Minimum pixel size

Discussion in 'Digital Photography' started by Alfred Molon, Jul 31, 2004.

  1. Alfred Molon

    Alfred Molon Guest

    Alfred Molon, Jul 31, 2004
    #1
    1. Advertising

  2. Of course 1 micron is possible, but noise is the key point.
    It is also worth mentioning that lens resolution would also need
    to be very good, which would cost more.

    You can assume that the camera companies trade off the cost
    of the lens verses the cost of the sensor. I would imagine relatively
    small sensors give the lowest total cost for both.

    The high end cameras have lower noise because of larger pixels.
    This costs more for the system.

    Joe


    "Alfred Molon" <> wrote in message
    news:...
    > Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    > to 1 micrometer and below (if we disregard noise issues for a moment) ?
    > --
    >
    > Alfred Molon
    > ------------------------------
    > http://groups.yahoo.com/group/Olympus_405080/
    > Olympus 5050 resource - http://www.molon.de/5050.html
    > Olympus 5060 resource - http://www.molon.de/5060.html
    > Olympus 8080 resource - http://www.molon.de/8080.html
    Joseph Schutz, Jul 31, 2004
    #2
    1. Advertising

  3. Alfred Molon

    dj_nme Guest

    Alfred Molon wrote:
    > Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    > to 1 micrometer and below (if we disregard noise issues for a moment) ?


    My first guess would be that the absolute lower limit would be the
    wavelength of near infra-red (around 3 microns). Any smaller and you
    truly run the risk of not being able to capture some of the visible
    spectrum.

    Just my 2c.
    dj_nme, Jul 31, 2004
    #3
  4. In article <CTGOc.219271$XM6.177658@attbi_s53>, Joseph Schutz
    <> writes
    >
    >
    >"Alfred Molon" <> wrote in message
    >news:...
    >> Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    >> to 1 micrometer and below (if we disregard noise issues for a moment) ?
    >> --


    >Of course 1 micron is possible, but noise is the key point.
    >It is also worth mentioning that lens resolution would also need
    >to be very good, which would cost more.


    The limit must be somewhere of the same order as the limit for computer
    chips (since the same manufacturing techniques are used). IIRC, these
    are now using 90nm (0.09 microns) technology. This involves using far-UV
    light and extremely sophisticated process lenses.

    However, to take this as a practical proposition overlooks the fact that
    there is a hard limit to how much a lens can resolve: diffraction. This
    means that any lens, even a perfect one, will produce not a single point
    image (of a single point object) but a slightly blurred spot. The size
    of this spot depends on the relative aperture of the lens (the
    f-number). Any attempt to resolve below this point will just make the
    fuzzy dots bigger.

    When do we reach this point? Well, figures I saw a couple of weeks ago
    suggest we are there about now. I can't remember where I saw them, it
    may have been here. However, a quick back-of-the-envelope calculation
    suggests that for a lens at f/2 the theoretical limit for resolving
    detail is about 750 lp/mm. Ignoring all the sophisticated aspects of
    MTF, sampling theory, Nyquist limits and the like, this implies at most
    1500 pixels/mm, i.e. 670nm per pixel. Taking the complex sampling issues
    into account, this is going to come out to 1 micron (1000 nm) or worse.

    And that is at f/2 - at f/4 it will be 2 microns, at f/8 it will be 4
    and so on. If someone can refer us to the more sophisticated table I saw
    that would of course be in all probability worked out on a more
    sophisticated basis, but the above should be "in the ball park".

    The solution? Just as for film: bigger sensor pixels, bigger sensors. It
    *will* happen, but only when commercial pressure makes it so.
    >
    >You can assume that the camera companies trade off the cost
    >of the lens verses the cost of the sensor. I would imagine relatively
    >small sensors give the lowest total cost for both.
    >
    >The high end cameras have lower noise because of larger pixels.
    >This costs more for the system.
    >

    Whilst true, this is not what was asked.

    --
    David Littlewood
    David Littlewood, Jul 31, 2004
    #4
  5. Gene Palmiter, Jul 31, 2004
    #5
  6. Alfred Molon

    Matti Vuori Guest

    Alfred Molon <> wrote in
    news::

    > Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    > to 1 micrometer and below (if we disregard noise issues for a moment) ?


