New 1.4x Lens Test Results Online

[lothman]

The problem here is that the host for the image, www.lolux.de, does not work with HTTPS (secure HTTP). There is some problem with its certificate.

is there a possibility to upload such testpics bigger than 1024 pixel to the forum?

[rjlittlefield]

is there a possibility to upload such testpics bigger than 1024 pixel to the forum?

No, the forum upload software has a strict limit of 1024 pixels.

--Rik

[physicsmajor]

Good looking lens mount Lothar. Custom machined I assume?

From memory both the Nikon 7 element 45mm lenses I have, one white dot, one pink dot, are best dot towards the camera sensor at magnifications, that is my memory is correct

...

Thanks Robert, yes custom machined but as a base I use a metal 42 body cap. Drilled in order to center the plastic bushing which then is glued on the M42 body cap.

So now I changed the dot facing towards the camera, what made a big difference but on full frame the extreme corners never get sharp. So for APS-C an excellent lens, but probably I will sell it.

I'd just like to thank lothman for asking this question, because my 7-element Scanner-Nikkor ED lens was not performing as I expected on APS-C and after checking... yup I had it backwards (dot was facing subject). Flipped it around with dot facing sensor and it makes a world of difference at 1.4x on APS-C.

[RobertOToole]

I'd just like to thank lothman for asking this question, because my 7-element Scanner-Nikkor ED lens was not performing as I expected on APS-C and after checking... yup I had it backwards (dot was facing subject). Flipped it around with dot facing sensor and it makes a world of difference at 1.4x on APS-C.

Glad you figured it out.

To be on the safe side I usually double check, even if I think I know the right direction, by trying the other way, just in case. This is especially true with tube lenses, sometimes they work better in the opposite direction you were sure would be best

All the best,

Robert

[lothman]

Glad you figured it out.

To be on the safe side I usually double check, even if I think I know the right direction, by trying the other way, just in case. This is especially true with tube lenses, sometimes they work better in the opposite direction you were sure would be best

All the best,

Robert

just a hint, I checked the pictures at your wonderful comparison and there the dot is orientated towards the objekct (compared to the other lenses around), that brought me on the wrong track. You might add some words about lens orientation on your webpage to make it safer for other users .

Perhaps like this:

[Olaf G]

Good looking lens mount Lothar. Custom machined I assume?

From memory both the Nikon 7 element 45mm lenses I have, one white dot, one pink dot, are best dot towards the camera sensor at magnifications, that is my memory is correct

...

Thanks Robert, yes custom machined but as a base I use a metal 42 body cap. Drilled in order to center the plastic bushing which then is glued on the M42 body cap.

So now I changed the dot facing towards the camera, what made a big difference but on full frame the extreme corners never get sharp. So for APS-C an excellent lens, but probably I will sell it.

I'd just like to thank lothman for asking this question, because my 7-element Scanner-Nikkor ED lens was not performing as I expected on APS-C and after checking... yup I had it backwards (dot was facing subject). Flipped it around with dot facing sensor and it makes a world of difference at 1.4x on APS-C.

This is also the case with the 14 element Scanner-Nikkor lens. The "normal" direction is with the dot towards the sensor. This is how they are mounted in the scanner.

[lothman]

just put of curiosity I tried 16-Shot_Pixelshift on the Sony A7riv, what shows that the laowa 100mm @1,5x and @2.8 still has resolving power
that is the whole frame 19008x12672 pixel downsized

one of the small squares is about 1,2x1,2mm and this is 100%crop of the Pixelshift, a little bit edited to my taste.

non pixel shift:

[ray_parkhurst]

Strangely, it seems the 16-shot composite has introduced more false color, where there was less in the single image.

[lothman]

Strangely, it seems the 16-shot composite has introduced more false color, where there was less in the single image.

the lightning were the same led lamps but probably at an different angle on the second one and the magnification on the second one was 1,5 instead of 1,4. But I think this lens does still outresolve a Sony A7r4 without pixelshift

[rjlittlefield]

Strangely, it seems the 16-shot composite has introduced more false color, where there was less in the single image.

I suspect that it has revealed more "real" color. These colors look to me like interference colors formed by small features on the subject, as opposed to false colors introduced by the Bayer filter. When the colored areas are small, the colors are less obvious when averaged together into pixels that cover more area on subject. The colors may have changed because of illumination, but that's not necessary to explain the results.

The situation in this test is different from increasing the magnification by using a higher mag objective or moving enough closer to double the magnification. In those cases the NA at subject increases, which tends to reduce false colors by reducing coherency. But in this case the NA is almost identical, so the colors in the optical image can be almost identical also.

--Rik

[lothman]

this taken from one of the pixelshift stack.

same upsampled

[ray_parkhurst]

Strangely, it seems the 16-shot composite has introduced more false color, where there was less in the single image.

I suspect that it has revealed more "real" color. These colors look to me like interference colors formed by small features on the subject, as opposed to false colors introduced by the Bayer filter.

Yes, may be the case. I think the telltale features are the longer ones, and they are showing a consistent color "fringe" along their lengths.

Lothar, I assume you shot the wafer with the scribe lines lined-up with the sensor, correct?

[lothman]

Lothar, I assume you shot the wafer with the scribe lines lined-up with the sensor, correct?

there is a slight tilt you can see in the overview

[Andy Davies]

Is the 5400 better for full-frame in terms of coverage?

[RobertOToole]

Is the 5400 better for full-frame in terms of coverage?

Yes. The Scanner-Nikkor 7 element, 45mm lens is only meant to cover the height, or short side, of the 35mm slide, not the long side. The Minolta seems to be over-engineered so even though its slower than the 7 element Scanner-Nikkor it seems to have a bigger image circle. The Minolta is a really special lens. When I tested it against the MP-E, the Min corners were even better, and the MP-E is very flat with a broad image circle, so much so I would say that is one of the MP-Es strengths.

Best,