Spherical and 180° panos wildly different

[mfc]

As others have reported, the spherical pano mode often (always?) gives wildly over-exposed results. I carried out a number of experiments today, and the result was that I found that the other pano modes work as expected.

Further, single frames taken of the same subject in the two pano modes show very similar shutter speeds, both entirely plausible and consistent with the exposures for single-shot photos of the same subject from the same position -- however the results (examples and details below) show that the frame from the spherical pano is seriously overexposed and quite unusable.

What's the best way to report this and have it improved? I want to use spherical panos to ensure I have a record of the drone's position at the start of a flight so I can later confirm the position of single shots, etc. I'm not looking for 'photographic art', simply a record of the environs.

-----
The two photos are attached (both ISO100); the frame from the spherical pano (the lighter one...) reports "1/1026 sec, EV 0.34" and the one from the 180° pano "1/800 sec, EV 0".

 

[mfc]

A little more information on this: it turns out that the exposure used for the frames of the panorama depends on the attitude of the camera before starting the panorama.:

• If the camera is pointing forwards when the panorama sequence is initiated, the entire panorama is badly overexposed (as described above).
• If the camera is pointing straight down when the panorama sequence is initiated, the chosen exposure is acceptable (about 1.3 stops darker), although perhaps somewhat overexposed, and good results can be recovered in post-processing.

In short, it looks as though the setting of shutter speed is being made too soon after the camera is moved from horizontal to vertical, and there is some over-compensation for the change in exposure reading. Surely the results of a photograph or panorama should not be influenced by the previous attitude of the camera?