[Peter Herman]

I really WANTED to love the Pentax DFA 15-30... (I shoot landscape genre, primarily).
Upon receiving my second copy (first similarly failed my tests) of the DFA15-30 from Amazon (B&H), I conducted very careful (tripod, e-shutter, delay timer, careful focus using max-zoomed live-view) lens tests and found serious trouble at ~20mm f8 (which should be a sweet spot!). It did better at 15 and 30. Online reports like HD PENTAX-D FA 15-30mm F2.8 ED SDM WR Review | ePHOTOzine don't hint at any situation like this.

As shown below, in the center area of frame, my copy of the DFA15-30 could not come even close to matching a 20 year old FA20-35 at f8 - the DFA is SIGNIFICANTLY softer in the important center area. Incidentally, my first copy result was similar. It began to outperform the old FA by f13, but that seems ridiculous to me, and doesn't jive with some online review test results which showed peaking resolution at f8. And is not acceptable for a $1200 new-technology lens vs. an affordable OLD old lens. Below are side-by-side high-zoom (like 200-300%) comparisons (lens noted on title bar – some are in reversed order). So the good news is that I already have a MUCH lighter, smaller, easier to carry ultrawide zoom in the FA20-35 (which is ALWAYS in my bag)! I should be happy, I suppose! Maybe I just have a "golden" FA20-35 and my expectations for the DFA15-30 (at 20mm) are unrealistic! But I doubt. I'm tempted to try a third time, but, maybe I just stick with my FA20-35 plus Rokinon 14mm for the wide end (which is competitive with the 15-30 from my tests, although I prefer the 15-30 rendering at 15 vs. rokinon14).

Just for context- I've attached a "full-image" shot (DFA at 20mm f8) from the indoor scene, just for context. Just to reiterate- both images were carefully and precisely focused on the center objected using max-zoomed live view (i.e. no "autofocus error" going on).

Any others out there who have carefully tested the DFA15-30 at ~20mm f8 and compared to an FA20-35 or even A20?
________________________________________________________________________

UPDATE - third lens received, tested, and passed muster.. "if at first you don't succeed, try, try again...". (Thank GOODNESS for easy-peasy no-cost-to-me return via Amazon). Seems odd that I received TWO problematic lenses (second one was even worse than first). But the third one is good, and in line with what other (glowing) reviews report, like NORTHCOASTGREG et.al. At 15mm, wide open, I do find some coma on stars in all corner areas (radial away from center) - I'll have to do some head-to-head compares with my beloved Rokinon 14, but my intent was to switch to this 15-30 for all astroscape. I haven't tested the 20-30mm range - perhaps improved and that gives great option for astroscape framing..

Attached Images

               

---

Last edited by Peter Herman; 12-06-2019 at 05:37 PM.

[aurele]

the rendering of the big red ... thing picture pretty much indicates that the lens as assembly issue : the lens should'nt be soft like that at all.

Just send it back, get another one from a reputable dealer, like B&H or Adorama.

[tvdtvdtvd]

(first similarly failed my tests

Are you sure they didn't send the same lens back?

[Peter Herman]

the rendering of the big red ... thing picture pretty much indicates that the lens as assembly issue : the lens should'nt be soft like that at all.

Just send it back, get another one from a reputable dealer, like B&H or Adorama.

That "big red thing" was a rosemaled plate located at top-center edge, fyi. But yeah, seems really really bad. I even re-shot it thinking I must have jiggled camera. First copy was procured from Adorama. Second attempt (6 months later) from B&H, so no way I got same lens back!

---------- Post added 11-28-19 at 11:48 AM ----------

Are you sure they didn't send the same lens back?

First copy was procured from Adorama. Second attempt (6 months later) from B&H, so no way I got same lens back!

[pschlute]

Send it back, it is a faulty lens.

This is what the 15-30 should look like (this is at 15)

[jatrax]

Peter I had what I thought was a quite good copy of the FA 20-35. I purchased the DFA 15-30 for an astrophotography project thinking that I would continue to use the FA 20-35 for most of my landscape work as it is so much lighter. But after comparing images from both I sold the FA 20-35. The DFA 15-30 was clearly sharper and produced better images under similar conditions. It also does not have the purple fringing the 20-35 often showed.

Anyway, there is clearly something wrong with the lens copy you have. Many of those images you posted are too dark for me to make out clearly on my monitor but regardless there is no reason at all that any of those images should not be pin sharp with that lens. Have you tried a fine focus adjustment? And are you using manual, live view or view finder focusing?

I would send it back and try another copy, there have not been a lot of reports of issues with this lens and I can say from experience that in my case the DFA 15-30 clearly outperformed my FA 20-35.

