[normhead]

The 18-135 got a shocking evaluation from Photozone.de. At the long end the IQ seems to break down to a resolution that corresponds to 2MP, IIRC.

Klaus (from Photozone.de) asked Pentax whether the copy he had tested was defect and received the answer "no".

The 18-250 was the undisputed king of superzooms during its time and never received such a bad review.

I'm happy for people to enjoy their 18-135 but to the best of my knowledge its main feature is the WR. This was also supported by many initial buyers. On the other hand, many people like it and are happy with its IQ. That's fine with me.

Did you even read the test reports?

In the center (not the edges) the 18-135 is rated as Excellent @ 5.6 and near excellent @ F8. By 70 mm the 18-250 is well below the excellent range and doesn't even approach it @135mm. With all due respect to klaus, sometimes he doesn't reflect his own test numbers with any accuracy in his reviews. If you look at the numbers, there is no point in it's range where the 18-250 is rated excellent that the 18-135 is not, the 18-135 even has excellent edges @ 24mm. The 18-250 does not have excellent edges anywhere in it's range. Photozone averaging program used to determine the number of stars it gets works against the 18-135. It simply wasn't designed to give an evaluation of a lens like this.

Maybe it's because it's older, but when you look at the actual numbers for MTF the 18-250 just doesn't compete. If you insist as I do, on judging a lens on what it can deliver not on what it's weakness are the 18-135 kills the 18-250 in every way. The 18-250 simply can't do what the 18-135 based on photozone numbers. You can also say you couldn't have the 18-135 as your only lens from 70-135, for while it's center sharp, it's not edge sharp, but there are lots of great primes in that range. So carry the 18-135 with a 70 or Tamron 90/FA 100WR and 200 2.8 and you're way ahead of the game.

Sometimes you can't read a book by looking at the cover.

18-250 numbers here.

18-135 numbers here.

Have a look, maybe you see something I'm not seeing.

[Docrwm]

Klaus only tested one copy. He was told it was "within spec" by Pentax Germany. Dummies, they should have exchanged the lens. I'm quite confident my copy performs significantly better than the one he tested, as per my test shots shown in the link above. There are favourable reviews of the 18-135mm, like the one on Pentax Forums and these:
Lens Test: Pentax-DA 18-135mm f/3.5-5.6 DC WR | Popular Photography
SMC Pentax-DA 18-135mm f/3.5-5.6 ED AL [IF] DC WR Lens Review

Several people who own both prefer the 18-135, and have talked about improved performance in favour of the newer superzoom. Similar comments can be found in every single thread where the 18-135 is discussed. IMO the weight of these actual owners outweigh Klaus' test of a probably faulty single copy.

I know how the lens fits in with my other consumer zooms. It beats the 18-55 and 18-250, not quite as good as the 55-300, and the 16-45 beats them all. It's a shame the PZ hatchet job has scared so many people away.

+1 absolutely. I keep the 18-250 for those rare instances when I can take ONLY 1 lens and may need that whole range. The 18-135 beats it in nearly every respect for IQ, speed, PF resistance, etc. Klaus' review made Zero sense to me after using the 18-135 and seeing the output from it. Relying on any single review is a mistake IMHO.

[Kozlok]

Got the 18-135 with my K-30, and I really adore it. It's not as good as my limited primes, but close enough for travel with the family.

[Class A]

Did you even read the test reports?

I did and -- on top of it -- I'm aware that the 18-250 figures were obtained with a K10D while the 18-135 figures were obtained with a K-5.

As a result, you cannot compare the numbers. Before you start converting them based on sensor resolution difference, know that Klaus sharpens the images and it is very likely that the sharpening levels are not the same.

Hence, your comparison of the two lenses based on the PZ numbers is invalid.

Note that Klaus says

"The Pentax lens is actually very good to even excellent in the image center but the borders/corners suffer from massive field curvature at the wide end and plain sofness at tele settings. This is certainly no issue for portraits and such but you don't really want use this lens for architecture or landscape photography here. This is also in so far surprising because the (cheaper) Pentax SMC-DA 18-250mm f/3.5-6.3 (a Tamron design) was actually very usable here. "

What can be gathered from the PZ figures is that the 18-250 has a much more balanced performance while the 18-135 loses performance outside the centre rather quickly and dramatically.

Klaus had the 18-135 checked by Pentax Germany and it was found to be " within factory specifications". How dumb would they have to be to not acknowledge an issue or lend him another copy?

To be clear: I've seen very nice images from the 18-135 and I trust that many owners enjoy it a lot. But I doubt that it beats the 18-250 optically. On the other hand, I don't really care personally as I'm much more interested in high quality prime lenses.

---

Last edited by Class A; 06-13-2013 at 08:56 AM.

