DA 18-135 on photozone

frogfoot · 04-21-2011, 10:58 AM

Originally posted by Parallax

Was SR off?

Oops.. Just checked Photome.. I forgot to turn it off

.. Will redo the test.

Menno · 04-21-2011, 11:01 AM

Originally posted by frogfoot

This is what I call unusable. Comparison between 18-135 and 50-200: 100% crop of right lower corner.
BTW at wide to middle range my 18-135 is noticeably sharper and more contrast than 16-45 (I'm talking about center. Corners are still weaker). So I don't know if my lens can be called faulty.

P.S. After more careful examination I can say that 18-135 is not worse than 50-200 in the range of 50-100 mm.

Well that's exactly the type of image I get too, see my link above. Wide open at 18mm I can get the same soft corners but then it depends on the scene (field curvature?) It's not motion blur. I see this from about 80mm and up on the corners and above 100mm also the center gets less sharp than on the DA50-200 and quite some CA appears.
But from 50 to 60mm my DA18-135 is much better than my DA50-200. From 18 to 70mm it mostly beats my 18-55/50-200 combo when stopped down so it's probably a keeper. For the good teleshots I will take the DA50-200 (or I'll buy the DA55-300) with me but I will have to switch much less often.
Most reviews mention this behavior so my guess it's the design of the lens, or there are more bad than good copies..

Pentaxor · 04-21-2011, 11:10 AM

I think the test should be redone. when using a tripod test, make sure to turn SR off or use a 2-sec timer for the shot.

my initial assessment for both shots however is that both shots are underwhelming, much so with the 18-135. if these were corner crops, I wouldn't like it at all. I'm not even sure if these were decentering issues since the images from 18-100 are what you say are acceptably very good to good. I would say that anything past 100mm is the lens' main weakness, thereby saying that 35mm went to waste considering the results here were in fact clearly ugly. so this lens became more of a 18-100mm lens which is bad.

frogfoot · 04-21-2011, 11:18 AM

Originally posted by Menno

For the good teleshots I will take the DA50-200 (or I'll buy the DA55-300) with me..

That sounds reasonable. I'm changing my 50-200 to 55-300 too. And I decided to add margin while shooting when it's possible, to crop bad edges during processing. Good center sharpness of 18-135 and 14.5 MP of my K-7 allows to do that.

frogfoot · 04-21-2011, 11:25 AM

Originally posted by Pentaxor

I think the test should be redone. when using a tripod test, make sure to turn SR off or use a 2-sec timer for the shot.

Ah.. I used 2 sec timer with SR on. So I don't need to retest..

**hyyz** · 04-21-2011, 11:46 AM

I noticed in general, a zoom lens (except ultra-wide) is weaker at long end than at wide end. I don't know why, but this seems to be a general fact. Actually, many criticisms to zoom lenses go to long ends.

And if we take 2 lens with comparable grade but in different range, e.g. 16-50 vs 50-135, we probably will notice 16-50 @ 50mm is weaker than 50-135 @ 50mm. And 50-135 @ 135mm will be weaker than 60-250 @135mm.

Above may not be true for all such comparisons, but it seems to be quite common. This behavior is probably designed on purpose.

IMHO, it is not quite fair to compare 18-135 @ 135mm (which would be its weakest link) to 50-200 @ 135mm (which is probably not its weakest link).

Pentaxor · 04-21-2011, 12:18 PM

Originally posted by hyyz

I noticed in general, a zoom lens (except ultra-wide) is weaker at long end than at wide end. I don't know why, but this seems to be a general fact. Actually, many critisms to zoom lenses go to long ends.

And if we take 2 lens with comparable grade but in different range, e.g. 16-50 vs 50-135, we probably will notice 16-50 @ 50mm is weaker than 50-135 @ 50mm. And 50-135 @ 135mm will be weaker than 60-250 @135mm.

Above may not be true for all such comparisons, but it seems to be quite common. This behavior is probably designed on purpose.

IMHO, it is not quite fair to compare 18-135 @ 135mm (which would be its weakest link) to 50-200 @ 135mm (which is probably not its weakest link).

this is true with regards to longer end IQ and I also believe it is not fair to make such comparisons. although one obvious criticism about the lens is not because how it performed against another lens but rather how poorly it performed at the longer end at it's own. it is not what you even call as acceptable nor mediocre but plain poor performance which makes a person hesitate on spending $500 dollars on such a lens. personally, I would rather have a 2 lens combo for less or same amount with a much better result.

audiobomber · 04-21-2011, 02:26 PM

I just did a quick and dirty test vs my 55-300mm. Four shots, I used the best two. No tripod but speeds were high in full daylight (1/750s). The target was the base of the bird bath; 135mm focal length @ f5.6, the 18-135 was wide open, the 55-300 was stopped down (would normally show f4.5 at this focal length). Click on the magnifying glass for a full resolution view.

https://picasaweb.google.com/bonhommed/18135Vs55300#

I've now tested the 18-135 at both ends of its range and wide open vs the 18-55 II, 18-250 and 55-300. It's noticeably better than the 18-55 and 18-250. I haven't done much to compare vs the 55-300 but from this result it seems to hold its own. So far the 18-135mm is all I could hope for, I guess I'm just one lucky SOB.

