[carrrlangas]

For Pete's sake...

If you only want to look at some charts based on one specific way of testing, ok. But, from the same review:

MTF (resolution)

Testing a fisheye lens is a bit experimental. The Imatest toolkit freaked out regarding all the wild curves at the image borders at the 10mm setting. The 17mm MTFs should be quite accurate whereas the border resolution figures at 10mm should be only taken as a very rough guidance.
Keeping this in mind the Pentax SMC-DA 10-17mm f/3.5-4.5 ED[IF] Fisheye produced good to very good results in the lab. At 10mm the center performance is already very high at wide-open aperture and the borders follow not far behind. Stopping down produces only marginally sharper results. The local sweet spot is at f/8. At 17mm there's a performance drop - not in the center but the borders suffer a bit. The situation is better at f/8 or f/11 where the border quality recovers to good+ levels.

The Tokina variant in Nikon mount performed somewhat better so the tested sample may not have been the very best one.

If you rather some real world samples, here´s a very unscientific comparison between 10-17 and 18-55WR. both wide open, same exposure, shot in RAW and developed without adjusting anything expect white balance and tint to be the same. The full resolution samples are there for download.

[interested_observer]

I don't believe that the 10-17 was designed to compete with the kit lens. It can't. Let's take the sensor, something near the K5 - 3000 x 4000 pixels. When you swap lenses, the sensor's size does not change, it remains the same. What changes is the focal length (which is a difficult indicator to gauge). A more important aspect to the lens is the Angle of View. The AoV is tremendously larger for the 10-17 than the 18-55. What that translates to is that each pixel has a much larger of area to represent. The 10-17 at the 17mm end has an AoV of 100 degrees, while the 18-55 at the 18mm end has an AoV of 76 degrees. So that is about 50% more area (when taking length x width into account).

So, its 2 different lenses, each of which has their own use, and the types of each of the lenses are quite different, so it's apples to oranges comparison. The similar focal lengths is a real red herring.

HUH?!? since when are we not allowed to compare the sharpness from lenses with a different focal length? (You should mail Photozone quickly; they're out of a job!) Sharpness, or the lack of it, can be compared just fine. The lack of sharpness in the DA 10-17 is not a characteristic of it being so wide. Otherwise the other UWA's would suffer from the same. It's just an indication. The DA 10-17 is not sharper then the 18-55 and the DA 12-24 is ultra-sharp, compared to the 18-55. Simple.

Ok, Came home from work, had some dinner and was reading some posts.....

• HUH?!? since when are we not allowed to compare the sharpness from lenses with a different focal length? (You should mail Photozone quickly; they're out of a job!) - I did not state you were not allowed to compare sharpness from lenses of different focal lengths. I was pointing out, when you have two lenses that differ in their type (fisheye vs. normal) and have a substantial difference in their Angle of View or Field of View (180 - 100 degrees for the fisheye vs. 76 degrees for the wide end of the kit lens), it makes little sense. The fisheye is 30% wider on the diagonal than the kit lens just on Pentax's field of view specification. Of course the fisheye is going to have worse sharpness, and that is what I was pointing out - and also listing the reasons why. Two different lenses were designed for completely different purposes. Their performance point that out.
• Sharpness, or the lack of it, can be compared just fine. - I did and do not disagree, it was just obvious to the casual observer. Folks tend to get confused - a fisheye at 17mm vs a normal lens at 18mm - gee they must be pretty similar. Two completely different projections on to the sensor. Two different uses. Two completely different optical designs - apples and oranges. Two completely different types of distortions, and one is corrected, while the other is not. You can certainly compare sharpness - but to me its a fool's errand.
• The lack of sharpness in the DA 10-17 is not a characteristic of it being so wide.- This is where we disagree. There are two main points in this respect. 1) The size and resolution of the sensor does not change when you swap lenses. Each pixel on the fisheye is going to represent more area than the same pixel with the kit lens mounted. It is just pure physics. 2) In addition, the fisheye has the fisheye distortion that is uncorrected, and will certainly pull the edges and especially in the corners. Even though the fishiness is pretty mild at the 17mm end, its still present. When the two of these are combined, there is no question that sharpness is degraded.
• Otherwise the other UWA's would suffer from the same. It's just an indication. - There are differences between UWAs, in the case of the 10-17, its a fisheye with a Angle of View of 180 to 100 degrees (corner to corner diagonal). The 12-24 at 12mm is 100 degrees wide (at 12mm). Other UWA's also suffer. There are no normal or rectilinear UWA lenses as wide as a fisheye - that is why fisheys lenses were developed. It is a function of each individual lens' actual width or Angle of View. They suffer (in sharpness) in direct proportion to their Angle of View. The wider the lens, the more area each pixel will need to represent, and that will degrade the overall sharpness.
• The DA 10-17 is not sharper then the 18-55 and the DA 12-24 is ultra-sharp, compared to the 18-55. Simple - The 10-17 complements the 12-24. Between the two lenses, you have a seamless coverage of AoV from 180 to 60 degrees. The 12-24 is much sharper than the 10-17. I never said it was. The two lenses however, also are substantially different in their optical design. One (10-17) is an uncorrected fisheye, while the 12-24 is a highly corrected (for distortion) normal rectilinear lens. Comparing the 12-24 to the 18-55 kit lens is easier, since each is a normal rectilinear zoom, and have a common overlap (18-24mm). I would suspect that the 12-24 being highly corrected, with better lens elements, would be sharper. The 12-24 has its optimum resolution at 18mm and it trails off into the 24 area. The 16-45 is sharper at the 24mm mark than the 12-24. The 16-45 is sharper and has better IQ than the kit lens, however that is expected, since that was its design purpose.

