How About a 3.1 MP B&W Camera? Nikon?

[tom jacobson]

I would love to see a camera like the Nikon 5700 that uses a mono sensor and shoots only in B&W. If you eliminate the Bayer Pattern of color filters laid out in grids on the sensor, the output of the sensor can be spectacular. Witness the Foveon! Maybe that will be the great B&W camera, come to think of it. If there are color problems with crossover, etc, it will still be a killer B&W camera. At the link below you will see the output of a 1.3 MP mono camera that will give a clue to what a 3.1 MP will do. Just imagine that it is only one quadrant of a larger file. I have made prints up to 20x30 from the 1.3MP files from the DCS 2000e Mono, and they look great, except for some jaggie diagonals. The image linked is straight from the camera, raw to tiff. No tonal correction, no sharpening at all, in camera or later. It's not a great example, but wil do to show the quality of the files. http://www.pbase.com/image/2383434/medium----
TJ

 

[Rich Shelton]

I'm with you. This would be a KILLER camera! My only concern is what wavelengths would be best to tune the CCDs to!? A camera intended for portraiture might be set up a lot different from a landscape camera...

Rich Shelton

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[tom jacobson]

I think the best thing is to try for a linear sensor. The wavelength that you refer to can be achieved by filters just like the ones used for B&W work on film, such as red, yellow, light orange, etc. The 2000e tends to be a bit red sensitive, but I like the look. I've tried filters and they work very well, but of course, you lose a stop or two. TJ

Rich Shelton

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[Aaron van de Sande]

I have thought about this too. Creating a ccd without the filters over the sensors is probably not that difficult, and there is no extrapolation for the firmware to do to create output. The d60 would create a 24mp perfectly sharp greyscale....

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[tom jacobson]

Aaron,

You mean upsampled? Yes, you could surely take the 6MP up to 24MP in GF and have a great image. I doubt that Canon would do that , but maybe they would. Kodak does a mono version of the 760 ( at only $10,000 list!). I think a cheaper model with B&W dedication would sell better, but who knows.
TJ

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[Rick Boden]

A great idea for the reasons given but not only that, you could shoot still lifes with tri-colour filters and create hi-rez colour images.

Rick

 

[Aaron van de Sande]

Not upsampled, real pixels. You remove the color filters from the sensors and treat them each as a seperate pixel.Currently they use 4 discreate sensors and combine them into one pixel. 6x4 24 MP.

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[Isaac Crawford]

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

Did you know that Kodak offers a black only ribbon for some of its dyesub printers? This seems to me to be the best option for people wanting continious tone B&W digital printing...

Isaac

 

[kessler]

I thought each of the four discrete sensors was used in creating four final pixels, thus in the end there are the same number of sensors as pixels you started with. Is this camera different?

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[Aaron van de Sande]

Well, then I am wrong. I just guessed that was the way they did it. No wonder people are so excited about foveon....
I would still like a 6mp b+w d60 : )

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[DavidMillier]

The D60 like all bayer cameras except fuji superccd, produces 1 output pixel for every photosite on the sensor i.e. the D60 has c. 6 million photosites and produces 6 MP images.

However, it doesn't do this directly - each output pixel is not in a one to one relationship with a specific photosite.

This is because the photosites cannot see the full colour spectrum. Instead a filter array is used to make each photosite sensitive to a particular narrow colour spectrum - either Red, green or blue.

The camera then takes the 6 million red, green or blue, photosite data items and feeds them into a computational algorithm that analyses groups of adjacent photosites.

From this analysis, the camera computes (guesses!) the appropriate colour and brightness values for each of the 6 million output pixels.

If you removed the colour filter array you would be left with a 6 million photosite brightness detector capable only of measuring a greyscale.

No bayer computation would be required of course, so although there would still be 6 million pixels in the output file they will be directly measured pixels rather than computed/interpolated/guessed pixels. This should produce more accurate pictures.

The foveon is somewhat like the monochrome sensor except it can directlly measure full colour as well as brightness and again it can do this without computation/interpolation/guessing.

The superccd is like the D60 sensor except the particular interpolation algorithm they use to deal with their funny shape photosites also inflates the final output pixel count by a factor of x2 or so - but this doesn't make it a true 12MP sensor because it still has only 6Million photosites.

