Sign in with
Sign up | Sign in
Your question

Does True B&W Digital Mode Exist?

Last response: in Digital Camera
Share
March 9, 2005 11:06:38 AM

Archived from groups: rec.photo.digital (More info?)

I recall reading that if the sensors were dedicated to B&W, it would be
possible to get really great quality B&W images, just like film, similar
physics. It seems not too hard to bracket three exposures with color
filters for color if you wanted too. This would eliminate the averaging
every third pixel bayer problems & produce considerably sharper images.
Does any such beast exist? I'm just curious.
March 9, 2005 6:16:14 PM

Archived from groups: rec.photo.digital (More info?)

paul wrote:
> I recall reading that if the sensors were dedicated to B&W, it would be
> possible to get really great quality B&W images, just like film, similar
> physics.

I doubt it. Part of the reason B&W film looks like B&W film is due to
the fact that it doesn't respond equally to different colors of light.
In fact, different films have different response curves.

At best you could emulate a single emulsion. With PS plugins you can
emulate your choice of emulsion.

Bob
Anonymous
March 9, 2005 7:08:37 PM

Archived from groups: rec.photo.digital (More info?)

"Ron Baird" <ronbaird@kodak.com> writes:

> There was a pro camera not long ago that did only BW. Also, most of the
> general use Kodak digital cameras have settings to yield BW or sepia, etc.
> Let me know if you have a particular need and I will try to help.

Lots of the other cameras have *settings* to produce B&W output, but
that doesn't change the color filter over the sensor.

Another point is that that color filter eats light; removing it would
considerably increase the sensitivity. God I'd love a camera that I
could push a button and have it move the color filter aside! (I see
problems with actually making one; the alignment required must be
*really extremely* precise, for one thing. And the people willing to
pay more for higher ISO and better B&W may not be common enough to
make it worth doing commercially.)
--
David Dyer-Bennet, <mailto:D d-b@dd-b.net>, <http://www.dd-b.net/dd-b/&gt;
RKBA: <http://noguns-nomoney.com/&gt; <http://www.dd-b.net/carry/&gt;
Pics: <http://dd-b.lighthunters.net/&gt; <http://www.dd-b.net/dd-b/SnapshotAlbum/&gt;
Dragaera/Steven Brust: <http://dragaera.info/&gt;
Related resources
Anonymous
March 9, 2005 9:35:29 PM

Archived from groups: rec.photo.digital (More info?)

paul wrote:

> I recall reading that if the sensors were dedicated to B&W, it would be
> possible to get really great quality B&W images, just like film, similar
> physics. It seems not too hard to bracket three exposures with color
> filters for color if you wanted too. This would eliminate the averaging
> every third pixel bayer problems & produce considerably sharper images.
> Does any such beast exist? I'm just curious.

I read a magazine article that refered to a B&W only digital camera,
possibly a Kodak that was made several years back.

Pete

--
http://www.petezilla.co.uk
Anonymous
March 9, 2005 9:35:30 PM

Archived from groups: rec.photo.digital (More info?)

Hi Peter,

There was a pro camera not long ago that did only BW. Also, most of the
general use Kodak digital cameras have settings to yield BW or sepia, etc.
Let me know if you have a particular need and I will try to help.

Ron Baird
Eastman Kodak Company




"Peter Chant" <pete@petezilla.co.uk> wrote in message
news:398tugF5uc7vvU2@individual.net...
> paul wrote:
>
> > I recall reading that if the sensors were dedicated to B&W, it would be
> > possible to get really great quality B&W images, just like film, similar
> > physics. It seems not too hard to bracket three exposures with color
> > filters for color if you wanted too. This would eliminate the averaging
> > every third pixel bayer problems & produce considerably sharper images.
> > Does any such beast exist? I'm just curious.
>
> I read a magazine article that refered to a B&W only digital camera,
> possibly a Kodak that was made several years back.
>
> Pete
>
> --
> http://www.petezilla.co.uk
Anonymous
March 9, 2005 11:42:04 PM

Archived from groups: rec.photo.digital (More info?)

On Wed, 09 Mar 2005 08:06:38 -0800, paul <paul@not.net> wrote:

>I recall reading that if the sensors were dedicated to B&W, it would be
>possible to get really great quality B&W images, just like film, similar
>physics. It seems not too hard to bracket three exposures with color
>filters for color if you wanted too. This would eliminate the averaging
>every third pixel bayer problems & produce considerably sharper images.
>Does any such beast exist? I'm just curious.


The issue isn't so much the capture (or the scanning.)

What's really missing are dedicated BW output devices.

For now, the best thing for that is an Epson inkjet
with a decent RIP, or Piezography, or maybe the
Harrington quadtone RIP. (ImagePrint and StudioPrint
are RIPs I hear mentioned often for dedicated BW
printing.)

There's really not much issue with the capture
devices. True, a "monochrome" sensor could give
you better resolution and a bit more sensitivity.
What you *really* want for BW is more bit depth
and dynamic range.

