h a l f b a k e r yResident parking only.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
| |
What a waste of technology. |
|
| |
Personally, I need the faces of the, you know, the same old subject, the 1000 people needing food assistance in this town, on film but that wouldn't be cool, showing their identity. I was explaining the large number to mayor and council this week. They had no idea that many go through the door, a half block away at the 'food bank'. Many of the donators I talk to have a poor idea of who the 1000 are. The 'food bank' in any city needs multiple exposures on film to capture what is transpiring there, in all the important aspects. |
|
| |
What mensmaximus said (the first time around) (the second bit is wierding me out). |
|
| |
I'm a bit unclear as to what you are going on about, [mens]. You seem to have a chip on your shoulder about anything involving image making but maybe I'm mis-reading your comments. |
|
| |
You need a bit of that "what if?" halfbakery spirit. |
|
| |
Perhaps were you to look at this idea as just the starting point for thinking about multiple captures of the same instant of time (which, admittedly, it's not) that allow for flexibility after the image is taken, then it becomes something far more than a waste or something that was solved 50 years ago. |
|
| |
Plenoptic cameras, for example, were first proposed only about 14 years ago ('91) and are a design that takes many exposures at once and allow for amazing adjustments to the image after the fact, including changes in the depth of field(!), image-based 3D model extraction and, to a lesser extent, exposure. The theoretical counterpart to plenoptics, the polydioptric camera, promises even more post-exposure adjustment, even if the exposures can only be of synthetic imagery. |
|
| |
There is also research going on with ultra-wide exposure latitude camera systems that take two exposures, one set for proper capture of the bright details and the other set for dark detail capture. These two are then combined for a single image with detail across the entire luminance range. |
|
| |
There are also post-production tools appearing that grok these "bracketing" camera image formats and provide real-world problem solving utility. |
|
| |
Finally, there is the RAW format that allows adjustments after the fact that were only a dream five years ago (or handled in crude, all-or-nothing, color timing and push/pull chemical processes). |
|
| |
I don't think any of these things are a waste and they riff on the basic, interesting and valid concept promoted in this idea: automatically bracketed image capture. An inventive and worthy subject. |
|
| |
Why would you even want to see different exposures? |
|
| |
God, [bristolz], I love it when you talk dirty. Don't stop, don't stop......[goes to dust off his pinhole film gear] |
|
| |
//camera systems that take two exposures, one set for proper capture of the bright details and the other set for dark detail capture// - sounds like a simpler way of achieving the same effect. Clever. |
|
| |
Ah ha! - so that's what bracketed means on my new mobile phone. |
|
| |
I now discover that it can take 9 quick-fire low-resolution pictures at different exposures whilst flashing its multi-coloured little light and making a sound like a machine gun. |
|
| |
I'm going to try it down the pub tomorrow evening. I hope it copes with immersion in beer. |
|
| |
//afterwards, you can take some program and tell it, "show me this images at 1/50 of a second" and the camera would add together 20 images// |
|
| |
This won’t be the same as a “real” 1/50 second shot. Just as an example: if you’re photographing a runner, twenty superimposed 1/1000 shots will have a “strobe light” effect. Every thousandth of a second will be in clear focus, so you get 20 different body positions overlapping your photo. In an ordinary 1/50-second photograph, the runner’s arms and legs would be more of a blur, so you could see the action in one still shot.
If you're going to superimpose 20 frames, they must not move in that 1/50th of a second. Otherwise, the moving arms and legs will under-expose and the immobile parts will over-expose.
Now for our runner's biggest hurdle: At 1/50th of a second, an ordinary camera shrinks the aperture. At 1/1000th, it opens up, to gather more light. The depth of field changes dramatically with aperture. If you want that runner to be in focus, but foreground and background out of focus, you’d select a wide aperture. The camera in the idea would not do this.
I kinda like the idea anyway, since it might lead to some neat photographic effects. But it’s no replacement for an ordinary camera. |
|
| |
[Kito] You're right. It's no replacement, but we might get some neat effects out of it. |
|
| |
1) Every time you read out a camera's CCD, there's some underlying noise. A single 1/50 second exposure has one dose of readout noise; 20 1/1000 exposures have 4.5 times (sqrt(20)) as much noise. |
|
| |
2) You don't need 1000 micro-images---that's too much information. You could automatically take 1/50, 1/100, 1/200 ... or some such sequence, and any arbitrary exposure could be added up from there. (1/66 = 1/100 + 1/200; 1/75 = 1/100+1/400 + 1/1000) |
|
| |
3) 1000 images on a 3 megapixel camera = 3 GB of raw data. That's much more than today's camera electronics and storage (not to mention batteries) can handle. |
|
| |
I've wanted to own one of these for a long time. I always struggle with an overexposed sky on my shots. Bun. |
|
| |
Sounds like you're describing a videocamera. |
|
| |