SIGMA releases an official update about their Foveon X3 sensor project

Canon Rumors Guy

EOS-1D X Mark III
CR Pro
Jul 20, 2010
8,998
1,711
Canada
www.canonrumors.com
SIGMA CEO Kazuto Yamaki has issued an official statement about the Foveon X3 1:1:1 sensor development which was scrapped last February due to a critical flaw in the sensor design that would make it impossible to mass-produce.
SIGMA has gone back to the drawing board and restarted the research and development from scratch. SIGMA cannot give a timetable as to when they can tell us more about the new image sensor.

Statement from SIGMA:
First of all, thank you very much for your continued support and interest in our products.
In February 2020, we announced our decision to restart our plans for a full-frame Foveon X3 sensor camera, beginning with the development of a new sensor technology. We have not been able to share any information regarding this project in the time since, and I would like to use this short message to provide an update about the situation.
As previously announced, we halted the development of the project after the sensor we were working with could not go into mass...

Continue reading...


 

JohnC

EOS 90D
CR Pro
Sep 22, 2019
158
174
Gainesville,GA
Sounds like it is in the beginning stages at this point. I do not know an extreme amount about Foveon sensors but from what I recall this would be an exciting development if they are successful.
 

amorse

EOS RP
Jan 26, 2017
784
1,043
www.instagram.com
I feel bad for their team - that must be just crushing to find a fatal design flaw when they're trying to break into the segment. Good on them for sticking with it and working through it.
 
  • Like
Reactions: dtaylor

slclick

Unsolicited & Always Free
Dec 17, 2013
4,440
2,625
I've been so close to pulling the trigger on various Siggy bodies but always found reasons to wait...maybe this will be the sensor. Still(s), if it's another teensy body or requires a tripod 80% of the time, I will once again pass.
 

Joules

doom
CR Pro
Jul 16, 2017
1,592
1,909
Hamburg, Germany
Oh, can we have a photo sites vs photodiodes resolution argument please?

After all if a Foven 20mp is classed as 60mp then an R5 45mp could just as easily be called a 90mp because of the dual pixel.
Wouldn't an R5 more reasonably be called a 15-ish MP? After all, 2/3 of the color information is just inferred from the little that's actually captured.

Dual pixel doesn't add to the resolution, although it has potential beyond what Canon uses it for.
 

jvillain

EOS RP
Sep 29, 2018
256
200
I wonder whether Sigma will survive a digital still camera market that only ships about 9 million annually.
Well they stated a couple of months ago that they aren't getting into RF glass yet due to lack of capacity so I doubt that is much of an issue. I would be much more worried about some of the other smaller players. But that makes a great segue to when are we getting Sigma RF glass?
 

privatebydesign

EOS-1D X Mark III
CR Pro
Jan 29, 2011
9,975
4,729
Wouldn't an R5 more reasonably be called a 15-ish MP? After all, 2/3 of the color information is just inferred from the little that's actually captured.

Dual pixel doesn't add to the resolution, although it has potential beyond what Canon uses it for.
I don’t see why, resolution could never be defined by color accuracy. Besides there is generally a complete misunderstanding of how the information collected through a Bayer array is subsequently processed, and depending on how that is done the color accuracy can be exceptionally high, certainly considerably higher than our eyes are capable of faulting.
 

Joules

doom
CR Pro
Jul 16, 2017
1,592
1,909
Hamburg, Germany
I don’t see why, resolution could never be defined by color accuracy. Besides there is generally a complete misunderstanding of how the information collected through a Bayer array is subsequently processed, and depending on how that is done the color accuracy can be exceptionally high, certainly considerably higher than our eyes are capable of faulting.
I got the impression you were wanting somebody to bring up a question about resolution, so I tried to do you the favor.

The argument was just that if anything, it should be argued the the resolution is lower, not higher. Just like it should not be argued that having true color sampling makes 20 MP spatial resolution anything other than 20 MP spatial resolution.

What the eye can or cannot see doesn't really matter either, does it? It's what we can extract from given image data through digital processing that ultimately reveals to the eye many things it could not capture otherwise. And there certainly are scenarios in which debayering can fail, leading to visual artifacts.

The theoretical appeal of foveon is more about increased signal gathering than color detail in my personal priorities. Although in practice that hasn't been working out that great, as far as I'm aware.
 