But there are lots of advantages to shooting BW
with a Bayer CCD. For example, you can simulate
virtually any filter you might imagine. You can
generate your BW image by varying the relative
contributions of the red/green/blue channels.
Really lots of possibilities.


rafe b.
http://www.terrapinphoto.com
March 9, 2005 11:42:05 PM

Archived from groups: rec.photo.digital (More info?)

OK thanks all. I was just pondering.

It seems you can do OK with a color bayer digicam (with options open) &
there's not really any market for this high end concept. I'm guessing it
will be an option in the future though.
Anonymous
March 10, 2005 12:37:38 AM

Archived from groups: rec.photo.digital (More info?)

harvey wrote:
i believe that the black and white version was called the 760M and it
was very pricey

paul wrote:
> I recall reading that if the sensors were dedicated to B&W, it would be
> possible to get really great quality B&W images, just like film, similar
> physics. It seems not too hard to bracket three exposures with color
> filters for color if you wanted too. This would eliminate the averaging
> every third pixel bayer problems & produce considerably sharper images.
> Does any such beast exist? I'm just curious.
Anonymous
March 10, 2005 1:11:03 AM

Archived from groups: rec.photo.digital (More info?)

"paul" <paul@not.net> wrote in message
news:6NGdnayrpoGSgrLfRVn-2w@speakeasy.net...
>I recall reading that if the sensors were dedicated to B&W, it would be
>possible to get really great quality B&W images, just like film, similar
>physics. It seems not too hard to bracket three exposures with color
>filters for color if you wanted too. This would eliminate the averaging
>every third pixel bayer problems & produce considerably sharper images.
>Does any such beast exist? I'm just curious.

My 20D has a BW mode and you can select what color filter effect you
want....
Anonymous
March 10, 2005 1:50:09 AM

Archived from groups: rec.photo.digital (More info?)

On Wed, 09 Mar 2005 19:21:00 -0800, paul <paul@not.net> wrote:

>OK thanks all. I was just pondering.
>
>It seems you can do OK with a color bayer digicam (with options open) &
>there's not really any market for this high end concept. I'm guessing it
>will be an option in the future though.


It is available, now, at the very high end,
in some Leaf backs. Very high end. Or at
least it was, just a couple of years back.

In the future, if Foveon technology ever makes
it into larger area arrays, you could have the
best of both worlds, since there are three
receptors at each physical photosite.

But that's a very big "if" and I'm not holding
my breath.


rafe b.
http://www.terrapinphoto.com
Anonymous
March 10, 2005 4:04:29 AM

Archived from groups: rec.photo.digital (More info?)

In message <XvKXd.3347$4k2.2376@fe2.columbus.rr.com>,
"Fitpix" <David@delawarestudioNOSPAM.com> wrote:

>"paul" <paul@not.net> wrote in message
>news:6NGdnayrpoGSgrLfRVn-2w@speakeasy.net...
>>I recall reading that if the sensors were dedicated to B&W, it would be
>>possible to get really great quality B&W images, just like film, similar
>>physics. It seems not too hard to bracket three exposures with color
>>filters for color if you wanted too. This would eliminate the averaging
>>every third pixel bayer problems & produce considerably sharper images.
>>Does any such beast exist? I'm just curious.

>My 20D has a BW mode and you can select what color filter effect you
>want....

It's just making a B&W image from multi-color-filtered data. It's not
true B&W.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
March 10, 2005 4:04:30 AM

Archived from groups: rec.photo.digital (More info?)

<JPS@no.komm> wrote in message
news:147v2119ud7uoaoi1aggumh907si14f7ku@4ax.com...
> In message <XvKXd.3347$4k2.2376@fe2.columbus.rr.com>,
> "Fitpix" <David@delawarestudioNOSPAM.com> wrote:
>
>>"paul" <paul@not.net> wrote in message
>>news:6NGdnayrpoGSgrLfRVn-2w@speakeasy.net...
>>>I recall reading that if the sensors were dedicated to B&W, it would be
>>>possible to get really great quality B&W images, just like film, similar
>>>physics. It seems not too hard to bracket three exposures with color
>>>filters for color if you wanted too. This would eliminate the averaging
>>>every third pixel bayer problems & produce considerably sharper images.
>>>Does any such beast exist? I'm just curious.
>
>>My 20D has a BW mode and you can select what color filter effect you
>>want....
>
> It's just making a B&W image from multi-color-filtered data. It's not
> true B&W.
> --
>
> <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
> John P Sheehy <JPS@no.komm>
> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

But it works just fine. But you're right, it's basically an enhanced
version of greyscale. The only true B&W digital camera I know of was a
version of the Kodak DCS-1.

--
Skip Middleton
http://www.shadowcatcherimagery.com
Anonymous
March 10, 2005 4:04:31 AM

Archived from groups: rec.photo.digital (More info?)

