... on the horizon. I don't think this has been covered before, but forgive me if it has. One of my magazines had this regarding some technology that I found interesting. I'll hide the company's name for the sake of whatever.
" .... new digital imaging technology. By fitting a camera's image sensor with a special lens and then processing the resulting data with new methods, xxxxx's technology will let photographers fix their photos and exercise new creative control after the shutter is released.
In examples demonstrated by xxxxx, sliders are used to change the point of focus in a photograph from foreground elements to those in the background. These examples also show how a user could bring different areas of an image into focus by clicking on them ....."
I can see it now .... a future great debate as to whether a Mac or a PC gives better bokeh
They've been working on this for a few years now. Theoretically it shouldn't be all that hard, it just takes a massive amount of computational power, which means big and expensive at this point.
And here we've got people arguing already over whether sensors are denser than makes any difference (not in my opinion), so this technology makes me wonder how much of the resolution of the sensor is sacrificed to allow the multi-focus principle. Even 4x6 prints with this sort of after-the-fact choice of focus could be quite impressive, but what would be the long term possibility for the level of IQ that we expect at this point, assuming processing power grows enough to not limit the results? I suspect that we might be up against more than computational limits, but who knows?
Wait, Photon, are you kidding? It's basically taking 3 different pics and you can choose between them, or a combination of all three, not that big a deal. You can currently take a photo at F11 and use a ton of different programs to select the type of focus you'd like with excellent results. Kinda like taking a RAW file but with an extended DOF.
Perhaps I am too dumb, but isn't this just shooting with a small aperture for maximum depth and "just" add selective focus and a shallow field of focus through post processing.
Granted the software must be pretty good to really produce a credible effect, but there is no magic involved. The numbers that explain the working of a 85/1.2 can be used to create the same effect digitally.
The real question is if a "perfect" digital effect looks as attractive as the "imperfect" lens.
Also I guess that shooting with a narrow aperture will demand better lighting relative to a large aperture.
Ruy
PS. you wouldn't even need a special camera, but a lens that can bring maximum dov with critical sharpness at the fastest possible speed.
PPS. as for multi focus sensors, again I may be too stupid, but isn't that overcomplicating matters? As in an expensive solution for something that can be done on the cheap? Even if it can be done that way with the same lens in a single exposure.
Kinda hit the nail on the head rhorta, the way this "special" camera from adobe records the image is WAY complicated and actually takes 1 photo with a ton of different apertures so you can adjust the focus on several different points in post processing. Sorta like a "mega" RAW kinda file. Still you can currently do someting really close to this by selecting a narrow aperture and messing with it.
I don't know the technology, but dang that'd be handy. I think it was mentioning microlenses at the top of the article.
Oh, and no offense to some anyone, but if you have time to shoot at F11, you have time to get the focus on the right subject. How dandy this would be at F .95 on that new leica.
5 years and this will be the new ISO.
At least whatever camera gets this won't have 1D III's AF problems
Adobe demonstrated this a long time ago. It uses multiple lenses to record the image so that effectively the data recorded effectively contains distance information as well as the usual pictorial stuff. Effectively it's a 3D camera. That allows d.o.f. to be set after the shot was taken since the distance of every component is can be calculated, and therefore the appropriate increase in blur can be determined for each pixel. Interestingly it also provides increased depth of field and sharpness due to multiple lenses seeing the same objects at calculable distances. If you have ever viewed good stereograms you will have noticed that the grain pattern seems to dissolve, since it is different in each image, and sharpness seems to be greater than the lens would normally supply. There is probably a low-light downside, since the composite lenses must have smaller apertures than one big one. Maybe their light is still cumulative, I don't know, but it would seem that there must be a downside for the trick of capturing 3D data on smaller lenses and sensor parts.
