Why only 44 seconds instead of reported totality of 2 minutes and 35 seconds? Clouds interfered with and distorted the capture, eventually totally obscuring the corona after ~43 seconds. My entire recording was 2 minutes and 26 seconds, but the viewfinder showed only darkness from about 44 seconds till the end. The video shows real-time speed (~44 seconds), then half speed.
Eclipse Day Decisions Made For Better or Worse
I couldn't order the weather. It was fickle for a good amount of Texas, including my viewing site, Austin. Weather forecasts and updates seemed to constantly change. Inclinations to travel were tempered by fears of encountering vehicular multiple-dimpling and hail-caused car window smashings. Also, destinations required real time to reach; available time required for traveling eventually evaporated. We stayed put.
From hearing stories about past eclipse events. area population swelling and travel would be challenging for residents and visitors alike. Highways and smaller roads could result in no-gos from traffic overflow or collisions or combinations. News frequently broadcast alerts about having certified solar glasses, stocking up on food, ensuring full gas tanks, and preparing for hours-long departures afterwards. We took the advisories seriously.
I should have thought of using a tripod for my Canon IXUS 180 (PowerShot) camera. As much as I tried to hold it steady, it shook. After I truncated the recording from 2 minutes and 26 seconds, I manually aligned the frames using GIMP. The process was similar to how I aligned the October 2023 annular eclipse frames.
Article:"Partial Annular Solar Eclipse Oct 14 2023, Austin"
YT video ~35 sec): Side by Side Eclipse Views, Austin Area Oct 14 2023"
Another should-have-done was aiming the camera downward a few seconds before totality, turning it on and starting to record, then aiming the camera upward. I would have captured the timer beeper for start of totality. (I did manage to capture the timer's end-of-totality beep.)
GIMP and OpenShot Image Processing
In short, I used GIMP, with grid display to align the images in layer views. The task was much more difficult than October 2023 annular eclipse because of moving-clouds distortions and obscuring. Several times, the view transitioned from disk to arc to smudge, in different directions and varying focus qualities. Clouds! Consolation is that some of the corona was viewable, and we got no rain.
As for OpenShot, I had help with converting the clip to individual frames, then re-assemblling the edited frames into a view-stable disk-centric movie. Turns out that trying to use OpenShot to assemble the frames froze the tool. We used ffmpeg in a command window.
Extra Touches
Some features I hadn't tried ever, or in a long time:
- I used some OpenShot's capabilitiese to crop, scale down, and move an inset movie of a kitchen timer.
- I extracted about 50 seconds of audio in OpenShot, then opened and edited it in Audacity. I added it to the video, along with the image of the graph.
- I created an outline circle to note the major visible prominence at 4 o'clock position.
If I Knew Then, ...
I discovered many of my actions generated a lot of heat before light. Glad to have learned shorter steps eventually. A thought kept me going for creating the video. I'd never think about processing another eclipse set of images. Not inclined to go to another eclipse site any time soon.
"Can’t get enough of the total solar eclipse or got clouded out? Here are the next ones to watch for" provides some years to consider next US eclipses.
The next U.S. taste of totality comes in 2033 when an eclipse brushes Alaska and Russia. And in 2044, one will cross Greenland and western Canada, touching swaths of North Dakota and Montana. ... An eclipse on the scale of Monday’s event won’t happen again until Aug. 12, 2045."Future Eclipses" provides some info in tables for future solar and lunar eclipses. Long haul to get there.