On Wed, 9 Mar 2005 17:30:35 -0800, "Skip M" <shadowcatcher@cox.net>
wrote:

><JPS@no.komm> wrote in message
>news:147v2119ud7uoaoi1aggumh907si14f7ku@4ax.com...
>> In message <XvKXd.3347$4k2.2376@fe2.columbus.rr.com>,
>> "Fitpix" <David@delawarestudioNOSPAM.com> wrote:
>>
>>>"paul" <paul@not.net> wrote in message
>>>news:6NGdnayrpoGSgrLfRVn-2w@speakeasy.net...
>>>>I recall reading that if the sensors were dedicated to B&W, it would be
>>>>possible to get really great quality B&W images, just like film, similar
>>>>physics. It seems not too hard to bracket three exposures with color
>>>>filters for color if you wanted too. This would eliminate the averaging
>>>>every third pixel bayer problems & produce considerably sharper images.
>>>>Does any such beast exist? I'm just curious.
>>
>>>My 20D has a BW mode and you can select what color filter effect you
>>>want....
>>
>> It's just making a B&W image from multi-color-filtered data. It's not
>> true B&W.
>> --
>>
>> <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
>> John P Sheehy <JPS@no.komm>
>> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
>
>But it works just fine. But you're right, it's basically an enhanced
>version of greyscale. The only true B&W digital camera I know of was a
>version of the Kodak DCS-1.


Leaf made several.


rafe b.
http://www.terrapinphoto.com
Anonymous
March 10, 2005 4:04:32 AM

Archived from groups: rec.photo.digital (More info?)

"rafe bustin" <rafeb@speakeasy.net> wrote in message
news:27cv21ldm2qnd789oje20vau1n8vv8fl8d@4ax.com...
> On Wed, 9 Mar 2005 17:30:35 -0800, "Skip M" <shadowcatcher@cox.net>
> wrote:
>
>><JPS@no.komm> wrote in message
>>news:147v2119ud7uoaoi1aggumh907si14f7ku@4ax.com...
>>> In message <XvKXd.3347$4k2.2376@fe2.columbus.rr.com>,
>>> "Fitpix" <David@delawarestudioNOSPAM.com> wrote:
>>>
>>>>"paul" <paul@not.net> wrote in message
>>>>news:6NGdnayrpoGSgrLfRVn-2w@speakeasy.net...
>>>>>I recall reading that if the sensors were dedicated to B&W, it would be
>>>>>possible to get really great quality B&W images, just like film,
>>>>>similar
>>>>>physics. It seems not too hard to bracket three exposures with color
>>>>>filters for color if you wanted too. This would eliminate the averaging
>>>>>every third pixel bayer problems & produce considerably sharper images.
>>>>>Does any such beast exist? I'm just curious.
>>>
>>>>My 20D has a BW mode and you can select what color filter effect you
>>>>want....
>>>
>>> It's just making a B&W image from multi-color-filtered data. It's not
>>> true B&W.
>>> --
>>>
>>> <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
>>> John P Sheehy <JPS@no.komm>
>>> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
>>
>>But it works just fine. But you're right, it's basically an enhanced
>>version of greyscale. The only true B&W digital camera I know of was a
>>version of the Kodak DCS-1.
>
>
> Leaf made several.
>
>
> rafe b.
> http://www.terrapinphoto.com

Leaf made backs, not cameras, as far as I know.
BTW, Kodak also made an infrared version of the same camera, I think.

--
Skip Middleton
http://www.shadowcatcherimagery.com
Anonymous
March 10, 2005 4:04:33 AM

Archived from groups: rec.photo.digital (More info?)

On Wed, 9 Mar 2005 20:48:10 -0800, "Skip M" <shadowcatcher@cox.net>
wrote:


>Leaf made backs, not cameras, as far as I know.
>BTW, Kodak also made an infrared version of the same camera, I think.


I can't find links to it but I remember a small,
self-contained camera... I think it was a Leaf.
This at least five years ago. The resolution
wasn't great, around 1 or 2 Mpixels. For color
work it used an external filter wheel. Obviously
for "studio" work only. I remember it because it
cost a lot more than some of the "consumer"
digicams available at that time. It did have
interchangeable lenses -- I think it used the
Nikon F mount, but my memory is very hazy.

I might be wrong about the vendor -- it might
have been PhaseOne or maybe BetterLight.

No mater.. it's clearly faded into oblivion, or
maybe just a product of my imagination.



rafe b.
http://www.terrapinphoto.com
Anonymous
March 10, 2005 4:54:27 AM

Archived from groups: rec.photo.digital (More info?)

In message <t49v21p6h39iv1vgvuvaa7t2s043obaldj@4ax.com>,
rafe bustin <rafeb@speakeasy.net> wrote:

>But there are lots of advantages to shooting BW
>with a Bayer CCD. For example, you can simulate
>virtually any filter you might imagine. You can
>generate your BW image by varying the relative
>contributions of the red/green/blue channels.
>Really lots of possibilities.

But these colors do not exist at the full resolution of the sensor;
green alone gives you 71%, red and blue, 50%.

