Hello, Sent a message to the user list, but think it's prolly more appropriate here. I have spent considerable time testing:
gphoto2 2.5.27 gcc, popt(m), exif, cdk, aa, jpeg, readline libgphoto2 2.5.31 standard camlibs, gcc, no ltdl, EXIF With my Nikon D7500 and Xfc cameras. gphoto2 clearly sees these cameras and --list-all-config and --summary produce significant information. Sadly, there are serious issues. D7500 capturetarget, iso, shutterspeed, shutterspeeds, f-number, and exposurecompensation are all "ReadWrite. I can successfully set these keys with the camera set to M mode and gphoto2 does not produce error output. But only capturetarget, iso and f-number produce expected results. ALL attempts to use shutterspeed (with Choice value or millisecond value), shutterspeed2 (with Choice value or fractional value), and exposurecompensation (using choice value or stop value) produce images with apparently random EXIF shutterspeeds. I noticed this when I set up a shell script to take seven images, that last with a 1/2 second shutter speed. When the sequence got to that point, it was audibly clear that the shutter opened for far longer than a half second. In fact, looking at the image EXIF showed it was 4 seconds. So, I started a rigorous testing process. I created shell scripts (last versions attached to this message). One version looped through 0-35, the range of shutterspeed and shutterspeed2 Choice values from 1/8000s to 30 seconds. I used --trigger-capture, with either a 1 second or .5 second wait between each loop cycle, to get an image set. For more image sets I then modified the script to use millisecond values for shutterspeed, fractional values for shutterspeed2, and finally exposurecompensation values using a fixed shutterspeed of 1/60 as the base from which to determine the compensation. At all times gphoto2 --trigger-capture resulted in a legit image I could list, get, delete, and view on my laptop. There were no errors reported unless I mistyped something in the script. But ... The EXIF Shutter Speed value of the images captured do not match the shutter speed values or the exposure compensation values as used in the script. In fact, on successive runs the image EXIF shutter speed values for some images **changed** from the previous run for the same shutterspeed(2) value (e.g., Choice value 0 should equal 1/8000s = 0.0000125, but on successive runs the image EXIF is 1/8000 on one and 1/4000 on the next). Xfc Sadly gphoto2 and the latest library that supposedly enabled Xfc support (which I built from source) is useless with this camera - capturetarget, iso, shutterspeed, shutterspeed2 are all ReadOnly no matter what mode I set the body to, despite --summary output showing that they are ReadWrite. ExposureCompensation, though --list-all-config shows it to exist and ReadWrite, is reported by gphoto2 as non-existent when used to attempt image capture. I am quite happy to help developers figure out what's going on (whether it's a user error [! certainly possible] or in the library/code) and create image sets using whatever commands you might need. I'm a retired 30-year IT guy, though my wheelhouse was perl/javascript/shell scripts/powershell/product management, so I'm unafraid to do this kind of work (it's what I used to get paid for). I am a Linux user. This testing was performed on a Kubuntu system with the camera USB plugged directly into the Surface Laptop USB port. I lost my testing and writeup from yesterday, but have the images from seven test runs today, 20240321. And I can always make more test image sets. Right now the image test sets with 35+ images are all sequenced to match the shutterspeed values from shortest to longest stopping at 30 seconds. It's easy to then look at the EXIF data and see that the images in that sequence clearly do not move precisely from 1/8000s to 30 second exposures, some exposure values are duplicated, and some do not exist in a set (but do exist in a subsequent set). A few more test sets exist where I was trying to finesse the problem with a 7 image set of exposures for a group of images following NASA Eclipse exposure recommendations. This is when I tried to use exposurecompensation. No joy. All that said, I CANNOT help further until mid-April. I was doing all this work to try and automate image capture for April's eclipse which is now but a few weeks away. I'm pulling an RV down to Missouri and I now have no time to try and get gphoto2 working as it must, and, frankly, I don't quite trust it given my testing results. Regards, Guy Stalnaker jimmyg...@gmail.com -- “Happiness is the meaning and the purpose of life, the whole aim and end of human existence.” ― Aristotle
gphotoD7500ShutterspeedTesting.sh
Description: application/shellscript
eclipse7ExposureF8ZFc.sh
Description: application/shellscript
eclipse7ExposureF8.sh
Description: application/shellscript
_______________________________________________ Gphoto-devel mailing list Gphoto-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/gphoto-devel