[HockeyDad]

Note that Klaus says

I'm curious to know if your own experiences with these two lenses contradict or reinforce Klaus's opinion. I'm certain my opinion is swayed by what I shoot. I'm terrible at landscapes and rarely photograph architecture so my preference of the 18-135 is likely based on shooting people, portraits, candids and sports/activities.

[Docrwm]

I did and -- on top of it -- I'm aware that the 18-250 figures were obtained with a K10D while the 18-135 figures were obtained with a K-5.

As a result, you cannot compare the numbers. Before you start converting them based on sensor resolution difference, know that Klaus sharpens the images and it is very likely that the sharpening levels are not the same.

Hence, your comparison of the two lenses based on the PZ numbers is invalid.

Note that Klaus says

"The Pentax lens is actually very good to even excellent in the image center but the borders/corners suffer from massive field curvature at the wide end and plain sofness at tele settings. This is certainly no issue for portraits and such but you don't really want use this lens for architecture or landscape photography here. This is also in so far surprising because the (cheaper) Pentax SMC-DA 18-250mm f/3.5-6.3 (a Tamron design) was actually very usable here. "

What can be gathered from the PZ figures is that the 18-250 has a much more balanced performance while the 18-135 loses performance outside the centre rather quickly and dramatically.

Klaus had the 18-135 checked by Pentax Germany and it was found to be " within factory specifications". How dumb would they have to be to not acknowledge an issue or lend him another copy?

To be clear: I've seen very nice images from the 18-135 and I trust that many owners enjoy it a lot. But I doubt that it beats the 18-250 optically. On the other hand, I don't really care personally as I'm much more interested in high quality prime lenses.

Interesting. I have both, plus 4 Limiteds. I use them on the same camera - a K-5 and now a K-5IIs. The 18-135 is a better optical performer in my experience at 120-130 than the 18-250. Are my limited lenses better optically - I sure hope so. Are they more versatile, nope. Tools are just that - tools. Some are more precise, some more versatile, but none do everything.

[normhead]

Hence, your comparison of the two lenses based on the PZ numbers is invalid.

Yet you can use them to try and justify the 18-250 being better? But just to further comment on this point, the one test I've seen , the 31 ltd, where a lens has been tested on both 10 and 16 mp cameras, it was not as good on 16 MP as it was on 10 Mp. So if anything we can assume that the test numbers would be even more weighted in the 18-135s direction even if they were both tested on the same sensor.

What can be gathered from the PZ figures is that the 18-250 has a much more balanced performance while the 18-135 loses performance outside the centre rather quickly and dramatically.

A more balanced performance? You mean because the 18-135 is so much more center sharp than the 18-250 that the difference between the center and edges is more dramatic, even though at 24mm the 18-135 is probably sharper on the edge than the 18-250 is in the middle? That's an interesting way of looking at things... not one I favour for myself, but if that's what is important to you then, I understand your perspective. At 24 mm the 18-135 is sharper than any comparable Pentax lens center and edge, and is limited only by it's speed, it's just not that fast. The only lens withing 4mm either way is the Tamron 17-50. It's worth buying just for it's performance just at 22mm to 30mm. I don't see any particular reason to buy the 18-250 at all. I insist that every lens I own give me something other lenses don't, that they have a unique characteristic that makes them worth purchasing. There is no where in the 18-250s range, that you can't find 4 or 5 other lenses that are better. That's just not the kind of lens I buy. And the 18-135 certainly isn't that kind of lens.

On the other hand, I don't really care personally as I'm much more interested in high quality prime lenses.

I don't really care that a lens is prime or zoom. I just want the best at that focal length. if it's a zoom it's a zoom, if it's a prime it's a prime. I don't get all doctrinaire about these things.

---

Last edited by normhead; 06-13-2013 at 09:28 AM.

[Class A]

Yet you can use them to try and justify the 18-250 being better?

Without looking at absolute numbers one can easily see that the 18-250 has a more balanced performance in terms of centre vs borders/corners sharpness.

But just to further comment on this point, the one test I've seen , the 31 ltd, where a lens has been tested on both 10 and 16 mp cameras, it was not as good on 16 MP as it was on 10 Mp. So if anything we can assume that the test numbers would be even more weighted in the 18-135s direction even if they were both tested on the same sensor.

You may want to check the tests again. On the 10MP sensor, the 31mm achieves 1848.5 LW/PH in the centre @ f/1.8. On the 16MP sensor, the 31mm achieves 2403 LW/PH in the centre @ f/1.8.
If you scale the 18-250 figures with the factor implied by the above figures, it outperforms the 18-135 numerically.

If you determine the factor by looking at the f/8 performances of the 31mm on the two different sensors then the 18-250 does not outperform the 18-135 but comes very close.