A channel mixer that separated high- and low-frequency components of
each channel, giving 6 contribution bands, would be good. That way, you
could have the overall tonality of the red channel, but
near-full-resolution for pixel-to-pixel detail.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
March 10, 2005 4:59:04 AM

Archived from groups: rec.photo.digital (More info?)

rafe bustin wrote:

>
> What's really missing are dedicated BW output devices.
>
> For now, the best thing for that is an Epson inkjet
> with a decent RIP, or Piezography, or maybe the
> Harrington quadtone RIP. (ImagePrint and StudioPrint
> are RIPs I hear mentioned often for dedicated BW
> printing.)

Kodak is making "old skool" silver B&W film, developed in like dektol to be
used in light jet type applications..



--

Stacey
Anonymous
March 10, 2005 5:25:53 AM

Archived from groups: rec.photo.digital (More info?)

rafe bustin <rafeb@speakeasy.net> wrote:
>
> But there are lots of advantages to shooting BW
> with a Bayer CCD. For example, you can simulate
> virtually any filter you might imagine.

You can do a lot of different things, but it is no where
near any filter I might imagine. Look through the Wratten
Light Filters book. Each of the filters were made for a reason.
If you need a particular spectral response, then mixing primary
colours isn't going to get you there.

> You can
> generate your BW image by varying the relative
> contributions of the red/green/blue channels.
> Really lots of possibilities.

Yes, there are lots of possibilities and many of the
pictorial effects you may be after can be done that
way, but except for variations on the basic set of
tricolour filters and minus-tricolour filters, it won't
do quite the same things.

Peter.
--
pirwin@ktb.net
Anonymous
March 10, 2005 5:25:54 AM

Archived from groups: rec.photo.digital (More info?)

On Thu, 10 Mar 2005 02:25:53 +0000 (UTC), Peter Irwin <pirwin@ktb.net>
wrote:

>rafe bustin <rafeb@speakeasy.net> wrote:
>>
>> But there are lots of advantages to shooting BW
>> with a Bayer CCD. For example, you can simulate
>> virtually any filter you might imagine.
>
>You can do a lot of different things, but it is no where
>near any filter I might imagine. Look through the Wratten
>Light Filters book. Each of the filters were made for a reason.
>If you need a particular spectral response, then mixing primary
>colours isn't going to get you there.
>
>> You can
>> generate your BW image by varying the relative
>> contributions of the red/green/blue channels.
>> Really lots of possibilities.
>
>Yes, there are lots of possibilities and many of the
>pictorial effects you may be after can be done that
>way, but except for variations on the basic set of
>tricolour filters and minus-tricolour filters, it won't
>do quite the same things.


I'm not a guru on shooting BW through filters, but
I know that chapters have been written on how to
get the best BW images from an RGB "original".

I haven't shot BW film since I began processing
my images digitally, six or seven years back.

Want contrast? Take more from the red channel.
Want detail? Take more from the green. And so on.

I often look at the individual channels of my RGB
images to see what sorts of "BW" images might be
lurking there.


rafe b.
http://www.terrapinphoto.com
Anonymous
March 10, 2005 8:42:03 AM

Archived from groups: rec.photo.digital (More info?)

Peter Irwin <pirwin@ktb.net> writes:
>rafe bustin <rafeb@speakeasy.net> wrote:
>>
>> But there are lots of advantages to shooting BW
>> with a Bayer CCD. For example, you can simulate
>> virtually any filter you might imagine.
>
>You can do a lot of different things, but it is no where
>near any filter I might imagine. Look through the Wratten
>Light Filters book. Each of the filters were made for a reason.
>If you need a particular spectral response, then mixing primary
>colours isn't going to get you there.

Yes. When you combine data from an RGB channel using the channel mixer
(or equivalent), effective filter that the image was effectively "shot"
through can be calculated by taking the spectral response of the three
Bayer filters and adding them in the same ratio.

But the *only* filter responses you can get are the linear combinations
of these three filters. You can't get anything with a steep slope (deep
red, or perhaps orange).

So you can simulate the effect of colour balancing or correction
filters, and maybe some 3-colour separation filters, but not most of the
special ones oriented to B&W shooting.

Dave
March 10, 2005 1:17:56 PM

Archived from groups: rec.photo.digital (More info?)

Dave Martindale wrote:
>
> Yes. When you combine data from an RGB channel using the channel mixer
> (or equivalent), effective filter that the image was effectively "shot"
> through can be calculated by taking the spectral response of the three
> Bayer filters and adding them in the same ratio.
>
> But the *only* filter responses you can get are the linear combinations
> of these three filters. You can't get anything with a steep slope (deep
> red, or perhaps orange).

What if you apply levels and/or curves to the channels before mixing.
That would change the slope, would it not?

There are PS plugins that claim to simulate such filters (and different
films, too) but they cost more than I've wanted to spend.

