Like Tree19Likes

The end of focus?

Closed Thread
Page 7 of 7 FirstFirst ... 4 5 6 7
  1. #61

    Join Date
    Nov 2010
    Posts
    3,377

    Why doesn't this then happen already? I assumed tgat jpeg was a compromise on compression to quality compared to raw files.


  2. #62

    Join Date
    Sep 2008
    Posts
    3,677
    Quote Originally Posted by luckycat:
    Why doesn't this then happen already? I assumed tgat jpeg was a compromise on compression to quality compared to raw files.
    It is, and a higher res picture will of course have a larger file size than a lower res one, even the content is exactly the same.

    What DeletedUser is saying that you can compress a series of pictures that have similar content more effectively than a series of pictures that have a different content.
    Currently they use a proprietary format however and claim it's not much larger than other RAW formats.
    Last edited by Raccon; 09-11-2011 at 08:22 PM.

  3. #63

    Join Date
    Sep 2007
    Posts
    4,012
    Quote Originally Posted by vinyljunky:
    So they've finally revealed the camera:
    Lytro Camera | Uncrate
    Lytro

    There's also a demo of the UI & zoom:
    Lytro light field camera zoom and UI walkthrough - YouTube

    The thing looks TINY compared to even a P&S, and they seem to have got the pricing right.
    My thoughts on the Lytro video:

    This camera is supposed to use "ground breaking" technology, however the main observation from the video is someone holding the camera and remarking that "They can feel it move inside etc." - Does that mean a motorised lens is groundbreaking tech?

    Also, the original info said that sophisticated software on a computer was required to construct the images from the complex camera files, however during the video they demonstrate it working on the camera!

    If I'm being a total cynic, I'd say this is a bog standard lens assembly hooked up to an ipod nano (Similar size of screen!) used to continue the illusion/scam.

    If it is an actual working camera I reckon that, due to the low resolution, it is able to take a very fast sequence of photos while going through a sweep of focal lengths.


    Thinking out loud here but if you are capturing this supposed "Light field", would you need a main lens?

  4. #64
    Quote Originally Posted by Stoob:
    My thoughts on the Lytro video:

    This camera is supposed to use "ground breaking" technology, however the main observation from the video is someone holding the camera and remarking that "They can feel it move inside etc." - Does that mean a motorised lens is groundbreaking tech?

    Also, the original info said that sophisticated software on a computer was required to construct the images from the complex camera files, however during the video they demonstrate it working on the camera!

    If I'm being a total cynic, I'd say this is a bog standard lens assembly hooked up to an ipod nano (Similar size of screen!) used to continue the illusion/scam.

    If it is an actual working camera I reckon that, due to the low resolution, it is able to take a very fast sequence of photos while going through a sweep of focal lengths.


    Thinking out loud here but if you are capturing this supposed "Light field", would you need a main lens?

    The sensor will be high resolution - say, 18MP but due to the way the light field information is gathered the output photos (at a particular focus) will always (essential) be lower resolution, say, 4MP (but probably worse than that). So it won't be fast like a low res camera because its not low res sensor. The raw files will be the same size as the 18MP raw file. They cannot compress that to jpeg as it will lose information before they have post processedthe photos.

    It doesn't sweep through focal length, it captures everything in focus at each depth for that focal length in the same exposure.

    On the little screen they can can get away with rubbish pictures.

    You would still need a main lens to set the focal length - the zoom. Focusing at different distances is the bit that you can do after the event.

    Or its like having several low res sensors focus at different depths - but the output is the sum of all the pixels.
    Last edited by zerocred; 10-11-2011 at 12:42 AM.

  5. #65

    Join Date
    Sep 2008
    Posts
    3,677
    Quote Originally Posted by Stoob:
    My thoughts on the Lytro video:

    This camera is supposed to use "ground breaking" technology, however the main observation from the video is someone holding the camera and remarking that "They can feel it move inside etc." - Does that mean a motorised lens is groundbreaking tech?
    You do realize that this is a private video / commentary and the motor is for zooming?

    Also, the original info said that sophisticated software on a computer was required to construct the images from the complex camera files, however during the video they demonstrate it working on the camera!
    Can you provide a source where it states that in-camera processing will not be possible?

    Just because they say you need to have sophisticated software on a PC does not mean it won't / can't be done on the camera - if the light field engine is powerful enough to process the data for storage it shouldn't be too hard to generate a picture for displaying, just like any DSLR can show you a picture of the RAW data it captured (and saved, if that option was selected).
    Needless to say but you also need sophisticated software on your PC to construct a picture from RAW files, so I don't see any contradiction here.

    If I'm being a total cynic, I'd say this is a bog standard lens assembly hooked up to an ipod nano (Similar size of screen!) used to continue the illusion/scam.

    [...]
    The principle is a century old but if it's a scam one has to wonder what took so long. The Germans are in on the scam btw. (Link from the article below)

    Here is a good article that provides some more information and that it's not just about that new camera but also what else the technology could offer: The Lytro Light-Field Camera: How It Works

    If you are into the really juicy details here is the dissertation of Lytro's CEO.

    I might sound like a fanboy but I am actually not interested in the Lytro camera itself, I just find the technology and possibilities quite interesting. The naysayers haven't provided any evidence so far that it doesn't work, instead it's mostly suspicions and opinions. That a new product has certain limitations should not be held against it IMO, just think about what resolution the first camera phones had, or all the "concerns" about the first digital cameras and DSLRs. And today? Some phones rival PnS cameras and film SLR are dead.
    Last edited by Raccon; 10-11-2011 at 01:16 AM.

