Sign in with
Sign up | Sign in
Your question

How many pixels make a "pixel?"

Last response: in Digital Camera
Share
Anonymous
June 23, 2005 5:43:25 PM

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

I'm too lazy to count all 8 million
to see if they are there, but is
it true that the pixels you see in
a digicam image are made up of four
different coloured pixels? In other words,
is the pixel count in an 8 meg image actually
2 meg of visible pixels?

More about : pixels make pixel

Anonymous
June 23, 2005 7:11:03 PM

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

RichA <none@none.com> wrote in news:72tlb15tfef8ujufrqpnddkq1ud1f9e45j@
4ax.com:

> it true that the pixels you see in
> a digicam image are made up of four
> different coloured pixels? In other words,
> is the pixel count in an 8 meg image actually
> 2 meg of visible pixels?

An 8 megapixel camera image is (usually) created from 4 million green
sensors, 2 million red sensors, and 2 million blue sensors arranged in a
GRGB pattern. Then the camera (or RAW software) interpolates the missing
RGB values.

Foveon sensors actually have RG and B at every sensor site. So then to keep
up with marketing numbers they refer to a 2 million site ccd as having 6
megapixels.
Anonymous
June 23, 2005 10:43:50 PM

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

1

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Related resources
Anonymous
June 23, 2005 11:12:43 PM

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

RichA wrote:
> I'm too lazy to count all 8 million
> to see if they are there, but is
> it true that the pixels you see in
> a digicam image are made up of four
> different coloured pixels? In other words,
> is the pixel count in an 8 meg image actually
> 2 meg of visible pixels?

Away with laziness! Read:

http://www.photozone.de/3Technology/digital_3.htm
Anonymous
June 23, 2005 11:12:44 PM

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

On Thu, 23 Jun 2005 19:12:43 GMT, "David J Taylor"
<david-taylor@blueyonder.co.not-this-bit.nor-this-part.uk.invalid>
wrote:

>RichA wrote:
>> I'm too lazy to count all 8 million
>> to see if they are there, but is
>> it true that the pixels you see in
>> a digicam image are made up of four
>> different coloured pixels? In other words,
>> is the pixel count in an 8 meg image actually
>> 2 meg of visible pixels?
>
>Away with laziness! Read:
>
> http://www.photozone.de/3Technology/digital_3.htm
>

Thanks, interesting reading. It does seem that the images
we see on our computer screens are in-fact made up of the
same number of square pixels as claimed by the camera to have,
an 8 meg is an 8 meg, etc.
-Rich
June 23, 2005 11:12:44 PM

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

On Thu, 23 Jun 2005 19:12:43 GMT, "David J Taylor"
<david-taylor@blueyonder.co.not-this-bit.nor-this-part.uk.invalid> wrote:

>RichA wrote:
>> I'm too lazy to count all 8 million
>> to see if they are there, but is
>> it true that the pixels you see in
>> a digicam image are made up of four
>> different coloured pixels? In other words,
>> is the pixel count in an 8 meg image actually
>> 2 meg of visible pixels?
>
>Away with laziness! Read:
>
> http://www.photozone.de/3Technology/digital_3.htm
>

Someone should tell the people who advertize Sony TVs... they still claim to
have 3 times more pixels then everyone else...
Anonymous
June 24, 2005 1:26:12 AM

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

On Thu, 23 Jun 2005 21:24:02 GMT, Mr. Mark <e.cartman@southpark.com> wrote:
>> http://www.photozone.de/3Technology/digital_3.htm
>
> GRGB...
>
> Why are the green sensors so over represented?

From what I've read, it seems that human vision is most sensitive in
the green spectrum.

For whatever reason, I find my B&W pictures look best when made
mostly from the red and blue channels. So I do occasionally wonder
whether I'd be better off with a more balanced representation of
channels. But of course what I really want is no Bayer filter ...

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 1:35:30 AM

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

> > Why are the green sensors so over represented?
>
> From what I've read, it seems that human vision is most sensitive in
> the green spectrum.