Don't forget you can also use the channel mixer in CMYK mode, too, which
can be handy as the sky is more often cyan than blue.

In fact, although I haven't done it, one could copy the image and make
two versions, RGB, and CMYK, and channel mix both of them and then
combine the images.

Bob
Anonymous
March 10, 2005 2:14:00 PM

Archived from groups: rec.photo.digital (More info?)

On Wed, 09 Mar 2005 08:06:38 -0800, paul <paul@not.net> wrote:

>I recall reading that if the sensors were dedicated to B&W, it would be
>possible to get really great quality B&W images, just like film, similar
>physics. It seems not too hard to bracket three exposures with color
>filters for color if you wanted too. This would eliminate the averaging
>every third pixel bayer problems & produce considerably sharper images.
>Does any such beast exist? I'm just curious.

An idea occurred to me a while back when people were discussing digital
B&W, but I never got around to posting. It's very much "off the top of my
head" (I've not looked into the maths involved), so could be a load of
complete tosh, but:

A conventional (bayer) digital camera demosaics the RGGB data to recreate
RGB triples. For B&W images, you would then use Photoshop to manipulate
this as required.

Would it be possible to use a different raw conversion algorithm that goes
more directly to intensity, without going through RGB first? Has this been
tried? Was it any good?

Just thinking aloud.


Regards,
Graham Holden (g-holden AT dircon DOT co DOT uk)
--
There are 10 types of people in the world;
those that understand binary and those that don't.
March 10, 2005 2:14:01 PM

Archived from groups: rec.photo.digital (More info?)

In article <9ga031ls3pnpj6q8cjkuo06fsq15ic4b31@4ax.com>, look@bottom.of.post
says...
> On Wed, 09 Mar 2005 08:06:38 -0800, paul <paul@not.net> wrote:
>
> >I recall reading that if the sensors were dedicated to B&W, it would be
> >possible to get really great quality B&W images, just like film, similar
> >physics. It seems not too hard to bracket three exposures with color
> >filters for color if you wanted too. This would eliminate the averaging
> >every third pixel bayer problems & produce considerably sharper images.
> >Does any such beast exist? I'm just curious.
>
> An idea occurred to me a while back when people were discussing digital
> B&W, but I never got around to posting. It's very much "off the top of my
> head" (I've not looked into the maths involved), so could be a load of
> complete tosh, but:
>
> A conventional (bayer) digital camera demosaics the RGGB data to recreate
> RGB triples. For B&W images, you would then use Photoshop to manipulate
> this as required.
>
> Would it be possible to use a different raw conversion algorithm that goes
> more directly to intensity, without going through RGB first? Has this been
> tried? Was it any good?
>
> Just thinking aloud.
>
>
> Regards,
> Graham Holden (g-holden AT dircon DOT co DOT uk)
> --
> There are 10 types of people in the world;
> those that understand binary and those that don't.
>

Just to jump in with something that I haven't seen mentioned.

You really cant judge the quality of a B&W photo on the computer screen..

Most (if not all) CRTs and LCD monitors add SOME color cast of their own to a
B&W photo.

We have grown, over the years, quite used to viewing B&W on color CRTs but
the best overall way to judge the quality of B&W is either on a very high
resolution B&W screen or on paper.

The very best color monitors are going to flavor the picture with something,
no matter haw hard you try to make it B&W.


--
Larry Lynch
Mystic, Ct.
Anonymous
March 10, 2005 2:14:02 PM

Archived from groups: rec.photo.digital (More info?)

Larry wrote:
>
> Just to jump in with something that I haven't seen mentioned.
>
> You really cant judge the quality of a B&W photo on the computer screen..
>
> Most (if not all) CRTs and LCD monitors add SOME color cast of their own to a
> B&W photo.
>
> We have grown, over the years, quite used to viewing B&W on color CRTs but
> the best overall way to judge the quality of B&W is either on a very high
> resolution B&W screen or on paper.
>
> The very best color monitors are going to flavor the picture with something,
> no matter haw hard you try to make it B&W.
>
>
In fact, even black inks may have a slight color too. This is similar
to black and white photo print paper. Many brands were known for, and
advertised as having either a warm or a cool appearance. Warmer papers
had a very slight reddish cast, cooler ones less, or even a slight blue.
It is amazing how sensitive the eye is to color, especially to whites,
greys, or blacks. Ever try to mix paints to get the proper "shade" of
white?

I have a fairly new Canon printer. I had always used HP before, and one
Lexmark. The Canon black appears cooler to me than either the HP or
Lexmark blacks.
Anonymous
March 10, 2005 3:15:23 PM

Archived from groups: rec.photo.digital (More info?)

paul <paul@not.net> wrote:

> I recall reading that if the sensors were dedicated to B&W, it would
> be possible to get really great quality B&W images, just like film,
> similar physics. It seems not too hard to bracket three exposures
> with color filters for color if you wanted too. This would eliminate
> the averaging every third pixel bayer problems & produce
> considerably sharper images.

