Photography
Related: About this forumSnowshoe at Mt. Rainier, end of April 2022
When you're taking images like these -- multiple shots for a panorama, no tripod, rotating the camera as best you can around the exit pupil, bright, bright, so you have sunglasses on and can't see the live histogram -- stuff happens.
This is a new camera for me, and for some reason the autoexposure under exposed by a 1.5 stops!!! Obviously I have to figure out what happened.
But still, with an APSC sensor, some aggressive post processing, it turned out to be a usable panorama. Didn't have to worry about any blown out pixels
---
We are standing on Panorama Point (about 2 miles north, and 1500 feet higher elevation than the Paradise visitor center)
Front and center is the Tatoosh Range.
There are 4 volcanoes visible (barely) in this image. Mt. Adams, Mt Hood (if you know exactly where to look), Mt. St. Helens, and of course we are standing on Mt. Rainier.
For a full view of the image, click on it -- it's 7000 pixels wide.
MichaelSoE
(1,576 posts)Unless you have a "scene" mode for Snow, sand, etc.; anything brighter than 18% gray, the camera sees a whole lotta brightness and defaults the expose to automatically adjust to an average exposure.
Likewise, if you are shooting something very very dark it will automatically go to the average and over expose the image.
Back in the days before everything was "auto" one had to take an exposure reading and then manually compensate to get a proper exposure.
Also you did a great job shooting a pano without a tripod or pano-head. Did the camera do the stitching or did you use stand alone photo software?
Pobeka
(4,999 posts)These were shot in "A" mode, with the evaluative metering (looked at the whole scene). I dial in a exposure bias of -1/3 to avoid blown out pixels.
Apparently this camera (Canon G1XM3), really cranks down the exposure for this type of scene (as you say).
Normally I can see the live histogram and adjust the exposure bias with an expose to the right strategy.
I think I mainly need to adjust my habit to check the histogram in the EVF before shooting the pano, set the exposure bias, and then shoot the pano.
99% of all my shots are outdoors, and in "A" mode. I don't use any of the fancy scene settings because I want raw images for post processing.
When I'm shooting panoramas, I'll stand still, hold the camera in front of me and attempt as best I can to rotate around where I believe the exit pupil to be. Gets a little difficult at the extremes because the LCD display can be difficult to see. I try to get a 33% to 50% overlap between images.
For post processing, I start with Darktable, I only do TCA correction, make sure there is exposure "headroom" for every image, and export as 16 bit tiffs at 50% scale which become input for Hugin.
It is possible to have Hugin adjust the camera position on every frame, but it's extremely tricky. But it makes a huge improvement in the final stitched image when done right. (I see there is some chatter about a possible feature enhancement for handheld pano's).
Thanks for your comments!
George McGovern
(5,420 posts)Pobeka
(4,999 posts)usonian
(9,680 posts)And open source! I have been an open source software fan forever.
I sympathize over finding your way through a new camera and so many options. I never used scene mode, though perhaps it's the smart way to go.
Keeping my processing incredibly simple at this point, just Preview on the mac and GIMP when the going gets tough.
Darktable and hugin sound great.
Happy photographing.
One problem I've had with scene modes is they don't produce a raw image. Often is is because those modes are actually taking multiple images and combining them in smart ways to produce a final image.
If I can see the live exposure histogram when taking the shot, I can manipulate the exposure bias to prevent blown out highlights, and it's all good.
Raw images save the day in sooo many ways. Typically with the extra bits of depth you can pull out reasonable colors from shadows, and wonderfully, what is a blown out pixel in a jpeg is quite often not blown out in the equivalent RAW, so you can recover the highlights with an editor that handles raw.
I was a heavy GIMP user before I found darktable. You can do a a lot with it, but it's an incredible memory hog if you have a lot of layers and need to save the project to edit later. Darktable does what is called nondestructive processing -- in that it doesn't change the original image data, it only save the edits you have made as a list of commands with parameters (if you will). And parametric editing, oooh, it is so powerful... Imagine layer masks in GIMP, but just a slider, and you could apply it to *any* operation with a couple of clicks.
All right, I'll get off my unsolicited endorsement of darktable
Happy photography to you too!