I didn't know that.

> For whatever reason, I find my B&W pictures look best when made
> mostly from the red and blue channels. So I do occasionally wonder
> whether I'd be better off with a more balanced representation of
> channels. But of course what I really want is no Bayer filter ...

I would REALLY like a true B&W mode in-camera.

I noticed that the Fuji sensor represented in the article was really nothing
more than the bayer filter tilted 45 degrees with funky shaped sensors. Who
do they think they're fooling? :) 

--
Mark

Photos, Ideas & Opinions
http://www.marklauter.com
Anonymous
June 24, 2005 1:35:31 AM

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

Mr. Mark wrote:
>>>Why are the green sensors so over represented?
>>
>>From what I've read, it seems that human vision is most sensitive in
>>the green spectrum.
>
>
> I didn't know that.
>
>
>>For whatever reason, I find my B&W pictures look best when made
>>mostly from the red and blue channels. So I do occasionally wonder
>>whether I'd be better off with a more balanced representation of
>>channels. But of course what I really want is no Bayer filter ...
>
>
> I would REALLY like a true B&W mode in-camera.


The Sigma DSLRs come close to that. Too much money for so few MP though.
Anonymous
June 24, 2005 1:39:15 AM

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

On Thu, 23 Jun 2005 21:35:30 GMT, Mr. Mark <e.cartman@southpark.com> wrote:
[Ben Rosengart wrote:]
[psst -- Mr. Mark -- please attribute quoted text. Thanks.]
>
>> For whatever reason, I find my B&W pictures look best when made
>> mostly from the red and blue channels. So I do occasionally wonder
>> whether I'd be better off with a more balanced representation of
>> channels. But of course what I really want is no Bayer filter ...
>
> I would REALLY like a true B&W mode in-camera.

Hear, hear. I'm seriously considering buying a more or less
disposable digital camera and trying to scrape the Bayer filter
off. There's a guy on photo.net who's done it and is willing to
share tips.

> I noticed that the Fuji sensor represented in the article was really nothing
> more than the bayer filter tilted 45 degrees with funky shaped sensors. Who
> do they think they're fooling? :) 

Heh. Well, it does have the extra small photosite for extended
dynamic range. Pretty cool idea. You can "expose to the right"
and not worry as much about clipping, I guess.

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 1:39:16 AM

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

Ben Rosengart wrote:
> On Thu, 23 Jun 2005 21:35:30 GMT, Mr. Mark <e.cartman@southpark.com> wrote:
> [Ben Rosengart wrote:]
> [psst -- Mr. Mark -- please attribute quoted text. Thanks.]
>
>>>For whatever reason, I find my B&W pictures look best when made
>>>mostly from the red and blue channels. So I do occasionally wonder
>>>whether I'd be better off with a more balanced representation of
>>>channels. But of course what I really want is no Bayer filter ...
>>
>>I would REALLY like a true B&W mode in-camera.
>
>
> Hear, hear. I'm seriously considering buying a more or less
> disposable digital camera and trying to scrape the Bayer filter
> off. There's a guy on photo.net who's done it and is willing to
> share tips.
>
Why? Oh, why?

--
John McWilliams
Anonymous
June 24, 2005 2:28:41 AM

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

On Thu, 23 Jun 2005 15:24:55 -0700, John McWilliams <jpmcw@comcast.net> wrote:
> Ben Rosengart wrote:
>
>> Hear, hear. I'm seriously considering buying a more or less
>> disposable digital camera and trying to scrape the Bayer filter
>> off. There's a guy on photo.net who's done it and is willing to
>> share tips.
>>
> Why? Oh, why?

The Bayer filter absorbs/scatters light. But I want that light to
go into the sensor! I bet it eats a stop and a half. Wouldn't you
like to have a stop and a half more light without bigger, heavier,
more expensive lenses?