And much more accurate colour, because true colorimetric RGB filters
can be used.

See http://www.research.ibm.com/image_apps/tdisp.htm.

Andrew.
Anonymous
March 10, 2005 3:27:52 PM

Archived from groups: rec.photo.digital (More info?)

Peter Chant wrote:

>paul wrote:
>
>> I recall reading that if the sensors were dedicated to B&W, it would be
>> possible to get really great quality B&W images, just like film, similar
>> physics. It seems not too hard to bracket three exposures with color
>> filters for color if you wanted too. This would eliminate the averaging
>> every third pixel bayer problems & produce considerably sharper images.
>> Does any such beast exist? I'm just curious.
>
>I read a magazine article that refered to a B&W only digital camera,
>possibly a Kodak that was made several years back.

http://www.luminous-landscape.com/reviews/cameras/kodak...

--
Szymon 'Zbooy' Madej -- ysmadej[ma³pka]cyf-kr.edu.pl
OZSR => http://www.cyfronet.krakow.pl/rowery/
Szersze spojrzenie => http://pepperoni.fm.pl/zbooy/panoramy/
Stara Górska Biblioteczka => http://pepperoni.fm.pl/zbooy/gory/biblio/
March 10, 2005 3:27:53 PM

Archived from groups: rec.photo.digital (More info?)

Szymon 'Zbooy' Madej wrote:
> Peter Chant wrote:
>>
>>I read a magazine article that refered to a B&W only digital camera,
>>possibly a Kodak that was made several years back.
>
>
> http://www.luminous-landscape.com/reviews/cameras/kodak...

Quotes from that:

"We immediately ordered a 60" print of one of my first test images. The
Lightjet print came back breath taking. But it was obvious that optics
would be a prime consideration in my continued work. I needed to find
the right glass to take advantage of the camera's resolution. Without an
anti aliasing filter and no Bayer color matrix, the resolution of a 6
mega pixel monochrome camera is astonishing. In monochrome, 6 mega
pixels effectively does what it takes 12-24 mega pixels with a color
matrix."


"I ended up in shock at watching exposure times go from 1/60 or 1/125 of
a second with my Leica M6 and film, to 1/800, 1/1200 and even 1/1600 of
a second for the same aperture with the DCS 760m. With a base ISO of 400
exposures times are brisk – another advantage of a digital monochrome
over a color based sensor."

"So why am I crying in my soup at this point?

Along the way, a serious problem came up with the images from my 760m.
There is a form of "banding' horizontal to the frame as the data comes
out of camera. The banding shows up under certain lighting conditions in
clear skies. The banding issue is generated in the camera hardware and
not in the software. In short, the problem cannot be fixed."

"The interest in DCS monochromes has fallen on deaf ears with Kodak
management. A full frame CMOS monochrome camera was prototyped, but was
never put into production. My pleas with Kodak for a forward path in
digital monochrome photography have resulted in unreturned faxes and
emails. The door seems to be closed. At management level, my work does
not seem to be welcome or viewed as an asset in moving DCS monochrome
forward."
Anonymous
March 10, 2005 6:45:05 PM

Archived from groups: rec.photo.digital (More info?)

bob <not@not.not> wrote:
> Dave Martindale wrote:
>>
>> Yes. When you combine data from an RGB channel using the channel mixer
>> (or equivalent), effective filter that the image was effectively "shot"
>> through can be calculated by taking the spectral response of the three
>> Bayer filters and adding them in the same ratio.
>>
>> But the *only* filter responses you can get are the linear combinations
>> of these three filters. You can't get anything with a steep slope (deep
>> red, or perhaps orange).

> What if you apply levels and/or curves to the channels before mixing.
> That would change the slope, would it not?

No, because the sensor records a whole range of different wavelengths
of light and calls them all the same colour.

For example, a digital camera might not be able to distinguish between
a strong orange at 600nm and a red+yellow pair at 570nm and 650nm.
These different mixtures would give the same RGB values. However, a
filter that cuts red will render them very differently.

Andrew.
March 10, 2005 6:45:06 PM

Archived from groups: rec.photo.digital (More info?)

andrew29@littlepinkcloud.invalid wrote:

> For example, a digital camera might not be able to distinguish between
> a strong orange at 600nm and a red+yellow pair at 570nm and 650nm.
> These different mixtures would give the same RGB values. However, a
> filter that cuts red will render them very differently.

That's interesting. Can you give any real life examples -- I've got a
set of B&W filters and I'd like to experiment to see the difference.

Bob
Anonymous
March 10, 2005 7:08:42 PM

Archived from groups: rec.photo.digital (More info?)

JPS@no.komm wrote:
>
> It's just making a B&W image from multi-color-filtered data. It's not
> true B&W.

A b/w film registers a b/w image from a multi-colored original - the
scene in front of the camera.

A color negative printed on non-color-sensitive paper yields a b/w image
from a colored original.

A b/w image derived from a color digital image is still a b/w image, as
are the first two.