Antono Refa

EOS R
Mar 26, 2014
1,266
418
Sigma already makes crop Foveon sensors, so what is it they had a problem with? As far as I can gather, it has something to do with taking only one sample at the top, blue layer, rather than splitting it into four.
 

usern4cr

R5
CR Pro
Sep 2, 2018
1,055
1,643
Kentucky, USA
Since "the bear has been poked" (above), I'd suggest that to really compare a Foveon sensor to a Bayer array you would have to compare a QE (quantum efficiency) chart for each pixel element type of R, G, and B for the Foveon and Bayer arrays. The level of QE sensitivity matters greatly, but so does the crosstalk within each pixel element (that is, if a "green" pixel element is also partially sensitive to red or blue then you have crosstalk which has to be dealt with). To me this is an issue that is as important (or even more so) than just comparing how many pixel elements there are since it affects overall sensor sensitivity and contrast.

And, for what it's worth, I think "dual pixels" are mainly for AF use (and incredibly great ones at that!), but I consider them so close to each other in a Bayer array that they behave more as a single pixel element in resolution than 2 pixel elements. But AF ability is just as important (or more) to consider as just QE & crosstalk and #pixel elements. And future quad pixels will even further improve AF ability.

So, to me, even if Sigma can get their Foveon sensor to work with good QE and acceptably low cross-talk, it still won't be enough to compete with Canon DP or QP technology unless it has some really good phase detection added to it, which may be asking too much of it in such a competitive and shrinking market of today. While I am rooting for them to succeed, I think it is going to be a very difficult uphill battle for them.
 
  • Like
Reactions: JohnC and slclick

Atlasman

EOS R5
CR Pro
May 14, 2020
66
79
Sigma already makes crop Foveon sensors, so what is it they had a problem with? As far as I can gather, it has something to do with taking only one sample at the top, blue layer, rather than splitting it into four.
I suspect it has to do with video.
 
Feb 25, 2013
2
6
I’m so grateful to the Engineers who work so hard on these projects.

Listening to CEO Kazuto Yamaki reminds me of all their hard work and patience.

I love this CEO and his sincere and humble attitude, and I love Sigma as a company. I own the ART 50mm and the ART 85mm and could not be happier with the quality and value from these lenses.

Great thanks to you Kazuto Yamaki and great thanks to your team. I appreciate that you have the bravery to undertake this innovation in sensor technology and I look forward to the result when it finally arrives.
 

usern4cr

R5
CR Pro
Sep 2, 2018
1,055
1,643
Kentucky, USA
I’m so grateful to the Engineers who work so hard on these projects.

Listening to CEO Kazuto Yamaki reminds me of all their hard work and patience.

I love this CEO and his sincere and humble attitude, and I love Sigma as a company. I own the ART 50mm and the ART 85mm and could not be happier with the quality and value from these lenses.

Great thanks to you Kazuto Yamaki and great thanks to your team. I appreciate that you have the bravery to undertake this innovation in sensor technology and I look forward to the result when it finally arrives.
I'll second this. Yamaki San is IMHO the only corporate owner/leader who talks to us, the photographers, with an open, honest, informative and humble attitude. It's a joy to see and hear him talk. I've always hoped whatever he would do would succeed. If he was in charge of Olympus from the beginning then they'd probably still be in business with some wonderful and competitive camera bodies & lenses even now. Now if he was allowed full access to the RF protocols, imagine what lenses he could come up with for us. :unsure:
 

slclick

Unsolicited & Always Free
Dec 17, 2013
4,440
2,625
This is do or die for Sigma's camera development. I hope they pull it off but not optimistic.
They keep chugging along, you could have said that same thing 15 years ago. The lenses keep them afloat. It's Tokina that I cannot for the life of me continues as a business.
 
  • Like
Reactions: JohnC

JohnC

EOS 90D
CR Pro
Sep 22, 2019
158
174
Gainesville,GA
They keep chugging along, you could have said that same thing 15 years ago. The lenses keep them afloat. It's Tokina that I cannot for the life of me continues as a business.
I haven’t analyzed numbers but my impression is that Sigma, and their image, have improved significantly over the last 20 years. Perhaps it is just my personal impression but their products have gone from “well I can’t pay the price for x brand so this Sigma will work” to “I think this Sigma outperforms the OEM in this case”. I’m not disparaging all of their older products by any means but they have become a player. Granted I am referring g to their lenses, not camera bodies...but it still speaks to the health and trajectory of their company in this market. Tonkina, Tamron...I’m not sure they have the same thing going. Again no facts just my gut feeling.
 
<-- start Taboola -->