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 2:28:42 AM

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

Ben Rosengart wrote:
> On Thu, 23 Jun 2005 15:24:55 -0700, John McWilliams <jpmcw@comcast.net> wrote:
>
>>Ben Rosengart wrote:
>>
>>
>>>Hear, hear. I'm seriously considering buying a more or less
>>>disposable digital camera and trying to scrape the Bayer filter
>>>off. There's a guy on photo.net who's done it and is willing to
>>>share tips.
>>>
>>
>>Why? Oh, why?
>
>
> The Bayer filter absorbs/scatters light. But I want that light to
> go into the sensor! I bet it eats a stop and a half. Wouldn't you
> like to have a stop and a half more light without bigger, heavier,
> more expensive lenses?
>
Yes, but of course. However, most of my photography is accomodated with
the 1600 ISO and 50mm 1.4 lens. And the drawback of not having a color
image, if I later prefer, is large.

As it is, I shoot everything in color, and have the luxury of making
very good conversions to B+W in software; gaining even two stops for 99%
of my stuff wouldn't be worth not having the option of color.

Are you sure you'd gain a full stop and a half?

--
John McWilliams
Anonymous
June 24, 2005 2:28:42 AM

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

Ben Rosengart wrote:

> The Bayer filter absorbs/scatters light. But I want that light to
> go into the sensor! I bet it eats a stop and a half. Wouldn't you
> like to have a stop and a half more light without bigger, heavier,
> more expensive lenses?

I doubt the filter eats more than 1/3 of a stop. The filters are clear.

In any case even if it did eat more, the ISO is calibrated with the A-A
filter in place so the glass makes no difference. It would only
contribute to noise (as more amplification would be required the more
the A-A eats light).

Cheers,
Alan



--
A beaver and a rabbit are sitting in the woods near Hoover Dam.
"Did you build that one?" the rabbit asked.
"No, but it was based on an idea of mine," the beaver replied.
-Jack Kilby on his invention of the IC.
Anonymous
June 24, 2005 2:58:14 AM

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

On Thu, 23 Jun 2005 15:39:51 -0700, John McWilliams <jpmcw@comcast.net> wrote:
> Ben Rosengart wrote:
>
>> The Bayer filter absorbs/scatters light. But I want that light to
>> go into the sensor! I bet it eats a stop and a half. Wouldn't you
>> like to have a stop and a half more light without bigger, heavier,
>> more expensive lenses?
>>
> Yes, but of course. However, most of my photography is accomodated with
> the 1600 ISO and 50mm 1.4 lens. And the drawback of not having a color
> image, if I later prefer, is large.

I've been shooting with an f/2.8 zoom, so I'm already two stops
behind you. I own a couple of f/1.8 primes (28 and 50), and that
does help ... but I like the flexibility of a zoom.

> As it is, I shoot everything in color, and have the luxury of making
> very good conversions to B+W in software; gaining even two stops for 99%
> of my stuff wouldn't be worth not having the option of color.

I respect that. It's not as much the case for me. In light that
low, I don't usually find that my pictures have very attractive color.
It's quite unusual for me to shoot a picture while "thinking in B&W"
and later decide that it's better in color.

> Are you sure you'd gain a full stop and a half?

Nope. Here's my reasoning. If the filters are truly complementary,
then on average, each one should pass 1/3 of the visible spectrum.
Assuming that light is evenly distributed in the spectrum -- which
is probably not true -- that means that 1/3 of the light is passed.

Under those questionable assumptions, the light level should triple
without a Bayer filter. A factor of 3 is a stop and a half; that
is, a doubling, and then halfway to the next doubling.

They say that the fastest way to discover an error is to post it to
Usenet and wait for the corrections to roll in.

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 3:10:28 AM

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

On Thu, 23 Jun 2005 19:02:40 -0400, Alan Browne
<alan.browne@FreeLunchVideotron.ca> wrote:
> Ben Rosengart wrote:
>
>> The Bayer filter absorbs/scatters light. But I want that light to
>> go into the sensor! I bet it eats a stop and a half. Wouldn't you
>> like to have a stop and a half more light without bigger, heavier,
>> more expensive lenses?
>
> I doubt the filter eats more than 1/3 of a stop. The filters are clear.
>
> In any case even if it did eat more, the ISO is calibrated with the A-A
> filter in place so the glass makes no difference. It would only
> contribute to noise (as more amplification would be required the more
> the A-A eats light).

