If this is not false advertisement, what is?

Dominic,

Thanks I needed that!

That could be the next Mel Brookes movie???

The name could be something like "Indiana X3 and the Temple of Trolls"
 
Mega = one million. This has been true for the last four centuries.
Except when it refers to bytes of memory or, more recently, pixels in an image, which case it means 2^20. This has been true for at least five decades (i.e. since people "invented" bytes).

BTW, I still claim that the Foveon sensor has 3.4 mega pixels; the Bayer ones have (e.g.) 6.0 "one-third" pixels ;-)
 
Pixels do not equal sensors. A pixel is the smallest unit of a
final image, regardless of how it was made. My monitor has 2 Mega
pixels, not 6 (2M red + 2M green + 2M blue). The Foveon puts out
3.4 Mega pixels, period...
And the Fuji S2 puts out 12 MP, so why do some people insist on
calling it 6 MP?

j
And my 17x11 600 dpi print has 67.3 pixels, so that means I can tell people my camera takes 67 megapixel pictures????

Basing quality on output attributes is ludicrous, it's like saying my AM radio broadcast is super high fidelity because I have great speakers.
 
Pixels do not equal sensors. A pixel is the smallest unit of a
final image, regardless of how it was made. My monitor has 2 Mega
pixels, not 6 (2M red + 2M green + 2M blue). The Foveon puts out
3.4 Mega pixels, period, however the quality of those pixels is
twice as good as those from other DSLR's which makes the final
image look as good as a standard 7Mpixel image would.
This gets at the heart of the misunderstanding many "Bayer people" have. Your monitor does not have 2M red, 2M green and 2M blue. It has a lot more than that, because there are several phosphor dots for each pixel (at least at reasonable monitor resolutions). This is called "dot pitch".

The Bayer sensor is different: there is only one dot (sensor) at each pixel location which detects only one oif the colours. the other colours are not detected and have to be guessed from neighbouring pixels.
 
I doubt it would stop the whiners.
Some people are just not happy unless they have something to complain about.
To please the pixel counters perhaps Foveon should adopt a new
revolutionary
name for their revolutionary Sensor.

I suggest the following..

"Foveon 3-CMOS Sensor Technology"
"3.5 Million Pixels X3"

Would that stop the whiners?
 
Also are there really 5 mirrors in a Canon penta-mirror ?
My Dad always told me it was a five sided piece of glass. More like a prism silvered on some sides (i.e. it reflects on the inside, not the outside) than a mirror.

I have no idea if this is true, but there must be a zillion sites out there with explanations.
 
Also are there really 5 mirrors in a Canon penta-mirror ?
My Dad always told me it was a five sided piece of glass. More like
a prism silvered on some sides (i.e. it reflects on the inside, not
the outside) than a mirror.

I have no idea if this is true, but there must be a zillion sites
out there with explanations.
Sorry, I was thinking of a Pentax, of course.
 
It was my feeble attempt at "false advertising" concept. A penta prism has five sides. A penta mirror probably only has 3 mirrors and 2 blank sides. So why didn't they call it their new tri-mirror finder? Why label a 3 sided mirror system a "penta" mirror finder
 
Dominic,

Thanks I needed that!

That could be the next Mel Brookes movie???

The name could be something like "Indiana X3 and the Temple of Trolls"
Back when I worked for Cummins Electronics (a Michigan based operation of the Indiana based Cummins Engine) our top level manager was Ron Temple. Next thing you know, some coworkers came up with "Indiana Joe and the Ron Temple of Doom".

--
Ciao!

Joe

http://www.swissarmyfork.com
 
If you have a beam-splitter camera w/ 3 6MP (one for each color)
sensors how would you rate the resolving power of the camera? Would
you just compare it to any Bayer 6MP camera? surely it would be a
better image. You probably wouldn't call it a 18mp camera but you
really would like to somehow differentiate it from your normal 6mp
cameras.
I've been tinkering with some new terms to describe this...

http://forums.dpreview.com/forums/read.asp?forum=1027&message=6503328

If the alignment was perfect, you could view your 3 sensor camera an 18 MegaSomething camera with 6 MegaPops of output.
Also are there really 5 mirrors in a Canon penta-mirror ?
No. The mirror angles duplicate those in a glass pentaprism, but in the glass prism, light strikes two of the five sides perpendicular and can pass through, and three sides at angles so it's reflected. Mirrors reflect all light regardless of angle, so two of the five mirrors have to be "missing" for light to be able to enter and exit.

--
Ciao!

