I'm sorry but I just read DPR's review of the Lytro camera, and my understanding of the camera from that review is that it's a croc of **it.
Now that someone has partially explained it's inner workings, it's become clear that it doesn't appear to operate as the maker says it does, and it's more of a in camera photoshopped image that requires dedicated software to view the files, to reprocess the images to behave as the maker states.
(even tho DPR make claims to the contrary).
All it appears to do is to capture an immense DOF(where it can) and then apply gaussian blur to the image, and then the image is de blurred depending on where you click on the file.
DPR then go on to say that there are two modes one of which is this 'everyday mode' where there is a refocusability range that extends back from infinity. This mode only works up to a specific focal length range, and once you pass this focal length range you use a creative mode. The cross over point where everyday mode stops is at approx 150mm(equiv FOV).
Considering the small sensor size, this probably equates to about 10mm of actual focal length from the small lens, and this is where some DOF limitation probably starts for the little camera.
If the camera was truly as the maker states and that all light rays can be focused, then there should be no limitation on DOF at all .. ever!
The lens's DOF limitations should never enter into the equation, and as we know, all lenses can be focused from near to far without any issues, other than doing it all at the same time(in a single shot), with no focus bracketing(stacking).
Sharpness should never 'drop off' at any point because the camera is supposed to be focusing everywhere at the same time!
Me thinks that:
1. Lytro were in a rush to bring this device to market probably due to financial pressure given the current economic climate
2. Due to this rushed intro, they didn't have the resources to 'properly develop' the software's features.
3. Due to the imperfectly developed feature set in the software(both camera and support, the camera doesn't in fact do what the maker lead everyone to believe.
4. The camera/lens is still bound by the same laws of physics as every other camera/lens system is, and is nothing more than a modern day box brownie.
5. All your images from this device need to be uploaded to the Lytro site to be viewable!!
6. Because of this need to upload all images to the Lytro site, me thinks that they don't want to release any software(well at least not yet) because of a fear that someone will hack into it to expose it for what it really is .. software manipulation of simple images, rather than something actually new in the way images are captured!
7. (from the DPR review), Lytro has the right to refuse any image uploads, and while it's T&C's are pretty much OK, and they don't impose any limitations of rights and claims that a photographer would baulk at, this right to refuse a photographer to upload any images means one thing .. there is a chance that if you capture an image(and it may be a controversial topic) there is a chance that the image may never be displayed, as the images are at the mercy of Lytro!!
Now I've never shot an image that I thought was controversial in any way and so I'd never be exposed to that last point of concern, but do we know what image types Lytro will refuse to allow?
Nudes? Political unrest? .. inner working parts of the camera itself? ... who knows!
Until a proper analysis of the inner workings, it's seems to be an expensive little box of nothing more than a cheapie P&S from yesteryear, with some sort of image processing ability built in!
About 10 years ago, I received a (joke) pressie that was a small key ring camera .. my first ever digital camera. It cost about $20. It took images of immense DOF, but had no internal image processing, and had a 600K pixel count.
This seems to be the development of this little key ring camera!
Most writeups of the camera seem to concentrate more about the operation of the camera, rather than the merit of the claimed inner workings of it, and the validity of these inner workings.
If the device is supposed to be able to capture light rays independent of how a lens focuses these light rays, and the lightfield re processing engine is supposed to then turn those light rays into a focused image, why is it bound to the same physical limitations as other camera/lenses?
DPR weren't full of praise for the device, they seemed to be 'duped' by the product design and interface .. my interpretation of their review is that they feel that because it's different to any other camera, it must be a good thing(they've listed these qualities in the 'pros' section).
(DPR's final word).As it is, it feels like a product arriving before the underlying technology is really ready.
Considering DPR's commercial nature, I don't think they've been as hard on it as they could have been. But as they'll inevitably end up with this device on their advertising program, this seems to make sense .. there's limited scope for commercial internet success if bridges are burnt too early, isn't there!
8. Adobe will catch on to the mass appeal of 'living' images, and produce an action or software version of their own to do the same thing as this focus anywhere-anytime too soon.