Really Alan, if you can't be bothered to read the posts, why respond?

We're talking about Bayer filters, not antialiasing filters.

I don't think you're right about the ISO. On a regular digicam, the
metering is done off the CCD; so without a Bayer filter, the meter
will read the light level as higher. On a DSLR, the meter will be
off, so one would apply some exposure compensation. ISO doesn't
come into it at all.

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 3:10:29 AM

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

Ben Rosengart wrote:

> On Thu, 23 Jun 2005 19:02:40 -0400, Alan Browne
> <alan.browne@FreeLunchVideotron.ca> wrote:
>
>>Ben Rosengart wrote:
>>
>>
>>>The Bayer filter absorbs/scatters light. But I want that light to
>>>go into the sensor! I bet it eats a stop and a half. Wouldn't you
>>>like to have a stop and a half more light without bigger, heavier,
>>>more expensive lenses?
>>
>>I doubt the filter eats more than 1/3 of a stop. The filters are clear.
>>
>>In any case even if it did eat more, the ISO is calibrated with the A-A
>>filter in place so the glass makes no difference. It would only
>>contribute to noise (as more amplification would be required the more
>>the A-A eats light).
>
>
> Really Alan, if you can't be bothered to read the posts, why respond?

er, It's never stopped me before (?).

>
> We're talking about Bayer filters, not antialiasing filters.

Oops!

Well regardless, at some point there has to be filtering. Whether split
into three sensors (a la video CCD) or a Foveon, you can't escape
filtering. So it's a moot point unless monochromatic is your game.


> I don't think you're right about the ISO. On a regular digicam, the
> metering is done off the CCD; so without a Bayer filter, the meter
> will read the light level as higher. On a DSLR, the meter will be
> off, so one would apply some exposure compensation. ISO doesn't
> come into it at all.

Sure. my reply was in the prev. context, not that of a modified sensor.

If you do "scrape off" the bayer filter (unlikely), then it should be
very easy to calibrate an offset ... as long as you measure a target
before the scraping.

Cheers,
Alan.



--
A beaver and a rabbit are sitting in the woods near Hoover Dam.
"Did you build that one?" the rabbit asked.
"No, but it was based on an idea of mine," the beaver replied.
-Jack Kilby on his invention of the IC.
Anonymous
June 24, 2005 7:25:46 AM

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

On Thu, 23 Jun 2005 19:37:11 -0400, Alan Browne
<alan.browne@FreeLunchVideotron.ca> wrote:
> Ben Rosengart wrote:
>
>> Really Alan, if you can't be bothered to read the posts, why respond?
>
> er, It's never stopped me before (?).

You could at least have the decency to pretend to be embarrassed.

> So it's a moot point unless monochromatic is your game.

Uh, but it is, mostly. See above about reading the posts.

> If you do "scrape off" the bayer filter (unlikely),

I have instructions from a guy who's done it.

> then it should be very easy to calibrate an offset ... as long as you
> measure a target before the scraping.

With a point and shoot, you don't have to. At least, I think on
a P&S, metering is done on the CCD. Could be wrong. With a DSLR
-- if you could bring yourself to attempt such a crude modification
of such an expensive device -- you can just see how far you need
to turn down the exposure compensation to avoid blowing out
highlights.

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 10:23:31 AM

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

Mr. Mark wrote:
>> http://www.photozone.de/3Technology/digital_3.htm
>
> GRGB...
>
> Why are the green sensors so over represented?

Because the majority of the luminance signal comes from green, so more
green sensors means better sensitivity and more accurate luminance.