________________________________

Ok, not to beat a dead horse errrr cow, here is an image comparison tool. I am using the same images that carrrlangas provided (thanks!)
I used the image comparison tool located at

• Image scraper and comparator - a BackroomWebDesign.com experiment

The left hand is the first image from carrrlangas the 10-17 at 17
The right hand is the second image from carrrlangas the 18-55 at 18

The difference across the three supplied images are increased scaling - equal on each sample, and a shift left and right, respectfully in order to be able to compare the musical notes side by side. Also, assuming that all things are equal in terms of the images (focusing distance, rotation of the cow, etc.)

I think that you can start to see that the width and sharpness is slightly different between the left and right. The elongation of the vertical marks. Some of this is due to the distortion.

So, I'll just leave the rest of the analysis up to the collective.....

[Clavius]

Ok, so now we have both resolution charts AND real world examples that confirm that the DA 10-17 isn't any sharper then the DA 18-55. There is no shame in that at all, because the Pentax kitlens isn't all that bad (https://www.pentaxforums.com/forums/pentax-slr-lens-discussion/234406-defense...-disguise.html) Especially not in comparisson to other brands kitlenses.

[GibbyTheMole]

Ok, so now we have both resolution charts AND real world examples that confirm that the DA 10-17 isn't any sharper then the DA 18-55. There is no shame in that at all, because the Pentax kitlens isn't all that bad

It's pretty darn good, actually. (At least my copy was...) There used to be a guy here in the forum that took superb glamour portraits of professional models with the 18-55. I think his name was "Rambo"... Anyway, he did some impressive stuff with the 18-55.

[reeftool]

I can't see any point at all with taking pictures of resolution charts with any fisheye lens. I can say by real world experience that the DA 10-17 produces quite sharp and crisp images. It was my first lens purchase after I bought my K10D several years ago and I love it as much now as I did then. I also like my DA 18-55 and still use it. The two lenses are so different by design, you really can't do a fair comparison. At 17mm, the distortion isn't severe if you have the subject framed right and by cropping, you can almost eliminate the distortion but nobody should buy this lens thinking that it can be "de-fished" in PP because most of the time, it can't. I like the distortion it produces and I use it for shots where the distortion produces an interesting artistic effect.

A custom motorcycle at a car and tool show a few years ago.


I sold this shot to the Snap On dealers who sponsor the show and they use it in their show flyers.


This was a very close shot and the DA 10-17 is one of the few that you can use in a tightly packed crowd. This is a college womens crew team racing against the clock on ERG machine after the competition was cancelled due to wind conditions. At 17mm and cropped, you need to look at the exif to tell it's a fisheye.


My daughter climbing the rocks on Crane Mountain in the Adirondacks. This shot holds up pretty well when pixel peeping. The DA 10-17 is no slouch.

[erwinsie]

hi all, maybe bit out of topic since i dont own k5 but i have recently purchased 10-17mm with k-x.
need some advise on how to properly use this lens:

when using it outdoor, i got the sky white-ish, and not blue. what should i do? tried to shoot at different f-numbers (Av settings), all the same.
should i shoot with Tv settings ?

[normhead]

Some of my favourite images have been taken with it, but I've heard Pentax reps themselves define it as a "fun" lens, because of the distortion. If you want a "serious" lens for architecture or landscape you want a Sigma 8-16. Like one of the posters above, this was one of the first lenses I bought and it saw a lot of use, before I got the Sigma.

Position your wife off to the side of an image and it makes her look tall and skinny, wives like that....



The distortion can create a great special effect...



Sometimes you can get some really nice "serious " images out of it when things line up just right.



Altogether, a really fun lens that can take some pretty stunning images.

[lytrytyr]

hi all, maybe bit out of topic since i dont own k5 but i have recently purchased 10-17mm with k-x.
need some advise on how to properly use this lens:

when using it outdoor, i got the sky white-ish, and not blue. what should i do?

For most lenses on the K-x,
I dial in an exposure compensation of -2/3 EV,
and then brighten up the shadows in post-processing.
That way, you get more detail in a bright sky.

(BTW, this discussion doesn't really belong in the K5 section.)

[erwinsie]

Ok thanks. Tried to look but other 10-17mm threads were old. I thought my issue is because of this specific lens only, since with my kit lens II get ok photos.

For most lenses on the K-x,
I dial in an exposure compensation of -2/3 EV,
and then brighten up the shadows in post-processing.
That way, you get more detail in a bright sky.

(BTW, this discussion doesn't really belong in the K5 section.)

[LimitedFan]

I bought a 10-17mm this week and had a lot of fun with it on the first walk-around - see below.

[erwinsie]

I bought a 10-17mm this week and had a lot of fun with it on the first walk-around - see below.

I see your two pick, some sky is white. You are ok with that?

[LimitedFan]

I see your two pick, some sky is white. You are ok with that?

Well, there is always space for improvement;-) It's been a quick shot, more or less against the sun, so under non-optimal conditions.

[RGold]

These were taken with my 10-17.

[leitztozeiss]

I just picked one up today. Love it

@10


@17


@10


@17

[Andi Lo]

I see your two pick, some sky is white. You are ok with that?

Are you from Canonikon forum? This is the Pentax forum and we're here for fun

I like the shots posted