Quality, however, is a moot point, and most b/w afficionados will
declare that b/w film produces a superior b/w image compared to a
desaturated color image, whether film or digital.

This, though doesn't invalidate a b/w image from a color digital - it
is, obviously, a b/w image. It may not be as sharp, as well-toned, as
contrasty, whatever, as a print from a b/w negative - but it *is* a b/w
image.

However, a good color digital image, processed properly in Photoshop or
equivalent - not just desaturated - can produce a pretty good b/w image,
all the same.

We have to try to get prejudice out of these comparisons, and look at
them dispassionately.

Colin
Anonymous
March 11, 2005 1:26:15 AM

Archived from groups: rec.photo.digital (More info?)

In message <JcGdnXiVQtQh-a3fRVn-gQ@speakeasy.net>,
paul <paul@not.net> wrote:

>"So why am I crying in my soup at this point?

>Along the way, a serious problem came up with the images from my 760m.
>There is a form of “banding’ horizontal to the frame as the data comes
>out of camera. The banding shows up under certain lighting conditions in
>clear skies. The banding issue is generated in the camera hardware and
>not in the software. In short, the problem cannot be fixed."

Does the DCS 760m have masked pixels on the sensor?

If it does, the banding may be fixable in software, as long as it is
consistent along a horizontal line (the result of either an inconsistent
scaling or offset from line to line).
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
March 11, 2005 1:40:21 AM

Archived from groups: rec.photo.digital (More info?)

Peter Irwin wrote:
>
> rafe bustin <rafeb@speakeasy.net> wrote:
> >
> > But there are lots of advantages to shooting BW
> > with a Bayer CCD. For example, you can simulate
> > virtually any filter you might imagine.
>
> You can do a lot of different things, but it is no where
> near any filter I might imagine. Look through the Wratten
> Light Filters book. Each of the filters were made for a reason.
> If you need a particular spectral response, then mixing primary
> colours isn't going to get you there.
>
> > You can
> > generate your BW image by varying the relative
> > contributions of the red/green/blue channels.
> > Really lots of possibilities.
>
> Yes, there are lots of possibilities and many of the
> pictorial effects you may be after can be done that
> way, but except for variations on the basic set of
> tricolour filters and minus-tricolour filters, it won't
> do quite the same things.
>
Fine-tuning a digital b/w image from color isn't all-or-nothing as with
tricolor filters. The effect of most color filters can be simulated by
controlling the degree of desaturation or enhancement of the three
primary layers, with the possible exception of narrow-cut filters having
a spectral response curve sharper than the corresponding Bayer filter.

But, those filters are specialized, and not generally used for pictorial
b/w photography. Simulating an ordinary green or yellow filter for
example, is easily done as outlined above.

A big advantage is that you can do this after the fact of the shoot,
whereas a filter on the camera is a done deal even before you process
the film.

Contrary to popular belief, converting to b/w doesn't lose a lot in the
way of definition or sharpness.

Colin.
Anonymous
March 11, 2005 1:40:53 AM

Archived from groups: rec.photo.digital (More info?)

In message <9ga031ls3pnpj6q8cjkuo06fsq15ic4b31@4ax.com>,
Graham Holden <look@bottom.of.post> wrote:

>Would it be possible to use a different raw conversion algorithm that goes
>more directly to intensity, without going through RGB first? Has this been
>tried? Was it any good?

The RAW intensities are checkered in a CFA capture and you can't make
the checker pattern disappear just by scaling the channels, because what
makes the checkering disappear on a blue subject may make the checkering
more intense on a red subject. What you *can* do, as an alternative to
demoasaicing first, is to interpolate each channel independently to get
the full bitmap, and then mathematically manipulate these, but if you go
for only the red or blue channel, you will have 1/4 the pixels, and half
the resolution.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
March 11, 2005 2:26:22 PM

Archived from groups: rec.photo.digital (More info?)

bob <not@not.not> wrote:
> andrew29@littlepinkcloud.invalid wrote:

>> For example, a digital camera might not be able to distinguish between
>> a strong orange at 600nm and a red+yellow pair at 570nm and 650nm.
>> These different mixtures would give the same RGB values. However, a
>> filter that cuts red will render them very differently.

> That's interesting. Can you give any real life examples -- I've got a
> set of B&W filters and I'd like to experiment to see the difference.

Hmm, that's difficult. The easiest real world example might be RHEM
strips, which have stripes that are invisible in pure daylight but
appear when viewed with any other colour balance. These are only
about a buck apiece, but come in boxes of 50.

Andrew.
March 11, 2005 2:26:23 PM

Archived from groups: rec.photo.digital (More info?)

andrew29@littlepinkcloud.invalid wrote:
>
> Hmm, that's difficult. The easiest real world example might be RHEM
> strips, which have stripes that are invisible in pure daylight but
> appear when viewed with any other colour balance. These are only
> about a buck apiece, but come in boxes of 50.
>

I guess I'll just experiment then. Maybe flowers -- I know some of them
have some interesting properties wrt wavelengths of light to attract
bugs better.