Pentaxor · 04-21-2011, 02:39 PM

Originally posted by audiobomber

I just did a quick and dirty test vs my 55-300mm. Four shots, I used the best two. No tripod but speeds were high in full daylight (1/750s). The target was the base of the bird bath; 135mm focal length @ f5.6, the 18-135 was wide open, the 55-300 was stopped down (would normally show f4.5 at this focal length).

https://picasaweb.google.com/bonhommed/18135Vs55300#

I've now tested the 18-135 at both ends of its range and wide open vs the 18-55 II, 18-250 and 55-300. It's noticeably better than the 18-55 and 18-250. I haven't done much to compare vs the 55-300 but from this result it seems to hold its own. So far the 18-135mm is all I could hope for, I guess I'm just one lucky SOB.

I believe it would be better if you did the same magnification rather than to rely on the 55-300's 135mm marking. I did a similar 135mm test with other lenses before and since the 55-300 seems to show more magnification than the other 135mm's, I adjusted it to atleast show the same degree of magnification and show a much precise comparison. also, I would try and compare at same aperture openings like f5.6 and f8.

audiobomber · 04-21-2011, 02:39 PM

Originally posted by Pentaxor

I believe it would be better if you did the same magnification rather than to rely on the 55-300's 135mm marking. I did a similar 135mm test with other lenses before and since the 55-300 seems to show more magnification than the other 135mm's, I adjusted it to atleast show the same degree of magnification and show a much precise comparison.

Point taken, but hearing how the 18-135 is being gutted in the forums you would think there would be no comparison, whether the same magnification or not.

Originally posted by Pentaxor

I would try and compare at same aperture openings like f5.6 and f8.

The aperture was f5.6 for both lenses.

Pentaxor · 04-21-2011, 02:46 PM

Originally posted by audiobomber

Point taken, but hearing how the 18-135 is being gutted in the forums you would think there would be no comparison, whether the same magnification or not.

The aperture was f5.6 for both lenses.

fair point. aside from that, I would love to see more samples from MFD, midrange and infinity + crops (center, borders and corners) as well just to be more consistent. if it does show consistency, there is a huge possibility that there are duds out there.

GeneV · 04-21-2011, 02:51 PM

It seems to me that many copies of this lens are capable of taking good photos at the long end.

However, for real tests, we really need subjects that are either on a single flat plane, or all so distant that it doesn't matter. Otherwise, we tend to get blurry objects in the corners which can be taken or mistaken for bad lens performance.

Perhaps one needs to photograph something sparkling to test this, but I have not seen another example with elongated, astigmatic highlights like the mountain scene in the PZ example.

deadwolfbones · 04-21-2011, 02:56 PM

I'll try to do a brick wall test on my copy soon. :P

GeneV · 04-21-2011, 06:28 PM

Originally posted by Pentaxor

fair point. aside from that, I would love to see more samples from MFD, midrange and infinity + crops (center, borders and corners) as well just to be more consistent. if it does show consistency, there is a huge possibility that there are duds out there.

Yeah, it's not like every copy of the body released with this lens was a work of perfection.

ogl · 04-21-2011, 08:28 PM

I've played with another RAW files from K-r - situation is better at f11 in telerange.
The huge difference in results (but wide-angle is good in any cases) means POOR QC and assembly defects.

It's a pity.

04-21-2011, 11:46 AM	#141
hyyz Site Supporter Join Date: Jan 2008 Location: Michigan Posts: 478	I noticed in general, a zoom lens (except ultra-wide) is weaker at long end than at wide end. I don't know why, but this seems to be a general fact. Actually, many criticisms to zoom lenses go to long ends. And if we take 2 lens with comparable grade but in different range, e.g. 16-50 vs 50-135, we probably will notice 16-50 @ 50mm is weaker than 50-135 @ 50mm. And 50-135 @ 135mm will be weaker than 60-250 @135mm. Above may not be true for all such comparisons, but it seems to be quite common. This behavior is probably designed on purpose. IMHO, it is not quite fair to compare 18-135 @ 135mm (which would be its weakest link) to 50-200 @ 135mm (which is probably not its weakest link). Last edited by hyyz; 04-21-2011 at 12:33 PM.