Joe

http://www.swissarmyfork.com
 
Pixels do not equal sensors. A pixel is the smallest unit of a
final image, regardless of how it was made. My monitor has 2 Mega
pixels, not 6 (2M red + 2M green + 2M blue). The Foveon puts out
3.4 Mega pixels, period, however the quality of those pixels is
twice as good as those from other DSLR's which makes the final
image look as good as a standard 7Mpixel image would.
This gets at the heart of the misunderstanding many "Bayer people"
have.
As soon as you start labeling people like this, most people stop following your arguments. You should be more careful. I've read some of your other posts, and you usually have valid points. Work on that reputation, not the one that alienates folks because they think you've dismissed them as "Bayer people" who have a "misunderstanding".
Your monitor does not have 2M red, 2M green and 2M blue. It
has a lot more than that, because there are several phosphor dots
for each pixel (at least at reasonable monitor resolutions). This
is called "dot pitch".
That's quanta, not really spatial resolution. The combination of monitor and video card really does have 2 million displayable pxels, each of which has a red, ggreen, and blue parameter.
The Bayer sensor is different: there is only one dot (sensor) at
each pixel location which detects only one oif the colours. the
other colours are not detected and have to be guessed from
neighbouring pixels.
Again, try to avoid prejudicial terms like colors "guessed from neighboring pixels". Foveon sensors need color guessing too, to compensate for the metamerism issues.

--
Ciao!

Joe

http://www.swissarmyfork.com
 
Mega = one million. This has been true for the last four centuries.
Except when it refers to bytes of memory or, more recently, pixels
in an image, which case it means 2^20. This has been true for at
least five decades (i.e. since people "invented" bytes).
That is a good example of a "definition" vs. a "convention".

A definition often established by a standards body, such as the ISO, NIST, or the CIE. It usually has legal status, it will hold up in a court of law.

A "convention" is often specific to one field, or something in recent use, or something just a bit confusing.

For example, the metric prefixes have precise legal definitions, set down by standards organizations, and held up in countless legal tests all over the world.

Hard drive manufacturers adverise capacity in legal, official "megabytes" and gigabytes, where one megabyte = 1,000,000 bytes, and one gigabyte = 1,000,000,000 bytes.

By mathematical coincidence, 2^10 approximates 10^3, so there is a computer industry kilo, mega, and giga that are sueful conventions, but not legal definitions. The strict and precise terms are "kibi", "mebi", and "gibi".

http://physics.nist.gov/cuu/Units/binary.html

Despite these terms being more percise, I have never been able to say "kibibyte" out loud, and continue to use (or misuse) the conventions "kilo", "mega", and "giga".
BTW, I still claim that the Foveon sensor has 3.4 mega pixels; the
Bayer ones have (e.g.) 6.0 "one-third" pixels ;-)
You should join the "MegaSomethings" and "MegaPops" thread.

http://forums.dpreview.com/forums/read.asp?forum=1027&message=6503328

--
Ciao!

Joe

http://www.swissarmyfork.com
 
This is like the John Cleese, Michael Palin, Christopher Guest and Rob Reiner all arguing at once about a dead parrot that goes to eleven.

You guys are killing me.

When I got on this morning and saw that there were 111 messages in this thread I laughed out loud.

Thanks for the comedic break in an otherwise dreary work day.

--
[fjohn]
 
Joe,

Forgive me, I underestimated your depth.
Sarcasm accepted...
I had not realized that
you had refined Mr. Bayer's concept to where you can measure
luminance at every site in a Bayer-pattern color filter array,
I have "refined" nothing. The luminance availiable from a Bayer array is entirely dependent on the spectral distribution of the subject. For a mostly red scene, luminance (or any information, for that matter) is availiable from 1/4 the sensors. For a green scene, 1/2 the sensors. For a grey scene, all the sensors.

For general photographic purposes, Bryce Bayer concluded, quite correctly, that the green sensors produced more useful information than the red or blue, therefore based the array on twice as much green as red or blue. There are situations where this works agains the Bayer pattern sensor, just as there are some where it works in it's favor.
thereby getting as many measurements as you have pixel sensors, and
then still get extra information about color out of those same
measurements. That means that when your idea comes out, we can
expect Bayer cameras to be as sharp per output pixel as the X3
cameras. When get I get one?
I can't tell you where to get one, but your attitude is beginning to make me want to tell you where to put one.

Just keep in mind that, computationally, weather you call them "pixels", "pixels X3", "sites", etc. the Foveon X3 sensor produces 10.3 million chunks of data that have to be stored, moved around, and processed into an image. This makes it 1.7x more computationally expensive than a 6MP Bayer camera to process a picture (either slower shooting or more expensive electronics reqiured) and 1.7x more financially expensive to store the data.

--
Ciao!

