm. At some point I'll be taking a few shots to analyze. so I may as well post them if I get something useful.

If I were going to have two WA choices for the same mount, it would be the 15ltd + Sigma 8-16.

.

[northcoastgreg]

But [the DA 15] is not as sharp as the 16-45 at 16mm, across the frame at any aperture (on my lenses). You have claimed that the prime sharper in the center, but I can't see it

I own both lenses, and that's not quite been my experience. In the center and in most of the frame, the DA 15 is one of the sharpest lenses I have ever used. It's as sharp as my K 50/1.2 or DFA 100/2.8 WR. It loses some resolution toward the corners, but since it's so sharp in the center, it can lose some resolution and still be sharp. My 16-45 is not as sharp in the center as my DA 15, but often cannot detect any loss resolution toward the corners when the lens is stopped down, which is quite amazing for a zoom.

I think both these lenses are plenty sharp, and whatever differences may exist are trivial. The DA 15 has more microcontrast and better color rendition; it does a better job handling/taming difficult light; it tends to produce more spectacular results than the DA 16-45.

[selar]

I sold my 16-45 a long time ago, but while I had it, I had no complaints about it at the wide end: 16mm at f5.6

[cali92rs]

I'll throw in my 2 cents just to muddy the waters. I have the Sigma 8-16mm and I love it. But that thing is BIG. It's not much smaller than my DA*50-135mm.

Also, Lightroom 3 does a good job correcting vignetting and distortion if you are willing to modify the exif data to read at 8mm (thanks Pentax for that "oversight"). You can read more about it, and my struggles here:
https://www.pentaxforums.com/forums/post-processing-articles/176330-sigma-8-1...ally-8-mm.html

[selar]

You can read more about it, and my struggles here: Sigma 8- 16 mm lens correction / metadata in PS & LR - especially at 8 mm

Yes, I remember that post, most useful post, but coudn't find it later when I bought the 8-16. Thanks for surfacing it.

[ihasa]

Well, I just used the DA 15 on my E-PL2 today and I now changed my mind about it. Here are my new observations:

- The DA 15 is indeed very sharp if focused correctly (being a wide lens, it might sound like focus shouldn't matter that much, but at close distances it does, even at f/8)
- Even wide open at f/4, focusing towards the borders, sharpness is very good. (I didn't focus on a flat surface so I couldn't check the field curvature)
- The E-PL2 makes it easier for me to manually focus accurately - this was something I was struggling with on the K-7. With the Pen, even if I don't take an image, I can see the sharpness of the lens while focusing it in LiveView, by watching how details pop out in focus.
- My K-7 might backfocus a bit with this lens, but I will have to look more into this. To get over any issues, I shot a page at an angle, so that I could have something in-focus anyway - the in-focus area of the K-7 came out very well.

Thanks Laurentiu. Helpful observations! I haven't noticed any focussing issues specific to the DA15mm, but this lens and my Sigma 10-20mm seem to benefit from a little check in live-view on my K7. Sometimes I think the AF just doesn't lock on properly and totally misses. Roll on the K30 with choice of CDAF / focus peaking and PDAF / OVF!

[Marc Sabatella]

Focusing is tricky *on DSLRs*. It is not at all tricky with MILCs where you can zoom into the area you are attempting to focus on.

Point taken, although the distinction here isn't really DSLR vs MILC - this technique also applies to DSLR's with live view.

As for testing, it can be misleading too. If we compare two lenses and one scores a max of 40lpmm in one out of ten trial shots, while another scores the same in 8 out of 10 tests, they will both look equivalent in test scores, but the effort to get good results will not be revealed by their equal score.

What's misleading about that/ If the original testing methodology was so flawed that it could only produce good results in 1 out of 10 trials, why penalize the lens for that?

It does not matter to me how sharp a lens is in lab conditions because my photography is not done in lab conditions - if I cannot focus a lens reliably under regular use, what good is its *potential* sharpness to me?

The thing is, you *can* learn to focus accurately and well. It is not rocket science. It just requires a moment's thought on the scene, reflecting on the lessons one has learn from previous experience about how your AF system works.

You should be careful with this line of thinking - it leads to herd mentality.

True, one should take care. Doesn't invalidate the basic concept, though/

See above. Capability is not an indicator of performance. You can dig a trench with a spoon, but it's not going to be efficient.

True, but a spoon is *repeatably* bad at digging trenches. it's not the fault of bad methodology, it's not user error of any kind: the spoon will not ever be as efficient at digging trenches as, say, a shovel. There is nothing inconsistent here: a spoon is *always* less good than a shovel at this. But this has nothing to do with the case at hand. If the DA15 is capable of as good a shot as some other lens, then it *always* as good as that other lens. Only user error would produce different different results. Whereas with the spoon, it isn't user error - it really *isn't* as good.

A sharp lens that misfocuses

Yes, but we're not talking about defective lenses here. assuming I even believe it is possible for a lens to misfocuson a Pentax DSLR, which I am still very much in doubt about. OK, yes, if you happen to have a provably defective lens, then that particular copy of that lens isn't as good as some other lens. But still, most copies of the lens *will* be that good.

it requires accurate focusing to benefit from that sharpness.

Well, of course - that is true of every lens ever made.