21mm SEM corner performance

[Mark T]

Advertisement (gone after registration)

The OP obviously has discovered two defective lenses. Luckily it was picked up, or there was a danger of sub-standard photographs.

[vcnt]

Hi - I am having exactly the same problem and my lens has gone back to Leica - http://www.l-camera-forum.com/leica-forum/leica-m-lenses/327603-decentred-sem-21-a.html

 

Did you get the issue resolved with Leica in the end or did you give up? Really concerned that this seems to have slipped through Leica QC but hope that they will resolve the issue.

 

The only reason why I did purchase the Leica lens was that I wanted to avoid this sort of problem otherwise will keep my VC21/4 which does have more even corners after managing to find a decent sample although needs to be stopped down to f8 with the SEM being a far superior lens in any other way.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[tobey bilek]

If I compare the height og the number '3' in the four corners there is a differense of more than 10 %, enough to safely say that this was not shot completly square at the wall.

 

I would be a bit more carfull testing before suggesting this lens is bad right out off the box... IMHO

 

Before I read this, I was coming to the same conclusion.

 

Just looking at the center square, it tapers smaller to the right and even more top to bottom.

 

Redo the test with proper geometry. Set your tripod to same height as target area center using the lens. Then you need to MEASURE lens to target edge and make sure left and right are the same.

 

Or drop verticals from the sides to the floor and construct a right triangle from center out.

Some painters tape on the floor will help. Measure with string if you do not have a long tape.

 

The view finder is in the wrong position to be used. It is too high and all but the zoom finder are not offset to the side so they are over the lens.

 

I would not discount the lens mount being off to the center in which case maybe tolerance stack up is showing. Try some other lenses.

Edited April 10, 2014 by tobey bilek

[NJH]

0.7m? Seriously since when has there been a use case for a wide angle lens that requires perfect corner to corner sharpness at minimal focus distances

[jaapv]

Lens testing arrangements at optical factories , including elaborate laser alignment testing gear, cost millions and require trained staff to operate. I am amazed that so much money is wasted when brick walls and pieces of paper taped to the kitchen table do so much better….

[jto555]

I downloaded you sample shot and opened it in Photoshop. I would suggest that you are tilting the camera by a couple of degrees. This will throw your results off.

 

You could set the camera on a tripod with the test charts on the ground and use a split-level to level the camera and teat again. However, if you are getting soft corners in normal use, which made you test the lens in the first place, then just sent the lens back to Leica. As its a new lens it will be covered. The bad news is that the lens will be gone for at lease four weeks, plus delivery time!

 

I have bought 2 new and 2 second-hand lenses since November last, and ALL the lenses needed to go back for adjustment. The last one, a 135mm APO (secondhand) is in for adjustment now but the other lenses were repaired and now deliver spectacular results.

Edited April 11, 2014 by jto555

[mirekti]

Advertisement (gone after registration)

Thanks for all the advices. I sent the lens back few weeks ago. I tried two copies, one always had problems on the left side, the other on the right side, but was generally better. I understand the process itself is not simple, and accept my fault in testing, however, I wasn't able to make left side sharp whatsoever, and I did many various adjustements in the lens position towards the wall. Basically, the results with the first lens were always bad on the left side, and with the second one on the right side. Nhf towards Leica, I just had bad experience with this lens.