04-21-2011, 02:26 PM	#143
audiobomber Veteran Member Join Date: May 2008 Location: Sudbury, Ontario Photos: Albums Posts: 6,806	I just did a quick and dirty test vs my 55-300mm. Four shots, I used the best two. No tripod but speeds were high in full daylight (1/750s). The target was the base of the bird bath; 135mm focal length @ f5.6, the 18-135 was wide open, the 55-300 was stopped down (would normally show f4.5 at this focal length). Click on the magnifying glass for a full resolution view. https://picasaweb.google.com/bonhommed/18135Vs55300# I've now tested the 18-135 at both ends of its range and wide open vs the 18-55 II, 18-250 and 55-300. It's noticeably better than the 18-55 and 18-250. I haven't done much to compare vs the 55-300 but from this result it seems to hold its own. So far the 18-135mm is all I could hope for, I guess I'm just one lucky SOB. Last edited by audiobomber; 04-21-2011 at 02:37 PM.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
I don't understand photozone.de	justtakingpics	Pentax SLR Lens Discussion	15	04-16-2011 12:20 PM
K5 and photozone	bluekorn	Troubleshooting and Beginner Help	5	02-02-2011 02:15 AM
DA*55 at photozone.de.	ogl	Pentax SLR Lens Discussion	6	11-24-2009 09:51 AM
DA 15mm at photozone	Andi Lo	Pentax News and Rumors	33	10-23-2009 02:22 AM
DA 15mm Photozone Review!!	K206	Pentax SLR Lens Discussion	2	10-19-2009 08:17 AM

04-21-2011, 10:58 AM	#136
frogfoot Forum Member Join Date: Apr 2011 Location: Moscow Posts: 70	Originally posted by Parallax Was SR off? Oops.. Just checked Photome.. I forgot to turn it off .. Will redo the test.

04-21-2011, 11:01 AM	#137
Menno New Member Join Date: Sep 2006 Location: Netherlands Photos: Gallery Posts: 10	Originally posted by frogfoot This is what I call unusable. Comparison between 18-135 and 50-200: 100% crop of right lower corner. BTW at wide to middle range my 18-135 is noticeably sharper and more contrast than 16-45 (I'm talking about center. Corners are still weaker). So I don't know if my lens can be called faulty. P.S. After more careful examination I can say that 18-135 is not worse than 50-200 in the range of 50-100 mm. Well that's exactly the type of image I get too, see my link above. Wide open at 18mm I can get the same soft corners but then it depends on the scene (field curvature?) It's not motion blur. I see this from about 80mm and up on the corners and above 100mm also the center gets less sharp than on the DA50-200 and quite some CA appears. But from 50 to 60mm my DA18-135 is much better than my DA50-200. From 18 to 70mm it mostly beats my 18-55/50-200 combo when stopped down so it's probably a keeper. For the good teleshots I will take the DA50-200 (or I'll buy the DA55-300) with me but I will have to switch much less often. Most reviews mention this behavior so my guess it's the design of the lens, or there are more bad than good copies..

04-21-2011, 11:10 AM	#138
Pentaxor Veteran Member Join Date: May 2009 Location: Vancouver, B.C. Photos: Gallery Posts: 6,513	I think the test should be redone. when using a tripod test, make sure to turn SR off or use a 2-sec timer for the shot. my initial assessment for both shots however is that both shots are underwhelming, much so with the 18-135. if these were corner crops, I wouldn't like it at all. I'm not even sure if these were decentering issues since the images from 18-100 are what you say are acceptably very good to good. I would say that anything past 100mm is the lens' main weakness, thereby saying that 35mm went to waste considering the results here were in fact clearly ugly. so this lens became more of a 18-100mm lens which is bad.

04-21-2011, 11:18 AM	#139
frogfoot Forum Member Join Date: Apr 2011 Location: Moscow Posts: 70	Originally posted by Menno For the good teleshots I will take the DA50-200 (or I'll buy the DA55-300) with me.. That sounds reasonable. I'm changing my 50-200 to 55-300 too. And I decided to add margin while shooting when it's possible, to crop bad edges during processing. Good center sharpness of 18-135 and 14.5 MP of my K-7 allows to do that.

04-21-2011, 11:25 AM	#140
frogfoot Forum Member Join Date: Apr 2011 Location: Moscow Posts: 70	Originally posted by Pentaxor I think the test should be redone. when using a tripod test, make sure to turn SR off or use a 2-sec timer for the shot. Ah.. I used 2 sec timer with SR on. So I don't need to retest..

