Same MP rating

Discussion in 'Digital Photography' started by George Preddy, Nov 2, 2003.

  1. Bayer (5 pixels of full color information):

    RGGBRGGBRGGBRGGBRGGB




    Foveon (20 pixels of full color information):

    BBBBBBBBBBBBBBBBBBBB
    GGGGGGGGGGGGGGGGGGGG
    RRRRRRRRRRRRRRRRRRRR
     
    George Preddy, Nov 2, 2003
    #1
    1. Advertising

  2. George Preddy

    Ray Fischer Guest

    George Preddy <> wrote:
    >Bayer (5 pixels of full color information):


    Get a life, idiot. Nobody believes your bullshit propaganda anymore
    sicne you obviously don't understand what you write.

    --
    Ray Fischer
     
    Ray Fischer, Nov 2, 2003
    #2
    1. Advertising

  3. "George Preddy" <> wrote in news:Yc1pb.19608
    $:

    > Bayer (5 pixels of full color information):
    >
    > RGGBRGGBRGGBRGGBRGGB
    >
    >
    >
    >
    > Foveon (20 pixels of full color information):
    >
    > BBBBBBBBBBBBBBBBBBBB
    > GGGGGGGGGGGGGGGGGGGG
    > RRRRRRRRRRRRRRRRRRRR


    What are you talking about?
    A Bayer sensor looks like this

    R G R G R G R G R G R G
    G B G B G B G B G B G B
    R G R G R G R G R G R G
    G B G B G B G B G B G B
    R G R G R G R G R G R G
    G B G B G B G B G B G B
    R G R G R G R G R G R G
    G B G B G B G B G B G B

    And a Foveon sensor looks like this ..

    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C
    C C C C C C C C C C C C

    .... where each C detects three "colors" that
    can be mapped to R, G and B.

    Both examples above has a maximum resolution of
    12x8 pixels, even though the Foveon example
    has 3 times as many meassured values.


    Roland
     
    Roland Karlsson, Nov 2, 2003
    #3
  4. "Roland Karlsson" <> wrote in message
    news:Xns9427A72629251rolandkarlssonchello@130.133.1.4...
    > "George Preddy" <> wrote in news:Yc1pb.19608
    > $:
    >
    > > Bayer (5 pixels of full color information):
    > >
    > > RGGBRGGBRGGBRGGBRGGB
    > >
    > >
    > >
    > >
    > > Foveon (20 pixels of full color information):
    > >
    > > BBBBBBBBBBBBBBBBBBBB
    > > GGGGGGGGGGGGGGGGGGGG
    > > RRRRRRRRRRRRRRRRRRRR

    >
    > What are you talking about?
    > A Bayer sensor looks like this
    >
    > R G R G R G R G R G R G
    > G B G B G B G B G B G B
    > R G R G R G R G R G R G
    > G B G B G B G B G B G B
    > R G R G R G R G R G R G
    > G B G B G B G B G B G B
    > R G R G R G R G R G R G
    > G B G B G B G B G B G B


    Exactly what I said, but I used just your first line, for both sensors.

    > And a Foveon sensor looks like this ..
    >
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    > C C C C C C C C C C C C
    >
    > ... where each C detects three "colors" that
    > can be mapped to R, G and B.


    Exactly.

    > Both examples above has a maximum resolution of
    > 12x8 pixels, even though the Foveon example
    > has 3 times as many meassured values.


    Equally correct , good job...

    Your "Bayer sensor" senses 24 pixels of full color information.
    Your "Foveon sensor" senses 96 pixels of full color information.

    24 compared to 96, but with the exact same MP rating on the store shelf.
     
    George Preddy, Nov 2, 2003
    #4
  5. "George Preddy" <> wrote in news:xK9pb.28259
    $:

    > Exactly what I said, but I used just your first line, for both sensors.


    Nope - you did not. Please compare your line with the actual sensor
    that I described.

    > Your "Bayer sensor" senses 24 pixels of full color information.


    No it has not - that is not how Bayer interpolation works.