David
Anonymous
June 24, 2005 10:28:19 AM

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

Mr. Mark wrote:
[]
> I noticed that the Fuji sensor represented in the article was really
> nothing more than the bayer filter tilted 45 degrees with funky
> shaped sensors. Who do they think they're fooling? :) 

It's not that simple. Because the array is at 45 degrees, but all picture
representations - in JPEG or BMP files for example - are at 0 degrees, the
sensor output must be interpolated before use (and thus results in more
image pixels). However, the aliasing pattern for the Fuji sensor will be
more visible against 45 degree diagonal lines (rather than near-horizontal
or near-vertical lines), and may be less visible in the resulting image.

How do the images compare in practice?

David
Anonymous
June 24, 2005 11:45:01 AM

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

"Mr. Mark" <e.cartman@southpark.com> wrote in message
news:SLFue.109186$VH2.30360@tornado.tampabay.rr.com...
> > http://www.photozone.de/3Technology/digital_3.htm
>
> GRGB...
>
> Why are the green sensors so over represented?
>
They're not over-represented, but *proportionally* represented: The eye's
maximum sensitivity is in the yellow-green portion of the visible spectrum.
See http://www.amastro.org/at/ot/othcs.html for a graph.
Anonymous
June 24, 2005 11:48:21 AM

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

"Ben Rosengart" <br+rpdss@panix.com> wrote in message
news:slrndbmb03.gno.br@panix5.panix.com...
> On Thu, 23 Jun 2005 21:35:30 GMT, Mr. Mark <e.cartman@southpark.com>
wrote:
> [Ben Rosengart wrote:]
> [psst -- Mr. Mark -- please attribute quoted text. Thanks.]
> >
> >> For whatever reason, I find my B&W pictures look best when made
> >> mostly from the red and blue channels. So I do occasionally wonder
> >> whether I'd be better off with a more balanced representation of
> >> channels. But of course what I really want is no Bayer filter ...
> >
> > I would REALLY like a true B&W mode in-camera.
>
> Hear, hear. I'm seriously considering buying a more or less
> disposable digital camera and trying to scrape the Bayer filter
> off. There's a guy on photo.net who's done it and is willing to
> share tips.

If you scape the Bayer filters off, then all of the electrons will fall out.
Anonymous
June 24, 2005 12:02:40 PM

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

"Ben Rosengart" <br+rpdss@panix.com> wrote in message
news:slrndbmdsp.bvg.br@panix5.panix.com...
> On Thu, 23 Jun 2005 15:24:55 -0700, John McWilliams <jpmcw@comcast.net>
wrote:
> > Ben Rosengart wrote:
> >
> >> Hear, hear. I'm seriously considering buying a more or less
> >> disposable digital camera and trying to scrape the Bayer filter
> >> off. There's a guy on photo.net who's done it and is willing to
> >> share tips.
> >>
> > Why? Oh, why?
>
> The Bayer filter absorbs/scatters light. But I want that light to
> go into the sensor! I bet it eats a stop and a half. Wouldn't you
> like to have a stop and a half more light without bigger, heavier,
> more expensive lenses?


The real reason for removing the Bayer filter is to be able to see the
government mind-control rays which are only visible when a gray-scale sensor
is used. Not to worry, though: Many people have reported that wearing an
aluminum-foil hat completely blocks the rays so you don't have to destroy
your camera!

Seriously, here's the reason digital camera companies don't off b&w-only
models:

Number of color models sold: 10,000,000
Number of b&w-only models sold: 2
--------------------------------------------------------------------------

"Everything looks worse in black and white"-- Paul Simon, "Kodachrome"
Anonymous
June 24, 2005 2:07:59 PM

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

Ben Rosengart wrote:

> On Thu, 23 Jun 2005 19:37:11 -0400, Alan Browne
> <alan.browne@FreeLunchVideotron.ca> wrote:
>
>>Ben Rosengart wrote:
>>
>>
>>>Really Alan, if you can't be bothered to read the posts, why respond?
>>
>>er, It's never stopped me before (?).
>
>
> You could at least have the decency to pretend to be embarrassed.

