Full colour vs Bayer

Here is an interesting comparision of the full colour mode of the Pentax
K3 (via pixel shift) and the Nikon D810.
http://www.imaging-resource.com/PROD...k3-iiTECH2.HTM

Basically, by having the full colour information at each pixel, the K3
with its 24MP matches the resolution captured by the D810 with 36MP
(even outperforms the D810 in some shots).

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.
--
Alfred Molon

Olympus E-series DSLRs and micro 4/3 forum at
http://tech.groups.yahoo.com/group/MyOlympus/
http://myolympus.org/ photo sharing site

In article ,
Alfred Molon wrote:

Here is an interesting comparision of the full colour mode of the Pentax
K3 (via pixel shift) and the Nikon D810.
http://www.imaging-resource.com/PROD...k3-iiTECH2.HTM

Basically, by having the full colour information at each pixel, the K3
with its 24MP matches the resolution captured by the D810 with 36MP
(even outperforms the D810 in some shots).

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.

There are RAW conversions that will deliver almost 100% resolution from
a Bayer pattern with the cost of some distortion. (Apple Aperture did
it with my Canon T3i and it was sometimes annoying.) It looks like the
images are benefiting from a combination of avoiding a very safe but
dull RAW conversion, less noise, and better sharpening made possible by
the reduced noise. Overall it looks like a big improvement.


Some cellphone cameras can do this too. They take a burst of images,
compensate for motion, and stack the parts that align well. Unlike the
pro camera, it requires some shaking from being handheld and plenty of
luck.

--
I will not see posts from astraweb, theremailer, dizum, or google
because they host Usenet flooders.

In article , Alfred
Molon wrote:

Here is an interesting comparision of the full colour mode of the Pentax
K3 (via pixel shift) and the Nikon D810.
http://www.imaging-resource.com/PROD...k3-iiTECH2.HTM

Basically, by having the full colour information at each pixel, the K3
with its 24MP matches the resolution captured by the D810 with 36MP
(even outperforms the D810 in some shots).

no it doesn't, but even if it did, it would only work if the subject
didn't move.

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.

no it doesn't. it gets you better chroma, which the eye can't resolve.

In article ,
nospam wrote:

In article , Alfred
Molon wrote:

Here is an interesting comparision of the full colour mode of the Pentax
K3 (via pixel shift) and the Nikon D810.
http://www.imaging-resource.com/PROD...k3-iiTECH2.HTM

Basically, by having the full colour information at each pixel, the K3
with its 24MP matches the resolution captured by the D810 with 36MP
(even outperforms the D810 in some shots).

no it doesn't, but even if it did, it would only work if the subject
didn't move.

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.

no it doesn't. it gets you better chroma, which the eye can't resolve.

That depends on the RAW conversion. Interpolating nearly 100% luminance
resolution from a Bayer pattern produces significant distortion and
noise. It actually doesn't look that good. You can have a blurry and
clean image or a sharp and dirty image. The data to produce a clean and
sharp image does not exist without some kind of super-res trick like the
Pentax is doing.

--
I will not see posts from astraweb, theremailer, dizum, or google
because they host Usenet flooders.

In article , Kevin
McMurtrie wrote:

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.

no it doesn't. it gets you better chroma, which the eye can't resolve.

That depends on the RAW conversion.

no it doesn't.

Interpolating nearly 100% luminance
resolution from a Bayer pattern produces significant distortion and
noise.

only if you have a ****ty raw converter.

It actually doesn't look that good.

millions and millions of photos show that claim to be false.

You can have a blurry and
clean image or a sharp and dirty image. The data to produce a clean and
sharp image does not exist without some kind of super-res trick like the
Pentax is doing.

it's a gimmick, and if the subject moves, the artifacts are *really*
bad.

In article ,
RichA wrote:

You can have a blurry and
clean image or a sharp and dirty image. The data to produce a clean and
sharp image does not exist without some kind of super-res trick like the
Pentax is doing.

it's a gimmick, and if the subject moves, the artifacts are *really*
bad.

Everything is a gimmick, except true tri-colour imaging.

other way around.

moving the sensor to increase chroma which the eye can't even resolve
is the very definition of gimmick.

In article ,
nospam wrote:

In article , Kevin
McMurtrie wrote:

In simple words, the full colour information gives you a gain of about
50% in pixel count vs a standard Bayer imager.

no it doesn't. it gets you better chroma, which the eye can't resolve.

That depends on the RAW conversion.

no it doesn't.

Interpolating nearly 100% luminance
resolution from a Bayer pattern produces significant distortion and
noise.

only if you have a ****ty raw converter.

X single color pixels can't be converted into X full color pixels.
There must be blurriness or distortion. It's simple information theory
and math.



It actually doesn't look that good.

millions and millions of photos show that claim to be false.

I'm pretty sure you don't know what has been done to the photos since
none of them are yours.


You can have a blurry and
clean image or a sharp and dirty image. The data to produce a clean and
sharp image does not exist without some kind of super-res trick like the
Pentax is doing.

it's a gimmick, and if the subject moves, the artifacts are *really*
bad.

The artifacts don't have to be really bad. Even cell phone cameras that
came out a year ago can correct for some motion and discard areas that
can not match, like areas revealed or obscured rather than shifted.
Excessive motion prevents enhancement but it doesn't cause ghosting like
dumb HDR blenders do.

--
I will not see posts from astraweb, theremailer, dizum, or google
because they host Usenet flooders.

In article , Kevin
McMurtrie wrote:

Interpolating nearly 100% luminance
resolution from a Bayer pattern produces significant distortion and
noise.

only if you have a ****ty raw converter.

X single color pixels can't be converted into X full color pixels.

of course they can.

There must be blurriness or distortion. It's simple information theory
and math.

nope.

chroma is sampled at lower rate than luma, but it's still higher than
what the eye can resolve. all detail below nyquist can reproduced.

It actually doesn't look that good.

millions and millions of photos show that claim to be false.

I'm pretty sure you don't know what has been done to the photos since
none of them are yours.

that doesn't matter.

what matters is that they're not blurry or distorted, namely the ones
straight out of camera.

You can have a blurry and
clean image or a sharp and dirty image. The data to produce a clean and
sharp image does not exist without some kind of super-res trick like the
Pentax is doing.

it's a gimmick, and if the subject moves, the artifacts are *really*
bad.

The artifacts don't have to be really bad. Even cell phone cameras that
came out a year ago can correct for some motion and discard areas that
can not match, like areas revealed or obscured rather than shifted.
Excessive motion prevents enhancement but it doesn't cause ghosting like
dumb HDR blenders do.

the artifacts really are bad.