    > 24 compared to 96, but with the exact same MP rating on the store shelf.


    This is one of your main beliefs, is it not? To try to
    make wrong into right by telling the world that there
    is an injustice made against Foveon from the Bayer
    community?

    Unfortunately, there exist no Bayer community. Bayer filtered
    sensors is just a technology. A good technology to solve
    the problems of color sensors. Another good technology is
    the Foveon technology. It is patented and only Foveon can
    currently use it. Soon someone will come up with an even
    better method of doing color pictures, maybe with a true
    RGB sensor at each pixel.


    Roland
     
    Roland Karlsson, Nov 2, 2003
    #5
  6. George Preddy

    George Kerby Guest

    On 11/2/03 9:25 AM, in article bo37l6$cum$, "Ray Fischer"
    <> wrote:

    > George Preddy <> wrote:
    >> Bayer (5 pixels of full color information):

    >
    > Get a life, idiot. Nobody believes your bullshit propaganda anymore
    > sicne you obviously don't understand what you write.

    Don't candy coat it Ray. Tell us how you REALLY feel!!!
    Liberals have such understanding don't they?!?
    ;-)


    _______________________________________________________________________________
    Posted Via Uncensored-News.Com - Accounts Starting At $6.95 - http://www.uncensored-news.com
    <><><><><><><> The Worlds Uncensored News Source <><><><><><><><>
     
    George Kerby, Nov 2, 2003
    #6
  7. "Roland Karlsson" <> wrote in message
    news:Xns9427B14D01DEArolandkarlssonchello@130.133.1.4...
    > "George Preddy" <> wrote in news:xK9pb.28259
    > $:
    >
    > > Exactly what I said, but I used just your first line, for both sensors.

    >
    > Nope - you did not. Please compare your line with the actual sensor
    > that I described.
    >
    > > Your "Bayer sensor" senses 24 pixels of full color information.

    >
    > No it has not - that is not how Bayer interpolation works.


    It knows colors that it doesn't sense? That is neat!

    > > 24 compared to 96, but with the exact same MP rating on the store shelf.

    >
    > This is one of your main beliefs, is it not? To try to
    > make wrong into right by telling the world that there
    > is an injustice made against Foveon from the Bayer
    > community?


    You posted those samples, I agreed with them. Your diagrams were right,
    there is a 4:1 Foveon advantange at the same MP rating, since the Foveon X3
    physically senses 3X the information per pixel, but doesn't then have to
    throw away 25% of the color information due to 2x2 unit scaling
    inefficiencies when used to sense only 3 primary colors.

    > Unfortunately, there exist no Bayer community. Bayer filtered
    > sensors is just a technology. A good technology to solve
    > the problems of color sensors. Another good technology is
    > the Foveon technology. It is patented and only Foveon can
    > currently use it. Soon someone will come up with an even
    > better method of doing color pictures, maybe with a true
    > RGB sensor at each pixel.


    Too late, Foveon already senses true RGB at every pixel.
     
    George Preddy, Nov 2, 2003
    #7
  8. On 11/2/03 2:25 PM, in article RKdpb.19962$,
    "George Preddy" <> wrote:

    >
    > "Roland Karlsson" <> wrote in message
    > news:Xns9427B14D01DEArolandkarlssonchello@130.133.1.4...
    >> "George Preddy" <> wrote in news:xK9pb.28259
    >> $:
    >>
    >>> Exactly what I said, but I used just your first line, for both sensors.

    >>
    >> Nope - you did not. Please compare your line with the actual sensor
    >> that I described.
    >>
    >>> Your "Bayer sensor" senses 24 pixels of full color information.

    >>
    >> No it has not - that is not how Bayer interpolation works.

    >
    > It knows colors that it doesn't sense? That is neat!
    >
    >>> 24 compared to 96, but with the exact same MP rating on the store shelf.

    >>
    >> This is one of your main beliefs, is it not? To try to
    >> make wrong into right by telling the world that there
    >> is an injustice made against Foveon from the Bayer
    >> community?