    A CCD doesn't have pixels. They only exist after calculation from the
    sensor information.

    --
    Matti Vuori, <http://sivut.koti.soon.fi/mvuori/index-e.htm>
    Matti Vuori, Jul 31, 2004
    #6
  7. Alfred Molon

    DJ Guest

    On Sat, 31 Jul 2004 10:43:56 +0000 (UTC), Matti Vuori <>
    wrote:

    >Alfred Molon <> wrote in
    >news::
    >
    >> Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    >> to 1 micrometer and below (if we disregard noise issues for a moment) ?

    >
    >A CCD doesn't have pixels. They only exist after calculation from the
    >sensor information.


    I am in awe of the clarity of your reasoning and the lucidness of your
    explanation.
    DJ, Jul 31, 2004
    #7
  8. "Alfred Molon" <> wrote in message
    news:...

    > Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    > to 1 micrometer and below (if we disregard noise issues for a moment) ?


    It's hard to understand how a pixel could usefully be smaller than a
    wavelength of light.
    Andrew Koenig, Jul 31, 2004
    #8
  9. Alfred Molon

    Don Stauffer Guest

    Indeed it will be very hard to get a lens to produce a blur circle
    smaller than the wavelength of the light being imaged. Thus it is
    unlikely to make any sense trying to make a chip with a resolution any
    smaller than about 0.65 microns (red light).

    For some military purposes it may be reasonable to employ UV imaging,
    and go slightly smaller, but the pictures would not be very useful for
    consumer cameras.

    Joseph Schutz wrote:
    >
    > Of course 1 micron is possible, but noise is the key point.
    > It is also worth mentioning that lens resolution would also need
    > to be very good, which would cost more.
    >
    > You can assume that the camera companies trade off the cost
    > of the lens verses the cost of the sensor. I would imagine relatively
    > small sensors give the lowest total cost for both.
    >
    > The high end cameras have lower noise because of larger pixels.
    > This costs more for the system.
    >
    > Joe
    >
    > "Alfred Molon" <> wrote in message
    > news:...
    > > Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    > > to 1 micrometer and below (if we disregard noise issues for a moment) ?
    > > --
    > >
    > > Alfred Molon
    > > ------------------------------
    > > http://groups.yahoo.com/group/Olympus_405080/
    > > Olympus 5050 resource - http://www.molon.de/5050.html
    > > Olympus 5060 resource - http://www.molon.de/5060.html
    > > Olympus 8080 resource - http://www.molon.de/8080.html


    --
    Don Stauffer in Minnesota

    webpage- http://www.usfamily.net/web/stauffer
    Don Stauffer, Jul 31, 2004
    #9
  10. Alfred Molon

    Guest

    In message <>,
    David Littlewood <> wrote:

    >However, to take this as a practical proposition overlooks the fact that
    >there is a hard limit to how much a lens can resolve: diffraction. This
    >means that any lens, even a perfect one, will produce not a single point
    >image (of a single point object) but a slightly blurred spot. The size
    >of this spot depends on the relative aperture of the lens (the
    >f-number). Any attempt to resolve below this point will just make the
    >fuzzy dots bigger.


    It may not be efficient, but if you out-resolve the lens by a margin,
    then bayer demosaicing becomes nearly artifact-free, no AA filter is
    necessary, and pixel-level noise can be totally written off because you
    know for a fact that it is not image detail. You can remove the noise
    with a low-pass filter that doesn't affect image detail. The question
    is whether there will be enough dynamic range left after the noise of
    the tiny sensels.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
    , Jul 31, 2004
    #10
  11. Alfred Molon

    Guest

    In message <KEKOc.672$HN5.198@trndny02>,
    "Gene Palmiter" <> wrote:

    >PIXELS DON'T HAVE A SIZE!!!!!


    That's correct, in a totally different context, such as a discussion of
    what size the pixels in a JPEG file are. The pixels on a sensor
    (sensels) have area, and pitch.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
    , Jul 31, 2004
    #11
  12. In article <>,
    writes
    >In message <>,
    >David Littlewood <> wrote:
    >
    >>However, to take this as a practical proposition overlooks the fact that
    >>there is a hard limit to how much a lens can resolve: diffraction. This
    >>means that any lens, even a perfect one, will produce not a single point
    >>image (of a single point object) but a slightly blurred spot. The size
    >>of this spot depends on the relative aperture of the lens (the
    >>f-number). Any attempt to resolve below this point will just make the
    >>fuzzy dots bigger.