  6. #66

    Join Date
    Nov 2010
    Posts
    3,377

    No one is a naysayer here Raccon. It is all interesting technology and we all had very interesting conversations about it.

    Can you provide a source where it states that in-camera processing will not be possible?
    The fact that the screen for the Lytro is about (at a guess) an inch wide will severly limit the possibility for in-camera processing. The camera, as it is, is designed for computer processing, not in-camera processing.

    The fact that people are raising questions on the feasibility and effectiveness of the camera as produced does not give you a right to shut down their opinions.

    If Lytro had brought a product to market which replicated a DSLR and engineered this techonolgy for sharpening, rather than the ability to refocus, then I certainly would have been interested in it. As it is, a small postcard sized image and the abilty to refocus on points of it doesn't attract me - especially for US$400.

    P.S. Whoever told you that film SLRs are dead severely misinformed you.

  7. #67

    Join Date
    Sep 2008
    Posts
    3,677
    Quote Originally Posted by luckycat:
    No one is a naysayer here Raccon. It is all interesting technology and we all had very interesting conversations about it.
    I think Stoob qualifies as naysayer (no offense intended) as he thinks it's a scam but hasn't presented any hard facts let alone evidence to support this.

    The fact that the screen for the Lytro is about (at a guess) an inch wide will severly limit the possibility for in-camera processing. The camera, as it is, is designed for computer processing, not in-camera processing.
    I don't understand what the screen size has to do with the processing. Of course a small screen limits the effect the user can see on camera, but that's not any proof that the technology or the product doesn't work.

    The fact that people are raising questions on the feasibility and effectiveness of the camera as produced does not give you a right to shut down their opinions.
    I am not trying to put anyone down but isn't it part of a discussion to counter the other party and rebut their arguments?
    But opinions aren't worth much if they are based on unfounded claims or just "because I don't believe it can work" type of arguments when there are many facts to the contrary.
    The feasibility of light field cameras should be beyond questioning because the theory behind it is a century old and there are enough resources on the net that explain this in detail. As well I have linked to another manufacturer's website.

    Furthermore in this thread we talked about the 3D capability, so if Stoob's opinion would be correct that the Lytro employs trickery not only would the camera need to take a rapid succession of pictures with different focal planes (not focal length) but also offset the pictures multiple times in the X and Y direction for each plane.
    But since the Lytro doesn't have any focus capabilities this seems not workable.

    If Lytro had brought a product to market which replicated a DSLR and engineered this techonolgy for sharpening, rather than the ability to refocus, then I certainly would have been interested in it. As it is, a small postcard sized image and the abilty to refocus on points of it doesn't attract me - especially for US$400.
    Agree, as said it's not my kind of toy either. That being said Lytro has indicated that they may be able to process the data for focus stacking / EDOF (Extended Depth of Field), which could be quite interesting, if they get around the other limitations.
    Though perhaps what Stoob mentioned about taking several pictures and then combine them would be better suited for this (not trying to be sarcastic here).

    P.S. Whoever told you that film SLRs are dead severely misinformed you.
    I read it on the net so it must be true.
    Last edited by Raccon; 10-11-2011 at 11:30 AM.
    luckycat likes this.

  8. #68

    Join Date
    Nov 2010
    Posts
    3,377

    Sorry I thought Stoob was trying to be funny.

    I don't see how the actual technology could be in contest.

    I agree that the camera will be able to process. Why wouldn't it? It already will have a sophisticed sensor - it just needs a bespoke interface. That, as you say, will need a bigger screen. I just don't really see the need for it though. If all this Lytro is designed to do is take a photo which can be refocused, presumably you would want to do that on a computer and then print the photo at the desired focal range. On the camera, for me (although admittedly I am not massively interested in this part of the tech), I would just like to see a preview.


  9. #69

    Join Date
    Sep 2007
    Posts
    4,012
    Quote Originally Posted by Stoob:
    I think that even if it is feasible it would be impossible to achieve without using billions of sensors and super computer processing power!

    I'm reminded of Steorn - Wikipedia, the free encyclopedia !!

    I will be delighted if I'm proved wrong though!
    Quote Originally Posted by Raccon:
    I think Stoob qualifies as naysayer (no offense intended) as he thinks it's a scam but hasn't presented any hard facts let alone evidence to support this.
    Guilty as charged on the "naysayer" accusation but as per previous posts, I 'would' like to be wrong.

    I am not an expert in optical stuff (Hence "focal plane" v "focal length" correction, which I will accept), however I am naturally skeptical of this product.

    If they have a camera, let me test it without laying out USD500 up front, i.e. demo it in a shop etc.!

    Let me download the software and provide actual demo camera files instead of 'playing' with a Flash file.

    Provide specifications for the camera

    Again, What is a "light ray"?, e.g. how many "light rays" are reflected from, say, a pinhead?
    Last edited by Stoob; 10-11-2011 at 01:15 PM.

  10. #70

    Join Date
    Dec 2010
    Posts
    780
    Quote Originally Posted by Stoob:
    Again, What is a "light ray"?, e.g. how many "light rays" are reflected from, say, a pinhead?
    The most concise explanation I could find is on Light measurement.

    Alternatively, Madonna might have the answer you're looking for


Closed Thread
Page 7 of 7 FirstFirst ... 4 5 6 7