    >
    > You posted those samples, I agreed with them. Your diagrams were right,
    > there is a 4:1 Foveon advantange at the same MP rating, since the Foveon X3
    > physically senses 3X the information per pixel, but doesn't then have to
    > throw away 25% of the color information due to 2x2 unit scaling
    > inefficiencies when used to sense only 3 primary colors.
    >
    >> Unfortunately, there exist no Bayer community. Bayer filtered
    >> sensors is just a technology. A good technology to solve
    >> the problems of color sensors. Another good technology is
    >> the Foveon technology. It is patented and only Foveon can
    >> currently use it. Soon someone will come up with an even
    >> better method of doing color pictures, maybe with a true
    >> RGB sensor at each pixel.

    >
    > Too late, Foveon already senses true RGB at every pixel.
    >
    >

    George you JUST MUST get out for some air every once in awhile. I suggest
    that you read Miss Manners for an occasional break. She is SO stimulating.
    You horizons will be broadened immensely...


    _______________________________________________________________________________
    Posted Via Uncensored-News.Com - Accounts Starting At $6.95 - http://www.uncensored-news.com
    <><><><><><><> The Worlds Uncensored News Source <><><><><><><><>
     
    cultural neanderthal, Nov 2, 2003
    #8
  9. George Preddy

    Nils Rostedt Guest


    >> Soon someone will come up with an even
    > > better method of doing color pictures, maybe with a true
    > > RGB sensor at each pixel.

    >
    > Too late, Foveon already senses true RGB at every pixel.
    >


    Hmm... according to one source the Foveon "RGB pixel" is actually a
    concentric 3-ring "bullseye". This implies firstly, that different colors
    are sensed at spatially different locations. Secondly, that each primary
    color is sensed by only 1/3 of the pixel's total area. This brings the
    Foveon color resolution capability closer to that of a Bayer sensor than is
    claimed by some of the numbers we've seen. Comments?
     
    Nils Rostedt, Nov 2, 2003
    #9
  10. "Nils Rostedt" <> wrote in news:bo3rd1$16ed7p$1@ID-
    88261.news.uni-berlin.de:

    > Hmm... according to one source the Foveon "RGB pixel" is actually a
    > concentric 3-ring "bullseye". This implies firstly, that different colors
    > are sensed at spatially different locations. Secondly, that each primary
    > color is sensed by only 1/3 of the pixel's total area. This brings the
    > Foveon color resolution capability closer to that of a Bayer sensor than

    is
    > claimed by some of the numbers we've seen. Comments?


    No - not really.

    It is true that the areas for the three sensors are different,
    but it is not concentric circles. The different areas is a
    slight problem though.

    The main objective to an RGB sensor is that the three sensors
    do not detect anything closely resembling R, G and B. They
    have different color spectrums, but you have to apply a rather
    strong conversion matrix to get R, G and B. This conversion
    makes the purity of the R, G and B colors questionable,
    making it harder to get good color balance and large gamut.


    Roland
     
    Roland Karlsson, Nov 2, 2003
    #10
  11. "Nils Rostedt" <> wrote in message
    news:bo3rd1$16ed7p$-berlin.de...
    SNIP
    > Hmm... according to one source the Foveon "RGB pixel" is actually a
    > concentric 3-ring "bullseye". This implies firstly, that different colors
    > are sensed at spatially different locations. Secondly, that each primary
    > color is sensed by only 1/3 of the pixel's total area. This brings the
    > Foveon color resolution capability closer to that of a Bayer sensor than

    is
    > claimed by some of the numbers we've seen. Comments?


    Yes, the fill rate of the Foveon photosites is much less than for a simple
    CMOS photosite, because of the additional transistors and connections per
    site. It's probably more shaped like partially superimposed squares or
    rectangles than circles though.

    The storage capacity is also reduced because three wells have to be
    accommodated instead of one, and the possibility of charge leaking is much
    larger. This also means that there is less likelyhood for denser packing of
    pixels, so Foveon designs with more pixels will need to be physically larger
    and thus exponentially more expensive to produce (because the yield of a
    single wafer will go down).