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

--
TJ

 

[Hariette]

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[Hariette]

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[Hariette]

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[Hariette]

I would love to see a camera like the Nikon 5700 that uses a mono
sensor and shoots only in B&W. If you eliminate the Bayer Pattern
of color filters laid out in grids on the sensor, the output of the
sensor can be spectacular. Witness the Foveon! Maybe that will be
the great B&W camera, come to think of it. If there are color
problems with crossover, etc, it will still be a killer B&W camera.
At the link below you will see the output of a 1.3 MP mono camera
that will give a clue to what a 3.1 MP will do. Just imagine that
it is only one quadrant of a larger file. I have made prints up to
20x30 from the 1.3MP files from the DCS 2000e Mono, and they look
great, except for some jaggie diagonals. The image linked is
straight from the camera, raw to tiff. No tonal correction, no
sharpening at all, in camera or later. It's not a great example,
but wil do to show the quality of the files.
http://www.pbase.com/image/2383434/medium----
TJ

 

[hardsuit01]

hardsuit wrote:
Wait , a minute. If you remove the color filtering , how Is the sensor
gona read the unfiltered data . my, guess would be 2 bit data " 0" " 1"
only black and white no middle grey. forget full detail images.
that mono CCD will only be good for text or line drawings.
In film we see this. silver B+W negs have clear areas and silver areas.
fine for analogue printing, try takeing it to digital lab , not so easy.
enter mono/color film, its got all the color filters in place ,
essental in B+W imaging ! it only uses black dye clouds, silver of cource
need not apply. what does this mean ? images are superior to silver B+W
have better sharpness compared to silver neg, better shadow detail.
any way back to siliicon CCD's . Even with shooting in mono with my
finepix 4900, I use top shelf B+W filters to enhance the image .

' lets see what's out there.....engage'

 

[tom jacobson]

Please take a look at the first message on this thread for a link to a picture taken without color filters. TJ

hardsuit wrote:

Wait , a minute. If you remove the color filtering , how Is the sensor
gona read the unfiltered data . my, guess would be 2 bit data " 0"
" 1"
only black and white no middle grey. forget full detail images.
that mono CCD will only be good for text or line drawings.
In film we see this. silver B+W negs have clear areas and silver
areas.
fine for analogue printing, try takeing it to digital lab , not so
easy.
enter mono/color film, its got all the color filters in place ,
essental in B+W imaging ! it only uses black dye clouds, silver of
cource
need not apply. what does this mean ? images are superior to silver
B+W
have better sharpness compared to silver neg, better shadow detail.
any way back to siliicon CCD's . Even with shooting in mono with my
finepix 4900, I use top shelf B+W filters to enhance the image .

' lets see what's out there.....engage'

--
TJ

 

[netrox]

Sensors do sense grayscale. A photosite detects the brightness level and sends an electronic value from 0 (black) to 255 (white) to the processor chip which in turn forms an image. It's analog to digital conversion. It's definintely not a one-bit process. You do get full detail images, far more than color images since there's no color interpolation involved that can smear the fine details.

hardsuit wrote:

Wait , a minute. If you remove the color filtering , how Is the sensor
gona read the unfiltered data . my, guess would be 2 bit data " 0"
" 1"
only black and white no middle grey. forget full detail images.

 

[Rinus Borgsteede]

Of course when shooting in monochrome on any digital camera, the output is still RGB. The only thing that has changed is that the camera has mixed the channels together. It solves a few problems but the linearity of the image data needs color filters to change the look of the B/W image. Three images through individual RGB filters will combine into an awsome color image. There are some scan backs that already do that. The lack of practicallity is why we use bayer type interpolation in the first place. Shooting a B/W image with a color sensor works best if the RGB filters stay where they are so that selective filtering can be used either in the camera with a filter wheel (that selects color sensitivity of the individual channels) or so that the ratios of channels can be mixed later in Photoshop. I prefer a slight increased red sensitivity for portraits but a opposite green sensitivity for nature shots. A single monochrome CCD is like shooting with a single type of film without the opportunity to change. Contrast can be affected with a zone system like exposure method but as for color sensitivity you'll be stuck with filters unless you maintain the colored pixels.
Rinus

 

[Ken Phillips]

Each photosite on the chip is essentially a photon counter ... and can easily, with today's technology, register 64K or more different values (16 bits or more!). Of course the low photon counts have to compete with miscellaneous noise! If an unfiltered chip were employed you would have the option of using any type of external filter ... if the chip were IR sensitive this would be a lot of fun! Any old-timers ever shoot a cityscape through the haze on BW with a red filter?

hardsuit wrote:

Wait , a minute. If you remove the color filtering , how Is the sensor
gona read the unfiltered data . my, guess would be 2 bit data " 0"
" 1"
only black and white no middle grey. forget full detail images.
that mono CCD will only be good for text or line drawings.
In film we see this. silver B+W negs have clear areas and silver
areas.
fine for analogue printing, try takeing it to digital lab , not so
easy.
enter mono/color film, its got all the color filters in place ,
essental in B+W imaging ! it only uses black dye clouds, silver of
cource
need not apply. what does this mean ? images are superior to silver
B+W
have better sharpness compared to silver neg, better shadow detail.
any way back to siliicon CCD's . Even with shooting in mono with my
finepix 4900, I use top shelf B+W filters to enhance the image .

' lets see what's out there.....engage'