Aha, found them. The clue was that noise reduction is only enabled at 1/4 of a second and slower. I just looked at a TIFF shot at 1/20th of a second, and there are a number of hot pixels that match what my prior images had uncovered. The hot pixels disappear at 1/4 of a second, indicating noise reduction is enabled, and begin to reappear at 2 seconds. ============================== A few months ago I was shooting photographs at San Francisco's Asian Art Museum, among them an image of a small ivory Boddhidharma (the First Patriarch of Zen, who brought Buddhism to China) figurine with a black background. The image capture setting was highest quality JPEG, and using flash was a no-no in the museum, shutter speeds were approx 1/15th of a second (note noise reduction was on). A bit over a week ago, I was looking at the Bodhidharma image and noticed white flecks where there should have been only black. Zooming in uncovered the unmistakeable pattern of a single-pixel white dot surrounded by black, which had consequently undergone JPEG pixel-matrix compression.
I cleaned the sensor with a brush, and shot images (again JPEG highest quality) last weekend in Yosemite at 1/90th of a second and faster shutter speeds. Examining them afterwards, I again noticed luminance shifts in the 'white' direction in the same pixel locations, but not to completely- or near-white values. "Oh, oh", I thought, "a sensor with bad pixels", although my prior testing with StarZen's 'Dead Pixel Test' soon after obtaining the camera had indicated no problems. So I reran lenscap-on files through StarZen's utility (thanks Mark!) earlier today, with noise reduction still turned on in the camera. My first tests on JPEGs, with the default utility settings for bad (60) and dead (250) pixel luminance values, suggested there were bad (but no dead) pixels, albeit at locations other than the ones I'd noticed before, and appearing at shutter speeds even faster than 1 second. But when I reran the tests on TIFFs, no bad pixels appeared until a 2-second shutter speed setting, and even then they were marginal. So does anyone have an idea what might be going on? Perhaps I still have particles clinging to the sensor, but wouldn't that cause stuck-at-black pixels, versus white ones? I've swapped lenses several times between the Asian Art Museum and Yosemite tests, so I don't think particles on the lens elements are to blame. I'm stumped. Maybe I should just shoot RAW (and break down and upgrade to Photoshop CS)? Thanks in advance! ============================== Brian Dipert Technical Editor: Mass Storage, Memory, Multimedia, PC Core Logic and Peripherals, and Programmable Logic EDN Magazine: http://www.edn.com 5000 V Street Sacramento, CA 95817 (916) 454-5242 (voice), (617) 558-4470 (fax) mailto:[EMAIL PROTECTED] Visit me at http://www.bdipert.com