Bob
Anonymous
March 11, 2005 9:32:53 PM

Archived from groups: rec.photo.digital (More info?)

bob <not@not.not> writes:
>andrew29@littlepinkcloud.invalid wrote:
>
>> For example, a digital camera might not be able to distinguish between
>> a strong orange at 600nm and a red+yellow pair at 570nm and 650nm.
>> These different mixtures would give the same RGB values. However, a
>> filter that cuts red will render them very differently.
>
>That's interesting. Can you give any real life examples -- I've got a
>set of B&W filters and I'd like to experiment to see the difference.

Maybe a deep red filter. Some of them cut out all but the longest red
wavelengths. In comparison, the red filter in a Bayer sensor needs to
pass wavelengths from deep red up to about pure yellow (so there aren't
gaps between red and green).

Or an orange filter. Adding red and green RGB channels has to be
roughly equivalent to a yellow filter, which actually passes 2/3 of the
spectrum. An orange filter will pass red and some yellow, but not much
of green - maybe half of the visible spectrum or less.

Basically, even if the Bayer filters provide sharp cutoffs (which are
not optimum for colour reproduction), there are only 6 possible cutoff
patterns you can get by selecting RGB channels: red only, green only,
blue only, cyan, magenta, and yellow. All of these choices pass either
about 1/3 of the visible spectrum, or 2/3 of it. The transition from
red to green and green to blue always occurs at a fixed place. While
coloured filters give you much more choice in exactly where the cutoff
happens, and how steep it is.

How many different red filters are there in a Wratten catalog? The
Bayer filter plus channel mixing gives you just one. You can adjust
the effective height of the transmittance curve, but not change its
shape.

If you *really* wanted to be able to simulate the effect of most filters
in software, you'd need to capture the original scene with a series of
narrowband filters. You'd like perhaps 31 colour channels spaced 10 nm
apart, or better yet 61 channels spaced 5 nm apart. Then you could
simulate almost any filter (except ones with very deep very narrow
notches). But I don't know of any cameras that take such an image all
at once - either the whole image is sampled at once at one wavelength at
a time, or you get all wavelengths at once but only for one pixel or for
one scanline. Plus you've got to store 30 or 60 measurements per pixel
instead of 3!

Dave
March 11, 2005 9:32:54 PM

Archived from groups: rec.photo.digital (More info?)

Dave Martindale wrote:
> bob <not@not.not> writes:
>
>>andrew29@littlepinkcloud.invalid wrote:
>>
>>
>>>For example, a digital camera might not be able to distinguish between
>>>a strong orange at 600nm and a red+yellow pair at 570nm and 650nm.
>>>These different mixtures would give the same RGB values. However, a
>>>filter that cuts red will render them very differently.
>>
>>That's interesting. Can you give any real life examples -- I've got a
>>set of B&W filters and I'd like to experiment to see the difference.
>
>
> Maybe a deep red filter. Some of them cut out all but the longest red
> wavelengths. In comparison, the red filter in a Bayer sensor needs to
> pass wavelengths from deep red up to about pure yellow (so there aren't
> gaps between red and green).

I was actually looking for suggestions of subject matter. Things I can
photograph through my red, orange, and yellow filters to see the
difference for myself that will look different than anything I can do in
software.

Bob
Anonymous
March 14, 2005 5:39:43 AM

Archived from groups: rec.photo.digital (More info?)

rafe bustin <rafeb@speakeasy.net> wrote:

>I'm not a guru on shooting BW through filters, but
>I know that chapters have been written on how to
>get the best BW images from an RGB "original".

I have no doubt that you can get pretty good
B&W by mixing RGB. Normally you want the green
channel to dominate, the red to be significant
and very little from the blue. The green channel
normally contains a bit from the blue part of the
spectrum so that blues may look about right even
with none of the blue channel mixed in.

>I haven't shot BW film since I began processing
>my images digitally, six or seven years back.

I find the darkroom to half the fun, and I have
no intention of giving it up.

>Want contrast? Take more from the red channel.
>Want detail? Take more from the green. And so on.

>I often look at the individual channels of my RGB
>images to see what sorts of "BW" images might be
>lurking there.


I did an interesting experiment. I made a CYMK
inkjet copy of the Ilford Test Chart from my
1958 edition Ilford Manual. The original test
chart uses nine different colour inks plus black,
and is intended for testing the response of film,
filter and light combinations for colour rendering
in black and white.

My CMYK copy of the chart looks almost identical
to the original under white light with no filter,
but when I looked at the two charts through an
orange filter they looked quite different.
On the original chart, the orange filter lightens
the orange patch more than the orange-red which
in turn is lightened more than the red so that
the difference between orange, orange-red and red
is exaggerated. When I look through the filter at
the CMYK copy of the chart, the red is lightened more
than the orange so that the orange, orange-red and
red patches look almost alike.

Peter.
--
pirwin@ktb.net
!