Again, I don't really care. But please don't cast a bad light on the 18-250 with false claims.

[StephenHampshire]

Without looking at absolute numbers one can easily see that the 18-250 has a more balanced performance in terms of centre vs borders/corners sharpness.


You may want to check the tests again. On the 10MP sensor, the 31mm achieves 1848.5 LW/PH in the centre @ f/1.8. On the 16MP sensor, the 31mm achieves 2403 LW/PH in the centre @ f/1.8.
If you scale the 18-250 figures with the factor implied by the above figures, it outperforms the 18-135 numerically.

If you determine the factor by looking at the f/8 performances of the 31mm on the two different sensors then the 18-250 does not outperform the 18-135 but comes very close.

Again, I don't really care. But please don't cast a bad light on the 18-250 with false claims.

The OP asked if anyone had owned/used both these lenses, several of us who have responded that we had. My own copy of the 18-135 performs much better optically than the 18-250 Tamron version that I had. It seems sharper in "real world" use and certainly has more punch (due to better microcontrast I assume) It could be that I got a relatively bad copy of the 18-250 and a good one of the 18-135. I certainly think that PZ got a duff copy of the 18-135 (despite what the distributor said!) as the vast majority of users here on PF seem to report positively on this lens. I know I was vary wary of buying the 18-135 BECAUSE of the drubbing PZ gave it, but I got a good deal on a lens split from a kit.
Conversely, my 55-300 performs worse than the PZ review, again I assume they got a better copy than mine. My 70mm Ltd seems to match the PZ review.
I think the bottom line is that I put more store on the overall tone of the reviews of users of this forum (and this can aslo highlight lense where there is a wide variation in user experience)
So I stake my claim that MY 18-250 was optically inferior to MY 18-135. No false claims here

[Class A]

No false claims here.

Please note that my comment was meant to be directed at normhead, no one else.

[kevingt01]

I own the 18-250mm and occasionally use it as a walk-around where I anticipate needing some reach. I just ordered a K30 though and would like to swap my 18-250 for the 18-135. Looking at a few asking prices on the marketplace, it looks like they have roughly the same cost/value. Does that sound about right?

[normhead]

Please note that my comment was meant to be directed at normhead, no one else.

So If I quote Stephen Hampshire instead of quoting photo zone am I still wrong? When I was talking about the 31 not doing as well on 16 MP, I was talking about how it performed as compared to the maximum estimated possible performance. At 10 MP the performance was of the 31 was off the scale. If you take the 18-250 as a percentage of what's possible and apply that to 16 MP, it doesn't come close to the 18-135. Listen, I'm just passing off what I noticed. If your don't see what I'm saying, fine , ignore me. I don't care. You could have just as easily looked at the numbers and seen what I saw. You chose, for some reason I can't comprehend not to. No problem, if you can't see what I;m saying it's no sweat off my back. But nothing you said changed my opinion. I have too many reasons for disputing how you did your math. And I'm quite comfortable with how I do it.

However, you might want to notice that people with practical experience with both lenses would seem to confirm my opinion. And I'm certainly not going to buy a Pentax 18-250 just to prove I'm right.

[Class A]

If your don't see what I'm saying, fine , ignore me.

I'd love to but I don't like the thought of someone else reading your posts and not suspecting your numbers and hence getting the wrong idea.

You could have just as easily looked at the numbers and seen what I saw.

I looked at the numers, but I don't see what you saw, because I know that you cannot compare the numbers (as they are from two different cameras).

It would be good for you if you understood that you cannot compare the PZ measurements. I alluded to ways in which they could be compared and then the lens performances are rather similar. If you have any objections against what I did, it would be constructive to spell them out, rather than stating "I'm sticking to my opinion".

[normhead]

OK, here's how you do it. Look at the top of Klaus' scale and the lens score then divide the lens score by the score at the top of the scale , expressed as a percent...then take the 16 MP top of scale score and multiply by the percent expressed as a decimal. And while Kluas says you can't directly compare, what you can do.. make a number of comparisons where you work out the comparative numbers. The grab a couple of lenses and see how your results relate to real life. You start to get an understanding of what the numbers mean, through practical experience.

So once you're look at 10 lenses or so, you start to understand the relationships between the numbers. I often quote numbers because they explain what I'm saying, what my experience, so based on all of that, not just Klaus' numbers, I can look at the Pentax 18-250 and tell you what I told you. There is absolutely no evidence that I'm wrong here, and at least one maybe two persons who also say I'm right in my interpretation.

You might think you're saving someone from me. I'd say, you need to save your self. Others can look after themselves. I find it amusing you feel you need to save people from me. Save yourself brother. You might be wrong here. You might be trashing me for nothing.