Any thoughts /news on Foveon sensors?

"Peter A. Stavrakoglou" wrote in message
...
it's 3.4 MP
sensor resolves as well as any 6MP Bayer, sometimes even better.


I don't understand how that is a technical possibility.
Helen

"Gormless" wrote in message
...

"Peter A. Stavrakoglou" wrote in message
...
it's 3.4 MP
sensor resolves as well as any 6MP Bayer, sometimes even better.


I don't understand how that is a technical possibility.
Helen

Bayer needs 4 sites per pixel - Foveon does it with depth, 3 on top of one,
so to speak..

--
M Stewart
Milton Keynes, UK
http://www.megalith.freeserve.co.uk/oddimage.htm

"Peter A. Stavrakoglou" wrote in message
...
"Edmund" wrote in message
...

As far as I understand things, this foveon sensor
looks the best technology to me.
Why is it hardly anybody uses it?
Are there made to big investments in other systems
and are "they" therefore choosing a lesser system?

At this time, only Sigma and one Polaroid P&S (if you can find one) are
using these sensors in a consumer camera. There will be a new Sigma SLR
soon, patience will pay off in waiting for it.

Compared to Bayer technology, the Foveon is still in its infancy and yet
it produces very sharp photos with good color rendition in a Sigma DSLR.
As development continues, it can only improve like Bayer technology has.
Despite their being many more years of development down the road, the
current Foveon sensor can hold its own. There needs to be improvement in
noise handling and perhaps in-camera jpeg in the Sigma DSLRs.

It only holds its own with the bottom of the line cameras like the Nikon D50
and Minolta 5D. It has been left behind by such cameras as the Oly E-500
and Canon RebelXT/350D, and seriously dusted off by cameras like the Nikon
D200 (at the price point the Sigma SD-10 was introduced) and the admittedly
much more expensive Canon 5D.

--
Skip Middleton
http://www.shadowcatcherimagery.com

In article , Gormless
wrote:

I don't understand how that is a technical possibility.
Helen

It's not. The Sigma is a 3.42MP camera.

Peter A. Stavrakoglou wrote:
"Mark²" mjmorgan(lowest even number wrote in message
news:526yf.10707$V.7118@fed1read04...
Edmund wrote:
As far as I understand things, this foveon sensor
looks the best technology to me.
Why is it hardly anybody uses it?

Because its only incarnation came in the for of the extremely poor
Sigma body.
-Don't know if its the fault of Sigma or Foveon...but the skin
renditions were absolutely horrid, and many other texture renditions
were just plain awful. The lack of a filter over the sensor often
leads to problematic fine line renditions... The list goes on.

Saying that skin renditions are horrid is an overstatement. Depending on
the exposure, skintones tend to be yellow but horrid
they are not. Easily correctable but some take issue with havng to
do that.

Disagree.
It often went beyong mere color.
Texture was often the culprit in addition to color, which is pretty
impossible to recreate in PS.

The Sigma camera themselves are not "extremely poor", that
is only your opinion.

Right. My opinion.

It's a well-built body with vey good
ergonomics and in the right hands produces outstanding images. Search
pbase and you'll see. Those of us who use them know how good
it really is, t seems that statements like yours come from people who
never used the camera but I am not making a judgment on your
situation. Perhaps you've used one, I don't know, but I have for two
years and in my totally amateurish hands it has taken some great
photos. The level of sharpness is surprising at times.

Peter A. Stavrakoglou wrote:
"Darrell Larose" wrote in message
. ..

"Edmund" wrote in message
...

As far as I understand things, this foveon sensor
looks the best technology to me.
Why is it hardly anybody uses it?
Are there made to big investments in other systems
and are "they" therefore choosing a lesser system?

The lack of use, and the lack of any updates indicates that Foveon
is a dead end. There has been no new developments for several years.
Foveon is stuck at 3.4 megapixels.

Not true. Although the choice for a reseller using WWL for the
Polariod camera using a Foveon was not a good one, that alone shows
that there has been development in the last two years. That camera
could produce some good images. The Sigma SD10 is a little over two
years old and it's 3.4 MP sensor resolves as well as any 6MP Bayer,
sometimes even better.

Do you have images posted anywhere?
You seem to indicate better results than I typically saw from
Sigma;--particularly skin tones and textures.
If you've got the goods, I'm open to reinterpretation of my opinion...though
I have yet to see skin renditions that consistently un-do my perception of
this.

Scott W wrote:
Edmund wrote:

As far as I understand things, this foveon sensor
looks the best technology to me.
Why is it hardly anybody uses it?
Are there made to big investments in other systems
and are "they" therefore choosing a lesser system?

Given the noise limitation of the foveon sensor it really needed to
bring something to the table that was not otherwise available. With
its 3.4 MP it came very close to matching a 6 MP camera, but it did not
even manage to do this. If they could come out with a 8 MP version and
control the noise better they might get somewhere.

In truth the loss from the Bayer pattern is fairly small, say between
25 to 30% of the pixel count. This assumes you are shooting raw not
just using the in camera jpeg photos.

Scott

Scott, how can you say only a 25 to 30% loss? With a CCD fully 66% of
the image is interpolated.

Alan wrote:

Scott, how can you say only a 25 to 30% loss? With a CCD fully 66% of
the image is interpolated.

Here is a test I did
http://www.pbase.com/konascott/image/54437327
The top photo is a straight out of the 20D (well from raw), The middle
photo is about as sharp as be had for the number of pixels, it is down
sampled by a large amount. The difference between these two images
then is the loss from the Bayer pattern. To test how much loss there
is simple down sample and back up, when you find the amount of down
sampling need to degrade the middle image to match the top you pretty
much know what the loss is. Down sampling by 85% and back up come very
close to what the camera produces, square this and you end up with a
loss of 28%. But some of this loss is most likely due to the lens as
well as the Bayer pattern so the real loss from the Bayer pattern will
be somethin gless then 28%

Scott

"Randall Ainsworth" wrote in message
...
In article , Gormless
wrote:

I don't understand how that is a technical possibility.
Helen

It's not. The Sigma is a 3.42MP camera.

I think Peter A S must be a Sigma adherent.
If someone believes something hard enough if it becomes true for them.

Edmund writes:

As far as I understand things, this foveon sensor
looks the best technology to me.
Why is it hardly anybody uses it?

Because it's a LOT better in theory than it turns out to be in
practice. At least so far.
--
David Dyer-Bennet, , http://www.dd-b.net/dd-b/
RKBA: http://noguns-nomoney.com/ http://www.dd-b.net/carry/
Pics: http://dd-b.lighthunters.net/ http://www.dd-b.net/dd-b/SnapshotAlbum/
Dragaera/Steven Brust: http://dragaera.info/