04-21-2011, 12:18 PM	#142
Pentaxor Veteran Member Join Date: May 2009 Location: Vancouver, B.C. Photos: Gallery Posts: 6,513	Originally posted by hyyz I noticed in general, a zoom lens (except ultra-wide) is weaker at long end than at wide end. I don't know why, but this seems to be a general fact. Actually, many critisms to zoom lenses go to long ends. And if we take 2 lens with comparable grade but in different range, e.g. 16-50 vs 50-135, we probably will notice 16-50 @ 50mm is weaker than 50-135 @ 50mm. And 50-135 @ 135mm will be weaker than 60-250 @135mm. Above may not be true for all such comparisons, but it seems to be quite common. This behavior is probably designed on purpose. IMHO, it is not quite fair to compare 18-135 @ 135mm (which would be its weakest link) to 50-200 @ 135mm (which is probably not its weakest link). this is true with regards to longer end IQ and I also believe it is not fair to make such comparisons. although one obvious criticism about the lens is not because how it performed against another lens but rather how poorly it performed at the longer end at it's own. it is not what you even call as acceptable nor mediocre but plain poor performance which makes a person hesitate on spending $500 dollars on such a lens. personally, I would rather have a 2 lens combo for less or same amount with a much better result.

04-21-2011, 02:39 PM	#144
Pentaxor Veteran Member Join Date: May 2009 Location: Vancouver, B.C. Photos: Gallery Posts: 6,513	Originally posted by audiobomber I just did a quick and dirty test vs my 55-300mm. Four shots, I used the best two. No tripod but speeds were high in full daylight (1/750s). The target was the base of the bird bath; 135mm focal length @ f5.6, the 18-135 was wide open, the 55-300 was stopped down (would normally show f4.5 at this focal length). https://picasaweb.google.com/bonhommed/18135Vs55300# I've now tested the 18-135 at both ends of its range and wide open vs the 18-55 II, 18-250 and 55-300. It's noticeably better than the 18-55 and 18-250. I haven't done much to compare vs the 55-300 but from this result it seems to hold its own. So far the 18-135mm is all I could hope for, I guess I'm just one lucky SOB. I believe it would be better if you did the same magnification rather than to rely on the 55-300's 135mm marking. I did a similar 135mm test with other lenses before and since the 55-300 seems to show more magnification than the other 135mm's, I adjusted it to atleast show the same degree of magnification and show a much precise comparison. also, I would try and compare at same aperture openings like f5.6 and f8.

04-21-2011, 02:39 PM	#145
audiobomber Veteran Member Join Date: May 2008 Location: Sudbury, Ontario Photos: Albums Posts: 6,806	Originally posted by Pentaxor I believe it would be better if you did the same magnification rather than to rely on the 55-300's 135mm marking. I did a similar 135mm test with other lenses before and since the 55-300 seems to show more magnification than the other 135mm's, I adjusted it to atleast show the same degree of magnification and show a much precise comparison. Point taken, but hearing how the 18-135 is being gutted in the forums you would think there would be no comparison, whether the same magnification or not. Originally posted by Pentaxor I would try and compare at same aperture openings like f5.6 and f8. The aperture was f5.6 for both lenses.

04-21-2011, 02:46 PM	#146
Pentaxor Veteran Member Join Date: May 2009 Location: Vancouver, B.C. Photos: Gallery Posts: 6,513	Originally posted by audiobomber Point taken, but hearing how the 18-135 is being gutted in the forums you would think there would be no comparison, whether the same magnification or not. The aperture was f5.6 for both lenses. fair point. aside from that, I would love to see more samples from MFD, midrange and infinity + crops (center, borders and corners) as well just to be more consistent. if it does show consistency, there is a huge possibility that there are duds out there.

04-21-2011, 02:51 PM	#147
GeneV Veteran Member Join Date: Oct 2008 Location: Albuquerque NM Photos: Albums Posts: 9,830	It seems to me that many copies of this lens are capable of taking good photos at the long end. However, for real tests, we really need subjects that are either on a single flat plane, or all so distant that it doesn't matter. Otherwise, we tend to get blurry objects in the corners which can be taken or mistaken for bad lens performance. Perhaps one needs to photograph something sparkling to test this, but I have not seen another example with elongated, astigmatic highlights like the mountain scene in the PZ example.

04-21-2011, 02:56 PM	#148
deadwolfbones Veteran Member Join Date: Sep 2007 Location: La Crescenta, CA Posts: 7,450	I'll try to do a brick wall test on my copy soon. :P

04-21-2011, 06:28 PM	#149
GeneV Veteran Member Join Date: Oct 2008 Location: Albuquerque NM Photos: Albums Posts: 9,830	Originally posted by Pentaxor fair point. aside from that, I would love to see more samples from MFD, midrange and infinity + crops (center, borders and corners) as well just to be more consistent. if it does show consistency, there is a huge possibility that there are duds out there. Yeah, it's not like every copy of the body released with this lens was a work of perfection.

04-21-2011, 08:28 PM	#150
ogl Banned Join Date: Dec 2007 Location: Sankt Peterburg Photos: Gallery \| Albums Posts: 8,382	I've played with another RAW files from K-r - situation is better at f11 in telerange. The huge difference in results (but wide-angle is good in any cases) means POOR QC and assembly defects. It's a pity.