Hardly. Mistakes are made. I apologize. Last evening was the tail end
of a couple very long days. I really should not have been on usenet at
all.

>
>
>>So it's a moot point unless monochromatic is your game.
>
>
> Uh, but it is, mostly. See above about reading the posts.
>
>
>>If you do "scrape off" the bayer filter (unlikely),
>
>
> I have instructions from a guy who's done it.
>
>
>>then it should be very easy to calibrate an offset ... as long as you
>>measure a target before the scraping.
>
>
> With a point and shoot, you don't have to. At least, I think on
> a P&S, metering is done on the CCD. Could be wrong. With a DSLR

No, you're correct. Most P&S meter in the sensor. My comments were
along the vein of a DSLR where there is, in the exposure system
conmpensation for the differences in optical path to the exposure
measurement. For that matter, in a SLR, the exposure system has
different constants / weightings for different exposure modes. All of
these take into account losses at the sensor filtering stage.


> -- if you could bring yourself to attempt such a crude modification
> of such an expensive device -- you can just see how far you need
> to turn down the exposure compensation to avoid blowing out
> highlights.

It should be a direct offset in EV.

Cheers,
Alan


--
A beaver and a rabbit are sitting in the woods near Hoover Dam.
"Did you build that one?" the rabbit asked.
"No, but it was based on an idea of mine," the beaver replied.
-Jack Kilby on his invention of the IC.
Anonymous
June 24, 2005 2:19:55 PM

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

David J Taylor wrote:

> Mr. Mark wrote:
>
>>> http://www.photozone.de/3Technology/digital_3.htm
>>
>>GRGB...
>>
>>Why are the green sensors so over represented?
>
>
> Because the majority of the luminance signal comes from green, so more
> green sensors means better sensitivity and more accurate luminance.

This doesn't make sense to me. If most of the luminance comes from the
green, then you would need less sensors for green, not more.


--
A beaver and a rabbit are sitting in the woods near Hoover Dam.
"Did you build that one?" the rabbit asked.
"No, but it was based on an idea of mine," the beaver replied.
-Jack Kilby on his invention of the IC.
Anonymous
June 24, 2005 5:47:16 PM

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

On Fri, 24 Jun 2005 08:02:40 -0700, "Paul H."
<xxpaulhtck@zzcomcast.yycom> wrote:

>
>"Ben Rosengart" <br+rpdss@panix.com> wrote in message
>news:slrndbmdsp.bvg.br@panix5.panix.com...
>> On Thu, 23 Jun 2005 15:24:55 -0700, John McWilliams <jpmcw@comcast.net>
>wrote:
>> > Ben Rosengart wrote:
>> >
>> >> Hear, hear. I'm seriously considering buying a more or less
>> >> disposable digital camera and trying to scrape the Bayer filter
>> >> off. There's a guy on photo.net who's done it and is willing to
>> >> share tips.
>> >>
>> > Why? Oh, why?
>>
>> The Bayer filter absorbs/scatters light. But I want that light to
>> go into the sensor! I bet it eats a stop and a half. Wouldn't you
>> like to have a stop and a half more light without bigger, heavier,
>> more expensive lenses?
>
>
>The real reason for removing the Bayer filter is to be able to see the
>government mind-control rays which are only visible when a gray-scale sensor
>is used. Not to worry, though: Many people have reported that wearing an
>aluminum-foil hat completely blocks the rays so you don't have to destroy
>your camera!
>
>Seriously, here's the reason digital camera companies don't off b&w-only
>models:
>
>Number of color models sold: 10,000,000
>Number of b&w-only models sold: 2
>--------------------------------------------------------------------------
>
>"Everything looks worse in black and white"-- Paul Simon, "Kodachrome"
>

There are actually lots of black and white CCD cameras available,
along with development kits from mfgs you can assemble cameras out
of. But they are not like consumer cameras but can be used as such
if someone wanted to.
Anonymous
June 24, 2005 6:51:52 PM

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

On Fri, 24 Jun 2005 07:48:21 -0700, Paul H. <xxpaulhtck@zzcomcast.yycom> wrote:
>
> "Ben Rosengart" <br+rpdss@panix.com> wrote in message
> news:slrndbmb03.gno.br@panix5.panix.com...
>>
>> Hear, hear. I'm seriously considering buying a more or less
>> disposable digital camera and trying to scrape the Bayer filter
>> off. There's a guy on photo.net who's done it and is willing to
>> share tips.
>
> If you scape the Bayer filters off, then all of the electrons will fall out.