    Resolution is a function of sampling pitch and (to a lesser extend) of
    sampling area. Point samples have a different response than area samples,
    and the Foveon partial areas are somewhere in between, but the main
    criterium is sampling pitch and that is the same for R, G and B in the
    Foveon design.

    Bart
     
    Bart van der Wolf, Nov 2, 2003
    #11
  12. "Nils Rostedt" <> wrote in message
    news:bo3rd1$16ed7p$-berlin.de...
    >
    > >> Soon someone will come up with an even
    > > > better method of doing color pictures, maybe with a true
    > > > RGB sensor at each pixel.

    > >
    > > Too late, Foveon already senses true RGB at every pixel.
    > >

    >
    > Hmm... according to one source the Foveon "RGB pixel" is actually a
    > concentric 3-ring "bullseye". This implies firstly, that different colors
    > are sensed at spatially different locations.


    Close, they are concentric wells in a bullseye pattern. Spatial offset = 0.

    > Secondly, that each primary
    > color is sensed by only 1/3 of the pixel's total area. This brings the
    > Foveon color resolution capability closer to that of a Bayer sensor than

    is
    > claimed by some of the numbers we've seen. Comments?


    Yes, you don't understand the technology, they are concentric photon wells.
    Think of it like 3 stacked paper cups. And resolution has nothing to do
    with that, sensor count is what drives optical resolution and that is why a
    6M sensor Bayer is not competitive with an SD-9--except with a 2 color B&W
    target, were the Foveon only wins by a little due to 3-4X resolution
    inflation on the Bayer.

    In full color, the SD-9 has about 2X the 10D's optical resolution overall.
    Here is the red/blue test (the SD-9 has a big sensor count advantage in all
    color channels)...

    http://www.outbackphoto.com/artofraw/raw_05/crop_10D_0000__0171_RT8.jpg
    http://www.outbackphoto.com/artofraw/raw_05/crop1_sd9_0000_00200.jpg

    As you can see with you own two, completely unbiased eyes, the SD-9 resolves
    about 1300 lines, the 10D is good for a very blurry 550. Imagine that,
    outresolved by a factor of 2.3X. Lookie here...

    SD-9 RGB sensors = 3.43M/3.43M/3.43M
    10D RGB sensors = 1.58M/3.15M/1.58M

    SD-9 Red/Blue sensor advantage? 230%

    The SD-9 is not magic afterall.
     
    George Preddy, Nov 3, 2003
    #12
  13. "Bart van der Wolf" <> wrote in message
    news:3fa585a0$0$58701$4all.nl...

    > The storage capacity is also reduced because three wells have to be
    > accommodated instead of one, and the possibility of charge leaking is much
    > larger.


    Now that is interesting. I wonder why the Foveon, with 170% of a 6MP
    Bayer's sensor count, is slightly a smaller chip (1.7X crop factor vs 1.6X)?
     
    George Preddy, Nov 3, 2003
    #13
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Dr Halonfires LesGirl

    M$ Latest Hotfix..Rating Critical.

    Dr Halonfires LesGirl, Jul 9, 2003, in forum: Computer Support
    Replies:
    7
    Views:
    515
    Dr Halonfires LesGirl
    Jul 10, 2003
  2. bob

    Rating Adware,a good reference

    bob, Sep 5, 2004, in forum: Computer Support
    Replies:
    0
    Views:
    449
  3. Bodysnatcher

    Web rating

    Bodysnatcher, Sep 12, 2004, in forum: Computer Support
    Replies:
    3
    Views:
    464
    Blinky the Shark
    Sep 12, 2004
  4. Melv

    site rating

    Melv, Oct 30, 2004, in forum: Computer Support
    Replies:
    4
    Views:
    554
    Muppet
    Oct 30, 2004
  5. *BUSY*
    Replies:
    2
    Views:
    780
    Franz Leu
    Oct 21, 2009
Loading...

Share This Page