From: Alfred Molon on
In article <hm8203$aa3$2(a)news.eternal-september.org>, david-
taylor(a)blueyonder.co.uk.invalid says...
>
> So do you want a camera to produce images, or one for pixel peeping?

Make an enlargement and you will see the difference.

And by the way, also a 640x480 camera can produce images.
--

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
From: David J Taylor on
"Alfred Molon" <alfred_molon(a)yahoo.com> wrote in message
news:MPG.25f1fac43f1d0f9498c235(a)news.supernews.com...
> In article <hm8203$aa3$2(a)news.eternal-september.org>, david-
> taylor(a)blueyonder.co.uk.invalid says...
>>
>> So do you want a camera to produce images, or one for pixel peeping?
>
> Make an enlargement and you will see the difference.
>
> And by the way, also a 640x480 camera can produce images.
> --
>
> Alfred Molon

That would be an interesting test - making the same size print from a
4.6MP Foveon and a 12MP (or whatever) Bayer DSLR. How close would you
need to look at the print to see the difference? Probably far closer than
most people would naturally view the print.

Pixel-peeping on my monitor is the equivalent of a 1100mm (43-inch) wide
print.....not a width I would ever print. If your prints are that big, I
appreciate that you will have different requirements.

Cheers,
David

From: SMS on
Alfred Molon wrote:
> Is there any reason why Sigma won't increase the pixel count in its line
> of Foveon DSLRs?

The noise and color separation problems would get even worse, and
they're already pretty bad. To invest huge amounts of money in sensor
development for a line of D-SLRs with close to 0% market share does not
make any sense. The Foveon sensors are also very difficult to
manufacture because of the layering and the silicon color separation.

That's not to say that someone in the future might not come up with a
sensor that layers pixel sensors on top of each other, but there is no
practical reason to do that unless they can solve the technical problems
that have affected Foveon.
From: SMS on
Bruce wrote:
> On Tue, 23 Feb 2010 19:15:58 +0100, Alfred Molon
> <alfred_molon(a)yahoo.com> wrote:
>
>> Is there any reason why Sigma won't increase the pixel count in its line
>> of Foveon DSLRs?
>
>
> Because the amount of image information collected is three times the
> nominal resolution of the sensor. At each pixel position there are
> red, blue and green receptors, each on a different layer. The result
> is that the Foveon sensor collects the same amount of information as a
> 14.2 MP Bayer pattern sensor.

No it doesn't. It collects the same amount of color information (or it
would if silicon color separation worked as well in practice as it does
in theory. Interpolating on a 14 MPixel Bayer sensor provides just as
much color information, but 3x the spatial information.

> It is fairly widely recognised that the Foveon sensor delivers much
> the same image quality as a 12 MP Bayer sensor.

That has never been recognized. It's simply what Sigma/Foveon claims.
From: SMS on
Mr. Strat wrote:
> In article <MPG.25eee8cf5471cc2598c230(a)news.supernews.com>, Alfred
> Molon <alfred_molon(a)yahoo.com> wrote:
>
>> Current 4/3 cameras have 12MP and the 4.6MP of the Foveon should be on
>> the same level of a 7MP Bayer sensor camera, but not more than that. If
>> Foveon brought out a 10MP model they would be competitive.
>
> The bottom line is that Foveon is just crappy technology...unless
> you're fond of Homer Simpson skin tones.

The original Foveon camera used three sensors and prisms. It was the
best portrait camera on the market at the time, but such a system was
too large, heavy, and expensive to scale to D-SLRs. The stacked layer
sensor was their attempt to scale the full color pixel technology to
smaller camera. Unfortunately, silicon color separation did not work out
in practice as well as it did in theory. The sensors were difficult to
manufacture and had problems with noise and color accuracy, requiring
vast amounts of post-processing.

A 4.6MP Foveon sensor is no better than a 4.6MP Bayer sensor. They both
have the same number of pixels. The Bayer uses interpolation, the Foveon
uses silicon color separation. Neither is ideal, both work.