Sorry for the wall of text:
“Stops” confused me in the beginning too. There’s 3 different ways to increase or decrease a stop, each one does something different, and their broken down into (usually) 1/3 stop increments.
- ISO is light sensitivity. higher number = more sensitive = more grain/noise
- Shutter speed is the time the shutter is open in seconds. shorter time = less motion blur
- Aperture or F-stop - Smaller number = larger aperture in the lens = more depth of field.
ISO should stay the same for any series of bracketed images for HDR photography, so should your aperture. You really don’t have to worry about Depth of Field on the Theta because at ~190° field of view, and a small sensor, DOF is going to be basically non existent.
so that leave shutter speed.
That’s a great article. I remember reading that one a few years back and it really helped me out.
I don’t think the shutter speeds you listed would work for HDRs. In bright sunlight, exposures 5 through 9 would be completely white and over exposed. When shooting Bracketed HDR, a good rule of thumb is to keep a max of 2 stops between each exposure. I’ve found that since the V only shoots JPEG that it’s’ probably better to stick with 1 stop between each shot of a bracketed series.
I did a test with the Theta App set to multi bracket. I’m not sure what has changed since the last time I did this experiment with my Theta S, but the V shoots multi bracket in this app just as fast as the Dual Fisheye plugin. It did crash a couple times in the very short test, and I had to trigger each series of brackets manually, but it did work.
ISO: 64, WB 5200K
1/2500 (base line exposure for the environment)
1/1250 (-1 stop from base exposure)
1/640 (-2 stops from base exposure)
1/320 (-3 stops from base exposure)
1/160 (-4 stops from base exposure)
1/80 (-5 stops from base exposure)
1/40 (-6 stops from base exposure)
1/5000 (+1 stop from base exposure)
1/10000 (+2 stops from base exposure)
1/20000 (+3 stops from base exposure)
1/25000 (+3 1/3 stops from base exposure) toped out at fastest exposure
While this example has more over exposed than under exposed shots bright sunlight will always cap out the Theta V and the whites will clip, it illustrates the more subtle transitions of 1 stop between exposures in the bracket.This scenario gives an EXR with smooth transitions between each exposure. If there’s too much of a jump between exposures, a lot of detail and information is lost. On a DSLR I’d shoot this at f16 or 22, but that’s not an option with the V as the aperture is fixed at f2. The Z1 can stop down to a f5.6 so that should be way better.
Kesper’s plug in HDR2EXR achieves this smooth transition between exposures, gives 11 shots, but also shoots 3 shots for each exposure to reduce the noise in the image. Very brilliant, but adds to the time. Also, for some reason I can’t get it to create an EXR, but that’s a different discussion.
I prefer to stitch in PTGui rather than the in camera stitching. The results are far better and I can fit it when there’s a problem. I set up a template that is based off of yours that you uploaded a while back. It’s specific to my camera and the minute differences that occur in the camera production process. The seam completely vanishes. I love PTGui, I’ve been using it for years. When I did the new tests using the V and the Theta app in multi bracket mode, I noticed shifting in the stitching around the top and bottom of the image. it was very strange because I haven’t seen this before. It’s not there when I stitch them in PTGui from the Dual fisheye plug in. something odd may be happening with the internal stitching.
Yes. I’d love to set have the first exposure of the first bracket set to auto to evaluate the scene, Base the rest of the brackets on +/- 1 or 2 stop increments like the ones I listed above, and then once that series of 9 are done, it repeats the same exposure settings as the first one. Having the settings locked in from the first exposure would give the subtle changes in lighting that occur over time with out the flickering that happens when auto settings are used. This method wouldn’t adapt to a changing lighting scenario like sunset, but that would require something like bulb ramping and there a reason why it’s called the Holy Grail of time-lapse; it’s really hard.
The Dual Fisheye plug in also has a VERY handy feature in that it renames the files to group them by file name and exposure number. this is extremely helpful.
I’m using the time-lapse in an experiment that I’ve been working on. sorry to be vague, but I can show an example when my testing is finished.