Lytro

the physics/technique has been around for quite a while and i remember reading first reports on lytro (or whatever they used to be called back then) two years ago. nice to see this getting to mass market quality. personally I'm not yet convinced to how much use it will be and how much it will contribute.

the change from analog to digital brought us a lot of advantages in photography, but it also brought us the "wheeee" girls taking hundred photos of the same object and wondering why they are still not getting image composition right or exposure controlled. If lytro ends up the same, that it makes things easier for a few good photographers but leaves the entire rest of humanity with even less understanding of the actual process of making a good picture.

The real application for the mass market here is obviously just to not get blurry family photos. Or the possibility to just for once have a non-blurry picture after a party.

I swear to god, the camera has an evil devil inside which makes the pictures blurry after he notices that the guy taking pictures have had just one beer!

i thought it was an imp?

I'm amazed because it actually WORKS!

The sony nex-5 I have does something like this, although I'm not sure of the exact mechanism. Shutter time increases to almost a second in a normally-lit picture with the focusing maxed out. I hardly ever use it because the pictures don't look as nice(more artifacts) and its more difficult to take a picture without blurring from the camera moving, to the point where without a tripod or something stable to lean against it is useless. A picture with everything in-focus just looks weird, too. It does let you go in and add artificial DOF in photoshop or whatever, so you don't have to worry about messing up the focus when you take a shot.

I think this would be very cool if it just did a narrow range around where you set it. I've taken many pictures that looked in focus when I took them but were just a hair off when I saw them on the computer, and subsequently got trashed. If I could just bump the focus a bit one way or the other after taking the pic, that'd be great. That'd actually be a really cool feature for a camera with a traditional sensor - when you hit the trigger it rattles off like 10 shots, incrementing the focus just a hair between shots.

This is a true revolution IMO. Obviously it's going to make things a lot easier for casual photographers, but it's also good for professionals — and indeed, at the price point it's likely to debut at I'd be very surprised if it's anything but pros using it initially. I mean, how useful is it going to be to be capable of touching up focus fuck-ups when filming? Being able to experiment with different focuses on a single photo rather than having to decide prior to shooting it? Often once the moment's gone it's too late to sort these things out unless all you do is shoot macro objects.

I think your camera is doing something different, twiz. There's no shutter lag with Lytra because there's no shutter process — traditional focus/exposure delays are no longer relevant. There shouldn't be artifacts either, and it should perform far better than any traditional camera out there even when lighting is minimal. There're even less parts in the camera because most of the work is done by a single light field sensor and software does the rest. This is one of those rare innovations where there are literally no downsides, only positives.

There's nothing like this on any existing camera AFAIK, it's existed only in theoretical/experimental forms for a long time. It's the future of cameras and will completely obsolete all existing approaches, even though I'm sure every camera manufacturer will very quickly start debuting their own takes on the concept.

Here's a basic run-down of the new type of sensor it includes BTW:

http://www.lytro.com/science_inside

And here's a more objective look at the technology:

http://old.news.yahoo.com/s/digitaltren ... tography_1

Quote from Thrik

I think your camera is doing something different, twiz.

Oh for sure, it has a traditional sensor, so I'm not sure how it does it. Just sayin that something attempting to achieve the same thing is out there, but obviously with its drawbacks. Interested to see this tech develop (and drop in price)!