[Peter Herman]

Peter I had what I thought was a quite good copy of the FA 20-35. I purchased the DFA 15-30 for an astrophotography project thinking that I would continue to use the FA 20-35 for most of my landscape work as it is so much lighter. But after comparing images from both I sold the FA 20-35. The DFA 15-30 was clearly sharper and produced better images under similar conditions. It also does not have the purple fringing the 20-35 often showed.

Anyway, there is clearly something wrong with the lens copy you have. Many of those images you posted are too dark for me to make out clearly on my monitor but regardless there is no reason at all that any of those images should not be pin sharp with that lens. Have you tried a fine focus adjustment? And are you using manual, live view or view finder focusing?

I would send it back and try another copy, there have not been a lot of reports of issues with this lens and I can say from experience that in my case the DFA 15-30 clearly outperformed my FA 20-35.

Jatrax,
Thanks - I was using live view focusing, so no chance for focus error in my tests. I already have it boxed up to return, and have replacement in my "cart"- I'll try, try again, based upon your and pschlute's comments - especially yours, since you also were also a 20-35 owner/fan. Incidentally, I use my Rokinon 12, 14, and 24 for astrophotog- the 20-35 would fail miserably.

---------- Post added 11-28-19 at 12:31 PM ----------

Send it back, it is a faulty lens.

This is what the 15-30 should look like (this is at 15)

Ouch - I think you cut my eye a little! Yeah, that appears bloody sharp (although sometimes, at screen-res, anything looks sharp- at 200-300% zoom, does it still look sharp?). And by the way- at 15mm, my copy seemed fine/sharp too. The problem was at 20mm f8.

[pschlute]

anything looks sharp- at 200-300% zoom

Never, ever look at an image at greater than 100% view. 1 pixel for 1 pixel is the maximum you can use to analyse.

once you go over 100% your image can only degrade.

[Peter Herman]

Never, ever look at an image at greater than 100% view. 1 pixel for 1 pixel is the maximum you can use to analyse.

once you go over 100% your image can only degrade.

Maybe I’ll agree for looking at SINGLE IMAGE. But when doing side-by-side compares (between lenses, as in above comparison screen-grabs), I find ~200% makes the differences more immediately apparent at normal viewing distance on high-end monitor, so I disagree with you on that “never ever” point.

[limestonecowboy]

I haven't used mine enough to be definitive but loving the 15-30mm so far - NAL 13-11-2019 17-23-58 | Nick L | Flickr

[pschlute]

Maybe I’ll agree for looking at SINGLE IMAGE. But when doing side-by-side compares (between lenses, as in above comparison screen-grabs), I find ~200% makes the differences more immediately apparent at normal viewing distance on high-end monitor, so I disagree with you on that “never ever” point.

Your picture proves that point. But you could easily have provided a 100% crop.

[C_Jones]

If I was testing something like the 15-30 I would definitely use the center point AF area, AF in Continuous mode, and utilize the shutter button via viewfinder for autofocus function. I would also use a tripod and focus with the center point on a specific part of the subject area when testing. Not to say that there is anything wrong with testing with live view. Just a comment on how I would approach it.

Good luck with getting a lens that works well for you.

[pschlute]

If I was testing something like the 15-30 I would definitely use the center point AF area, AF in Continuous mode, and utilize the shutter button via viewfinder for autofocus function. I would also use a tripod and focus with the center point on a specific part of the subject area when testing. Not to say that there is anything wrong with testing with live view. Just a comment on how I would approach it.

Good luck with getting a lens that works well for you.

Everyone has their preferences for focussing. LV and manual focus when zoomed beats AF in my view hands down.

Certainly at the 15mm end I mostly use LV and turn off AF.

[mikesbike]

LV or no LV, it does look like a focus issue to me. Also, I wonder why so many examples shown are so dark?

[Peter Herman]

If I was testing something like the 15-30 I would definitely use the center point AF area, AF in Continuous mode, and utilize the shutter button via viewfinder for autofocus function. I would also use a tripod and focus with the center point on a specific part of the subject area when testing. Not to say that there is anything wrong with testing with live view. Just a comment on how I would approach it.

Good luck with getting a lens that works well for you.

The best way to focus for lens testing, in my opinion, is to use live view with digital zoom to the maximum (on the object at center). I never use auto focus for a lens test. Live view shows you exactly what the sensor will be seeing, except for the stopped down aperture DOF (which can also be seen/previewed by a rocking the shutter switch).

---------- Post added 11-28-19 at 08:17 PM ----------

LV or no LV, it does look like a focus issue to me. Also, I wonder why so many examples shown are so dark?

Trust me, it was NOT a focus issue. I know how to focus accurately (manually, using zoomed live-view) for lens testing which is the only proper way in my opinion. The darkness of a few images (clipped from much much larger image, of course) was because those areas were literally in the dark- I exposed for highlights.. SOOC jpeg, so no lifting of shadows, etc.