That's ok, the dog will get them.

--
Ben Rosengart (212) 741-4400 x215
Sometimes it only makes sense to focus our attention on those
questions that are equal parts trivial and intriguing.
--Josh Micah Marshall
Anonymous
June 24, 2005 6:51:53 PM

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

Ben Rosengart wrote:
> On Fri, 24 Jun 2005 07:48:21 -0700, Paul H.
> <xxpaulhtck@zzcomcast.yycom> wrote:
>>
>> "Ben Rosengart" <br+rpdss@panix.com> wrote in message
>> news:slrndbmb03.gno.br@panix5.panix.com...
>>>
>>> Hear, hear. I'm seriously considering buying a more or less
>>> disposable digital camera and trying to scrape the Bayer filter
>>> off. There's a guy on photo.net who's done it and is willing to
>>> share tips.
>>
>> If you scape the Bayer filters off, then all of the electrons will
>> fall out.
>
> That's ok, the dog will get them.

That's his job: barkuuming.

When the electrons fall out do the photons become "free-range"? Or
will the hold their noses and go, "Eeeeuuwww!" ?
Anonymous
June 24, 2005 7:36:20 PM

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

Alan Browne wrote:
> David J Taylor wrote:
>
>> Mr. Mark wrote:
>>
>>>> http://www.photozone.de/3Technology/digital_3.htm
>>>
>>> GRGB...
>>>
>>> Why are the green sensors so over represented?
>>
>>
>> Because the majority of the luminance signal comes from green, so
>> more green sensors means better sensitivity and more accurate
>> luminance.
>
> This doesn't make sense to me. If most of the luminance comes from
> the green, then you would need less sensors for green, not more.

I see what you mean - but John's explained it better.

What I meant to say is that the weighting used when deriving the luminance
signal uses more of green, and less of red and blue. This makes the
luminance component "look" approximately the same as normal panchromatic
black-and-white film - the spectral response is similar.

Given that the majority of the perceived sharpness comes from the
luminance component of the signal, and perhaps much of the visible noise
as well, it makes sense to try and collect as much green signal as
possible so that it has the best quality, and therefore the luminance from
the camera is of the best quality. You can filter the noise in the
chrominance signal in any case, as it doesn't need to be that sharp.

Much of this theory and practice come from colour TV, but I expect it
reads across reasonably well to digital cameras.

Cheers,
David
Anonymous
June 24, 2005 7:37:58 PM

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

On Fri, 24 Jun 2005 10:19:55 -0400, Alan Browne
<alan.browne@FreeLunchVideotron.ca> wrote:

>David J Taylor wrote:
>
>> Mr. Mark wrote:
>>
>>>> http://www.photozone.de/3Technology/digital_3.htm
>>>
>>>GRGB...
>>>
>>>Why are the green sensors so over represented?
>>
>>
>> Because the majority of the luminance signal comes from green, so more
>> green sensors means better sensitivity and more accurate luminance.
>
>This doesn't make sense to me. If most of the luminance comes from the
>green, then you would need less sensors for green, not more.

You got RGB - now pick which one you're going to duplicate
to make the array work. You want the greatest number of
photons to hit the sensor overall for best signal to noise
ratio.

Would you pick the red end of the spectrum, the blue end of
the spectrum, or the middle of the spectrum where most of
the luminosity (=more photons) is?

