Canon RF 7-14mm F2.8-3.5L FISHEYE STM

[Václav Klimeš]

Hello, when do you think the Canon RF 7-14mm F2.8-3.5L FISHEYE STM lens will be supported? Or just select an older version and everything will be fine. I just stitched an anorama from it and it is not stitched well. Now I don't know if it's because the lenses are slightly different, or if I'm doing something wrong. So far I've stitched panoramas from a 24mm lens, I don't have any problems there. But as I say, I don't have any experience with fisheye yet. I'm still at the stage where I'm stitching from 4 frames. So the overlap is sufficient, but I calibrated the nodal point to 1/3 of the image, because it seems to me that the panorama will be stitched +/- there, but at the end, beyond this direction towards the edge of the image, the parallax is already running away. But it occurred to me that I should have shot too far back, so that I would practically be shooting with a tripod, which I would therefore have in front of me. Is it possible that the nodal point is somewhere in front of the lens?

Best regards, Vaclav Klimes

[Erik Krause]

Am 15.04.2026 um 19:37 schrieb Václav Klimeš:

> I just stitched an anorama from it and it is not stitched well.

Did you tell your stitching app that it's a fisheye? Also the crop
circle should be adjusted, if not the whole sensor is covered. What do
you use for stitching?

--
Erik Krause
http://www.erik-krause.de

[Václav Klimeš]

I use PTGui and I think I've figured it out, the nodal point is just in front of the lens and I was afraid to put the camera further back. Yes, I have Fisheye set, when I didn't set it, the panorama couldn't be stitched at all. Tomorrow I'll try to experiment with it, I'll create my own lens profile and I'll let you know how it turned out. Here's a test image, the bad stitching can be seen, for example, on a painter's easel. http://obyvak.klimes.studio

Dne středa 15. dubna 2026 v 21:00:10 UTC+2 uživatel Erik Krause napsal:

[Erik Krause]

Am 15.04.2026 um 21:08 schrieb Václav Klimeš:

> the nodal point is just in front of the lens

For fisheyes the no-parallax-point actually moves, dependent from the
distance from the optical axis. It could well be, that it's just in
front of the lens for areas close to the image circle, especially if the
lens covers more than 180°.

BTW.: You can see the no-parallax-point, which is the apparent position
of the entrance pupil. Close the aperture and place a bright light
behind the lens (instead of the camera) Then look through the front lens
from sideways. You should see the image of the aperture (the entrance
pupil).

However, I think you should head over to the PTGui forum, where more
qualified help is available, and this lens is currently discussed:
https://groups.google.com/g/ptgui/c/gAuQchoiSv8/m/Ht29S9gWBAAJ