[tx0h]

hello,

i've got a new 18-135 and now i have some IQ questions.
i know that the 18-135 are known for it's issues but i also
ask myself if this particular lens is in the 'normal' range or
is it worse than normal.

therefor i uploaded three images to the forum and linked
them to the original sized, out of cam images at picaza.
where you can zoom in to a 100% view.

my concerns are the edges and especial the corners.
please click on the images and judge the original sized
images at picaza (use the magnifier lens).

here are the linked test images:

135mm/F5.6


18mm/3.5


135mm/5.6


i really hope for some comments of people who've some
experience with this lens.

thank you,
tx0h

Attached Images

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

View Picture EXIF

File Date

 

[Adam]

Based on that one shot i wouldn't say there's anything wrong with your copy.

[calsan]

I think that's the price you have to pay for such a long zoom range: It does look pretty mushy in the corners, at 100% (which is a how many meters wide print?)!!! Print these at A3 and I'm pretty sure they would look fine.

I know you're testing, but in real life, you would normally stop down for such a flat subject - even with a prime lens.

I would say, shoot again at f/8, increase exposure by 0.7 and process with the lens correction features turned on.
The only lens I can "trust" wide open is my 35 macro, but I would still use at least f /5.6 where the whole image is flat like your subject.
For the above scene with my 15 limited, I still would have gone for f/8.

I think you can still shoot close to wide open for most subjects, but just be aware of the limits of your lenses. Convenience of a long zoom range does come at a price, but you can always work around it.

[VisualDarkness]

That looks perfectly in line with the reviews. A WR super-zoom at that price is bound t have huge compromises. Pentax SMC-DA 18-135mm f/3.5-5.6 ED AL [IF] WR - Review / Lens Test - Analysis

[VladM]

I recently got one of these, and my results are similar to yours.
At wide open apertures centre is pretty sharp and the corners not so.
If you close the aperture down to f8, the results are much better in the corners.
That is the limitatation of zoom lenses in this category.
To get better corner sharpness at wide aperture you really need to use prime lenses.

[kh1234567890]

It does look pretty mushy in the corners, at 100% (which is a how many meters wide print?)!!!

Only 16.4 inches at the usual 300 dpi print resolution.

Is this lens really this cr*p ? The argument that this is the price you pay for the zoom range doesn't quite hold - the old 18-250mm managed better than this.

[VisualDarkness]

Just read the review I linked to and you will find that your findings is pretty much the same.

"The borders are good at f/3.5 but the extreme corners are soft and the very high field curvature doesn't really help here either. Stopping down to f/5.6 boosts the borders to very good levels and the corners well into good territory - technically. However, regarding the field curvature you should stop down to f/11 in order to make sure that the corners are at least somewhat in-focus. THIS DOES NOT SHOW UP IN THE CHART BECAUSE WE REFOCUS THE BORDERS/CORNERS!"

As you shot a flat subject without refocusing for the borders you will have even way lower resolution at the borders than the review graphs shows.

You can compare with your 18-250 which got less field curvature and better edge sharpness but instead if sacrifices center sharpness. It's all a story of compromises and to be fair Canon and Nikon don't fare better when it comes to cheapisch super-zooms in this range and they aren't event weather sealed and often cost more.

[tx0h]

..when it comes to cheapisch super-zooms in this range..

ok, that's all a bit off-topic, but you know what?
if i pay about 590US$ for a lens then i wouldn't
call it cheap and i would expect a decent quality.
it's even not a fast lens. if it would hold F3.5
or better over the whole range i would call it a
cheap lens.

i just like to know if this specific lens is in the
'normal' range or if it is worse then normal.
in fact, i don't think it's a better version :-(

oh, and i made the same images with F8.0 but
wouldn't say they got *much* better then open.

greetings,
tx0h

[VisualDarkness]

Well, nor would I call it a really cheap lens but it's cheapisch compared to what most other camera makers has on the market in this range, and I believe it's the only weather sealed one. Cheap is relative to how the market looks.

[calsan]

Well, cheap or expensive or whatever, I reckon it looks a whole lot better than my sigma 18-125, which I sold for similar reasons.
4x is absolute maximum zoom range I will now consider. 3x would be better.
This is one reason everyone ends up buying primes and putting up with the inconvenience. I reckon much of the problem is that the K5 is very high resolution. I doubt that this issue would be that unacceptable on a K10d.

[calculator01]

Only 16.4 inches at the usual 300 dpi print resolution.

Is this lens really this cr*p ? The argument that this is the price you pay for the zoom range doesn't quite hold - the old 18-250mm managed better than this.

Copy variation, my 18-135 knocks the socks off the samples I have seen from the 18-250. But, some samples of the 18-250 edge out the 18-135 samples.

[tx0h]

Copy variation, my 18-135 knocks the socks off the samples I have seen from the 18-250. But, some samples of the 18-250 edge out the 18-135 samples.

how can a statement like this help me to judge if my copy is a good or bad one?

[normhead]

I have found the 18-135 to be completely consistent with the photozone charts. I have also found that in 3d objects the lens is superior many better lenses rated better on flat charts. Shooting with the subject in front of a flat wall, the shift in focus helps keep the wall behind your centered subject in focus.

Not only that at 24 mm. It's superior on sharpness edge to edge than any lens in it's class, except for the Tamron 17-50.

I have to say, when shooting landscape over 50mm, I take this lens off the camera and put the DA* 60-250 on. However, shooting wildlife, where borders are bokeh and center sharpness is critical, this is a great lens.

SO stop criticizing the lens for what it isn't good at. If you want sharpness border to border below 50mm, you wanted the Tamron 17-50 not this lens. (I have both). If you wanted a top of range 24mm lens, then this is a great lens. If you wanted a great walk around lens, with the ability to give you great landscapes border to border in it's low end and great center sharpness in it's long end, then this is a great lens. But I have to say, some of the lenses being discussed incomparison, don't give you great anything at any focal length, and I'd be at a loss to figure out where I could even use them. Not so with this lens. Learn how and where to use it and it's a great lens. Sit and dwell on it's supposed weaknesses and you can go on for ever, and accomplish nothing.

[crewl1]

Your shots look similar to what comes out of my 18-135.

For me the value of this lens is it's WR, wide zoom range, fast/quiet AF, small size, great color rendering.
My primary use of it is for family outings so edge to edge sharpness is not required.

If you think it is not a good value due to lack of edge sharpness for your type of shooting then send it back and try another but I believe your copy is similar to mine in performance.

[calculator01]

how can a statement like this help me to judge if my copy is a good or bad one?

IMO, if to you it looks better than the 18-250 samples most the time, keep it. If it looks worse most the time, lose it.

But as another member stated, lens aren't really made to shoot charts and each excel at their own tasks.

Disclosure: I am on my phone and don't see any images so I am not commenting on the supplied sample