--
Regards

John Bean
June 25, 2005 12:48:18 AM

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

Alan Browne <alan.browne@FreeLunchVideotron.ca> wrote in news:D 9h9fd$9t0$1
@inews.gazeta.pl:

> Paul H. wrote:
>
>> "Everything looks worse in black and white"-- Paul Simon, "Kodachrome"
>
> http://gallery.photo.net/photo/2452591-lg.jpg
>
> Yeah. Right.

I don't see how that image proves anything; to me it is rather
uninteresting. Personally I would rather see it in colour. I guess it is
just a matter of personal preference.


--
Mark Heyes (New Zealand)
See my pics at www.gigatech.co.nz (last updated 3-May-05)
"There are 10 types of people, those that
understand binary and those that don't"
Anonymous
June 25, 2005 1:51:59 PM

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

In message <d9fh34$lmp$1@inews.gazeta.pl>,
Alan Browne <alan.browne@FreeLunchVideotron.ca> wrote:

>If you do "scrape off" the bayer filter (unlikely), then it should be
>very easy to calibrate an offset ... as long as you measure a target
>before the scraping.

The builtin light meter may already be color-filtered to match the
sensor/CFA combined response, and may vary more based on color after the
CFA is removed, even if you use an offset.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
June 25, 2005 2:10:24 PM

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

In message <Xns967E86677466sizernospamcom@216.40.28.85>,
Sizer <sizer@nospam.com> wrote:

>RichA <none@none.com> wrote in news:72tlb15tfef8ujufrqpnddkq1ud1f9e45j@
>4ax.com:
>
>> it true that the pixels you see in
>> a digicam image are made up of four
>> different coloured pixels? In other words,
>> is the pixel count in an 8 meg image actually
>> 2 meg of visible pixels?
>
>An 8 megapixel camera image is (usually) created from 4 million green
>sensors, 2 million red sensors, and 2 million blue sensors arranged in a
>GRGB pattern. Then the camera (or RAW software) interpolates the missing
>RGB values.

Mainly to create a bitmap in conventional form ... you really don't see
color at this increased, interpolated resolution, anyway, unless you're
standing right next to a billboard image.

>Foveon sensors actually have RG and B at every sensor site. So then to keep
>up with marketing numbers they refer to a 2 million site ccd as having 6
>megapixels.

That in and of itself is deceptive, as "megapixel" has been accepted to
mean 2D spatial locations, not 3D locations of sensors.

Besides, the Foveon color separation is pretty poor, and color has to be
interpolated between the three color samples for each pixel with a great
deal of posterization and deviation.
--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
Anonymous
June 25, 2005 6:21:46 PM

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

JPS@no.komm wrote:

> In message <d9fh34$lmp$1@inews.gazeta.pl>,
> Alan Browne <alan.browne@FreeLunchVideotron.ca> wrote:
>
>
>>If you do "scrape off" the bayer filter (unlikely), then it should be
>>very easy to calibrate an offset ... as long as you measure a target
>>before the scraping.
>
>
> The builtin light meter may already be color-filtered to match the
> sensor/CFA combined response, and may vary more based on color after the
> CFA is removed, even if you use an offset.

On those Nikons with the 3D color RGB matrix metering, yes, but on most
other cameras the metering is light intensity only.


--
A beaver and a rabbit are sitting in the woods near Hoover Dam.
"Did you build that one?" the rabbit asked.
"No, but it was based on an idea of mine," the beaver replied.
-Jack Kilby on his invention of the IC.
Anonymous
June 26, 2005 2:12:23 AM

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

"Alan Browne" <alan.browne@FreeLunchVideotron.ca> wrote in message
news:D 9h9fd$9t0$1@inews.gazeta.pl...
> Paul H. wrote:
>
> > "Everything looks worse in black and white"-- Paul Simon, "Kodachrome"
>
> http://gallery.photo.net/photo/2452591-lg.jpg
>
> Yeah. Right.

It was a trifle: a joke, a drive-by jape, not meant to be taken seriously.
Un-knot thy shorts, Winky.
!