    >
    >It may not be efficient, but if you out-resolve the lens by a margin,
    >then bayer demosaicing becomes nearly artifact-free, no AA filter is
    >necessary, and pixel-level noise can be totally written off because you
    >know for a fact that it is not image detail. You can remove the noise
    >with a low-pass filter that doesn't affect image detail. The question
    >is whether there will be enough dynamic range left after the noise of
    >the tiny sensels.


    Yes; if you look at the rest of what I said, you will see I acknowledged
    these factors but ignored them to give a simple approximation.

    In practice, you can probably resolve down to about half the wavelength
    of the light - but only with an extremely large aperture lens. About
    200-250nm is the limit for visible light.
    --
    David Littlewood
    David Littlewood, Jul 31, 2004
    #12
  13. <> wrote in message
    news:...
    > In message <KEKOc.672$HN5.198@trndny02>,
    > "Gene Palmiter" <> wrote:
    >
    > >PIXELS DON'T HAVE A SIZE!!!!!

    >
    > That's correct, in a totally different context, such as a discussion

    of
    > what size the pixels in a JPEG file are. The pixels on a sensor
    > (sensels) have area, and pitch.


    Correct, so let's call them what they are, photo sensors, not pixels.
    Sensors are input, pixels are output.

    Bart
    Bart van der Wolf, Jul 31, 2004
    #13
  14. On Sat, 31 Jul 2004 06:59:46 +0200, Alfred Molon
    <> wrote:

    >Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    >to 1 micrometer and below (if we disregard noise issues for a moment) ?


    The hard limit would governed by the wavelength of the lithography
    process used, of which the current process used to fabricate CPUs can
    create individual traces that are 0.09 micron. Based on that, and
    allowing for support circuitry, it would be theoretically possible
    make an individual sensor element which is perhaps one micron across
    I'd guess. You'd probably need to pay the price of the fab for a lens
    that could resolve that much detail, and I wouldn't expect a Bayer
    array that fine any time soon either.

    Why do you ask? Got bored with infra-red and want to try UV or even
    X-Ray photography, or just trying to predict when the megapixel "race"
    will end? ;)

    Andy
    Andy Blanchard, Aug 1, 2004
    #14
  15. Thanks...been out all day and could not defend myself....



    "Bart van der Wolf" <> wrote in message
    news:410c19f2$0$48933$4all.nl...
    >
    > <> wrote in message
    > news:...
    > > In message <KEKOc.672$HN5.198@trndny02>,
    > > "Gene Palmiter" <> wrote:
    > >
    > > >PIXELS DON'T HAVE A SIZE!!!!!

    > >
    > > That's correct, in a totally different context, such as a discussion

    > of
    > > what size the pixels in a JPEG file are. The pixels on a sensor
    > > (sensels) have area, and pitch.

    >
    > Correct, so let's call them what they are, photo sensors, not pixels.
    > Sensors are input, pixels are output.
    >
    > Bart
    >
    Gene Palmiter, Aug 1, 2004
    #15
  16. "Joseph Schutz" <> wrote in message news:<CTGOc.219271$XM6.177658@attbi_s53>...

    > The high end cameras have lower noise because of larger pixels.
    > This costs more for the system.


    Sigma SD9's (and SD10's) have the largest pixels of any digital camera
    currently being made (9.2 microns square, about 30% more area than the
    $8,000 Canon 1Ds's pixels). Cost is about $600.
    Georgette Preddy, Aug 1, 2004
    #16
  17. "Bart van der Wolf" <> wrote in message news:<410c19f2$0$48933$4all.nl>...
    > <> wrote in message
    > news:...
    > > In message <KEKOc.672$HN5.198@trndny02>,
    > > "Gene Palmiter" <> wrote:
    > >
    > > >PIXELS DON'T HAVE A SIZE!!!!!

    > >
    > > That's correct, in a totally different context, such as a discussion

    > of
    > > what size the pixels in a JPEG file are. The pixels on a sensor
    > > (sensels) have area, and pitch.

