Urgent: Need nodal points for Zeiss 55mm f1.4 Otus and Zeiss 135 f2 APO please

[Ligament]

Howdy all,

I'm in a time crunch and hoping one of you can share the nodal points for these two fine lenses. I'll be shooting them on a D800e and have the RRS pano head. I don't have enough time to find the optimal nodal point myself. Help!


 

[Bart_van_der_Wolf]

Howdy all,

I'm in a time crunch and hoping one of you can share the nodal points for these two fine lenses. I'll be shooting them on a D800e and have the RRS pano head. I don't have enough time to find the optimal nodal point myself. Help!


 

Hi,

Well, although it shouldn't take that long to determine (just look for relative foreground versus background shifts when you rotate), a quick and dirty method would be to look for the aperture blade position looking in from the front of the lens at a few angles. The entrance pupil will be situated approximately at that position, then shift that position over your axis of rotation.

It won't be perfect, but it's better than nothing ..., and is a good starting point for later tweaking.

Use a good stitching program (PTGUI, Hugin) that can handle image shifts (often called the 'd' and 'e' parameters, from the Panotools heritage). They will be able to compensate a bit for user errors. Also use liberal overlaps, 30-50%, to increase your options for blending and masking.

Cheers,
Bart

[NancyP]

Look on the Zeiss website for the lens design, the information ought to be there. But the quick and dirty way is as described, and the 5 minute and slightly less accurate way is to position your camera/lens/nodal slide/tripod so you can sight a vertical window frame edge a meter away, and a vertical window frame across the street, or even better, several streets distant. Shift your camera from side to side and move up and down the nodal slide until you find the nodal slide point where there is no change in relative position of your own window frame and the other building's window frame.

As you might guess, I live in a high-rise and see other high rise buildings the next block over.

[alan_b]

Here's a tutorial: http://dgrin.smugmug.com/gallery/2114189

It's pretty easy - even more so with live view, as you can see any parallax effect in real time as you pan.

[Ligament]

When Looking at the Zeiss lens spec sheets for these lenses:

http://www.zeiss.com/content/dam/Photography/new/pdf/en/downloadcenter/datasheets_otus/otus_1455.pdf
http://www.zeissimages.com/mtf/z/aposonnar_2_135.pdf

I see the Entrance pupil position ( in front of image plane) listed for Otus at 101mm, and the 135 f2 APO at 34.5mm

Are these number sufficient for setting my panoramic nodal slider distance, or does one still need to do it manually by eye and iterations? Is the 101mm and 34.5mm suggested starting points, or are they exact, best distances to set?

Sorry for the naive questions. If we could just use the published entrance pupils positions life would be a lot simpler...

[elf]

Bart's answer is the best for determining the entrance pupil.  It usually works well enough just estimating the location, but you can measure it fairly accurately by using a macro lens focused on the iris. 

Put the subject lens on a rail, I have a piece of aluminum channel, but any straight edge that can stabilize the lens will work. Mark the position of the flange. Focus the macro lens on the iris.  Replace the subject lens with a flat plane and move it until it's in sharp focus. Measure the distance from the plane to the marked position of the flange.  Add the flange focal distance to this to get the entrance pupil location relative to the film plane.

One small problem with entrance pupils is they rarely stay in the same place when you're changing focus or zoom.  I have a 70-300 zoom where the entrance pupil changes from an inch behind the front lens to 18 inches behind the camera.

I've found the technique on the site posted by Alan_b doesn't work well when the subject focus distance is different than the test focus distance (This really depends on the lens design, so you may have better luck).

[Bart_van_der_Wolf]

I see the Entrance pupil position ( in front of image plane) listed for Otus at 101mm, and the 135 f2 APO at 34.5mm

Are these number sufficient for setting my panoramic nodal slider distance, or does one still need to do it manually by eye and iterations? Is the 101mm and 34.5mm suggested starting points, or are they exact, best distances to set?

Hi,

They are probably very accurate, if you can detect the image plane position accurate enough. A slight complication though, as poster 'elf' mentioned, the position of the entrance pupil may shift with focus. So it is always wise to do a verification of the initial setting, based on an actual shooting scenario.

Since most parallax occurs at close distances, I tend to calibrate my No-Parallax-Point (NPP) settings with the lens focused at something like 3 metres for landscape panos. I then place something (sometimes hang a wire with a weight from a lamp) much closer, and align it with something in the distance (a building or a lamp post somewhere outside my window), and use a very narrow aperture to get enough DOF to see features in both fore- and background. Then 2 shots (one rotated to the extreme left and another to the extreme right, just overlapping a few percent) are enough, although (if the lens has considerable distortion) also one in the middle may be needed to verify.

My TS-E 24mm II required NPP calibration with 0.5 mm accuracy in order to get parallax free results. Half a millimetre more or less produced a 1 pixel parallax in opposite directions. Most lenses require not more than a 1 mm NPP position accuracy.

Fish-eye lenses can shift their entrance pupil a lot with focus distance, luckily they can also be used with just a 'fixed' hyperfocal distance setting and cover from foreground to infinity, thus requiring just a single NPP to make a note of.
 
Cheers,
Bart

[fdisilvestro]

Fish-eye lenses can shift their entrance pupil a lot with focus distance, luckily they can also be used with just a 'fixed' hyperfocal distance setting and cover from foreground to infinity, thus requiring just a single NPP to make a note of.
 
Cheers,
Bart

I understand that fish-eye lenses shift their entrance pupil depending on the angle of incidence of light, making it tricky to use for stitching. In theory you can find a "least-paralax-point" and shoot two opposite hemispherical photographs to get a 2-shot panorama