Joe

http://www.swissarmyfork.com
 
This is like the John Cleese, Michael Palin, Christopher Guest and
Rob Reiner all arguing at once about a dead parrot that goes to
eleven.

You guys are killing me.

When I got on this morning and saw that there were 111 messages in
this thread I laughed out loud.
So, you decided the obvious thing to do to a thread that is already too long and drifting too much was to throw in your own off topic contribution.

Your comment is definitely a case of the pot calling the kettle black.

--
Ciao!

Joe

http://www.swissarmyfork.com
 
Pixels do not equal sensors. A pixel is the smallest unit of a
final image, regardless of how it was made. My monitor has 2 Mega
pixels, not 6 (2M red + 2M green + 2M blue). The Foveon puts out
3.4 Mega pixels, period, however the quality of those pixels is
twice as good as those from other DSLR's which makes the final
image look as good as a standard 7Mpixel image would.
This gets at the heart of the misunderstanding many "Bayer people"
have.
As soon as you start labeling people like this, most people stop
following your arguments. You should be more careful. I've read
some of your other posts, and you usually have valid points. Work
on that reputation, not the one that alienates folks because they
think you've dismissed them as "Bayer people" who have a
"misunderstanding".
True. I am chastened. I do try not to use words that have a "slant" to them. I will have to try harder.
Your monitor does not have 2M red, 2M green and 2M blue. It
has a lot more than that, because there are several phosphor dots
for each pixel (at least at reasonable monitor resolutions). This
is called "dot pitch".
That's quanta, not really spatial resolution. The combination of
monitor and video card really does have 2 million displayable
pxels, each of which has a red, ggreen, and blue parameter.
Again, I apoligize. I do most of my work on Imax images (where 6MP is pretty low) so I jumped to the erroneous conclusion that you really were looking at a 6 megapixel monitor (just not thinking clearly) and that you thought two million were red (i.e. were confusing pixels with phosper dots). I now understand what you meant, and I agree.
The Bayer sensor is different: there is only one dot (sensor) at
each pixel location which detects only one oif the colours. the
other colours are not detected and have to be guessed from
neighbouring pixels.
Again, try to avoid prejudicial terms like colors "guessed from
neighboring pixels". Foveon sensors need color guessing too, to
compensate for the metamerism issues.
On this one, I'm not so sure I should feel guilty: perhaps i should have said "interpolated" but -- and the last thing I want is to start yet another argument over subtle distinctions of definition -- "interpolation" is very close to "intelligent guess". But I do concede the point: it is a "slanted" way of putting it.

As far as Foveon chips guessing, the Foveon process derives x values from x measurements (where x is 300 for a 10x10 image). The bayer process derives x values from x/3 measurements. Both processes need to do some math, but the amount of isn't of the same magnitude at all.

The original point I was trying to make is that I have noticed a number of posts (although not a large percentage) from people who appear to believe that a bayer pixel is in fact made up of four sensors (similar to the relationship between phosphor dots and pixels on a monitor). This is what triggered my misunderstanding of your post.
 
It was a beam splitter design with separate sensors for each color.......list as I recall was about $50,000....Canon lens mated to a laptop...not the easiest handholdable rig.....big viewfinder though (-:

Came with it's own separate work station computer .... high end machine at the time, and some pretty impressive software for it's day. I forget how many pixels...enough as I recall for a pretty darn good 45x45 inch print though--

Can't knock the company for some good ideas.

Richard Katris aka Chanan
 
I can't tell you where to get one, but your attitude is beginning
to make me want to tell you where to put one.
Joe, thanks for reacting to my needling.

My attitude was formed in reaction to yours, when you make authoritative assertions in contradiction to the ideas of others, yet won't provide a reference source to back them up and won't concede that they might just be your opinions.

Many of your assertions, opinions, and ideas are fine, but when you contradict what I believe, I want to know why, not just because you say so.

j
 
Alatar, thanks for taking the bait.

Since we now agree that "Basing quality on output attributes is ludicrous", how do you justify your previous statement that "I still claim that the Foveon sensor has 3.4 mega pixels;"?

j

Since we agree that
Pixels do not equal sensors. A pixel is the smallest unit of a
final image, regardless of how it was made. My monitor has 2 Mega
pixels, not 6 (2M red + 2M green + 2M blue). The Foveon puts out
3.4 Mega pixels, period...
And the Fuji S2 puts out 12 MP, so why do some people insist on
calling it 6 MP?

j
And my 17x11 600 dpi print has 67.3 pixels, so that means I can
tell people my camera takes 67 megapixel pictures????

Basing quality on output attributes is ludicrous, it's like saying
my AM radio broadcast is super high fidelity because I have great
speakers.
 

Keyboard shortcuts

Back
Top