    >
    > Correct, so let's call them what they are, photo sensors, not pixels.
    > Sensors are input, pixels are output.


    Read a dictionary.
    Georgette Preddy, Aug 1, 2004
    #17
  18. Alfred Molon

    DJ Guest

    On 31 Jul 2004 20:20:30 -0700, (Georgette Preddy)
    wrote:

    >"Joseph Schutz" <> wrote in message news:<CTGOc.219271$XM6.177658@attbi_s53>...
    >
    >> The high end cameras have lower noise because of larger pixels.
    >> This costs more for the system.

    >
    >Sigma SD9's (and SD10's) have the largest pixels of any digital camera
    >currently being made (9.2 microns square, about 30% more area than the
    >$8,000 Canon 1Ds's pixels). Cost is about $600.


    Well of course, there are so few of them that naturally they are larger.
    DJ, Aug 1, 2004
    #18
  19. Alfred Molon

    Guest

    In message <>,
    DJ <> wrote:

    >On 31 Jul 2004 20:20:30 -0700, (Georgette Preddy)
    >wrote:
    >
    >>"Joseph Schutz" <> wrote in message news:<CTGOc.219271$XM6.177658@attbi_s53>...
    >>
    >>> The high end cameras have lower noise because of larger pixels.
    >>> This costs more for the system.

    >>
    >>Sigma SD9's (and SD10's) have the largest pixels of any digital camera
    >>currently being made (9.2 microns square, about 30% more area than the
    >>$8,000 Canon 1Ds's pixels). Cost is about $600.

    >
    >Well of course, there are so few of them that naturally they are larger.


    In the SD9, they have no microlenses, and therefore, receive only about
    30% of the light hitting the sensor. The camera should have been
    designed for very low ISOs, as it can take 3.3x the exposure of the SD10
    before saturating the sensors.

    Bottom line is that the SD9 only benefits half of the larger-sensor
    benefits, as its sensitivity is low.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
    , Aug 1, 2004
    #19
  20. Alfred Molon

    jpc Guest

    On Sat, 31 Jul 2004 06:59:46 +0200, Alfred Molon
    <> wrote:

    >Is there a lower limit to pixel size, i.e. could pixel size in a CCD go
    >to 1 micrometer and below (if we disregard noise issues for a moment) ?



    I doubt we can disregard noise. From some data I've seen a CCD can
    collect somewhere between 400 to 800 photo electrons per square
    microns with large sensors. With much smaller sensors I'd expect the
    limit to decrease drastically because there would be more interaction
    between the potential fields that define the individual sensors.

    Even if we assume a 1 micron square sensor well can hold 400 photo
    electrons and there are no other noise sources this gives us a max S/N
    of 20 in the brightest highlights because of background shot noise and
    virtually no S/N in the shadows. In reality, the S/N would be far
    worse.

    As for CMOS, there would be even less silicon real estate for the
    photodiodes because of the electronics built into the sensor chips.

    jpc
    jpc, Aug 1, 2004
    #20
    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. Abrasha

    Hot pixel vs. stuck pixel

    Abrasha, Aug 2, 2003, in forum: Digital Photography
    Replies:
    5
    Views:
    8,468
    Steven Buglass
    Sep 2, 2003
  2. Tom Thackrey

    Re: Pixel size of individual Pixel

    Tom Thackrey, Sep 14, 2003, in forum: Digital Photography
    Replies:
    2
    Views:
    810
  3. Robert E. Williams

    Re: Pixel size of individual Pixel

    Robert E. Williams, Sep 14, 2003, in forum: Digital Photography
    Replies:
    2
    Views:
    536
    Don Stauffer
    Sep 16, 2003
  4. Peter H

    Fuji S3000 - 3.2 m/pixel or 6 m/pixel?

    Peter H, Nov 18, 2003, in forum: Digital Photography
    Replies:
    3
    Views:
    686
  5. stu7

    pixels / pixel size / sensor size

    stu7, Apr 25, 2012, in forum: Digital Photography
    Replies:
    4
    Views:
    596
    Wolfgang Weisselberg
    Apr 30, 2012
Loading...

Share This Page