Foveon X3 Quattro direct image sensor: any opinions?

Join the discussion
Ask a follow-up here, or get your own question answered by working scientists, mathematicians and engineers — people, not an autocomplete.
Real named experts · corrections over time · the nuance an AI answer skips
7 replies · 2K views
Messages
30,547
Reaction score
7,529
I just came upon this sensor, in a new(?) camera by Sigma. It seems to be a clever idea. As anyone here got any special knowledge of the device or the technology? It solves a number of problems but, no doubt, introduces some new ones. Incredibly high res!
I will stick with Pentax until I become very rich - enough to make a change of lenses- so I have no Interest in the company or the device.
 
on Phys.org
I don't know much about these things, but there seems to be a lot of marketing bs in there. For instance, the statement
In the Foveon direct image sensor, there are no color filters which would cause a loss of information transmitted by light.
is a stretch considering that
Leveraging the light absorption characteristics of silicon, the sensor comprises three layers of photodiodes, each at a different depth within the silicon and each corresponding to a different RGB color.
sounds functionallly equivalent to filtering.
 
sophiecentaur said:
I just came upon this sensor, in a new(?) camera by Sigma.

The camera may be new but the Fovean type of sensor has been around for a while now and never seemed to make significant inroads into the marketplace. Not sure if that's due to the sensor itself or Sigma's implementation of the sensor in their cameras. They certainly have come up with some nice lenses recently.
 
DrClaude said:
sounds functionallly equivalent to filtering
Well, it has to be because, if the blue sensor is using all the available light then it is 'filtering it out' as to goes through. But I think the point is that the whole available area of the sensor is available for each colour of sensor. It's almost as if the copy writer has missed the point.
Sigma:
pixel said:
They certainly have come up with some nice lenses recently.
They have always had some pretty good high end lenses. At several £k, some of them are pretty popular with the big boys. Their regular lenses seem to be at least as good as the main camera manufacturers' kit lenses.
 
sophiecentaur said:
I just came upon this sensor, in a new(?) camera by Sigma. It seems to be a clever idea. As anyone here got any special knowledge of the device or the technology? It solves a number of problems but, no doubt, introduces some new ones. Incredibly high res!
I will stick with Pentax until I become very rich - enough to make a change of lenses- so I have no Interest in the company or the device.

I've been aware of the foveon chip since it came out in the late 1990s. Seems fantastic, not sure why it has yet to penetrate the market.
 
Andy Resnick said:
I've been aware of the foveon chip since it came out in the late 1990s. Seems fantastic, not sure why it has yet to penetrate the market.
Some practical reasons that have been given:
- The color layers are not so orthogonal in response, hence some tricks need to be applied to get good color, hence reduction in the relative noise advantage
- Less aggressive development relative to Bayer sensors, leading to reduction in other Foveon advantages such as the inherent resolution advantage
- Sigma is not one of the "big" brands
 
olivermsun said:
- The color layers are not so orthogonal in response, hence some tricks need to be applied to get good color, hence reduction in the relative noise advantage
I guess they have the same problem as with colour film, with its layers. But the foveon layers are not 'use once and throw away', as with cheap film so Sigma can afford to make it work better. I remember working on colour Telecine systems and it was a nightmare to get good enough colour for TV. Each stock was different. Pictures that would look fine for visual viewing would be totally mangled when passing through the PAL TV analysis and display.
 
sophiecentaur said:
I guess they have the same problem as with colour film, with its layers. But the foveon layers are not 'use once and throw away', as with cheap film so Sigma can afford to make it work better. I remember working on colour Telecine systems and it was a nightmare to get good enough colour for TV. Each stock was different. Pictures that would look fine for visual viewing would be totally mangled when passing through the PAL TV analysis and display.
At this point in the evolution of Foveon, the output colors (and everything else) look great. The noise performance just isn't up there with the best conventional sensors, despite using "all" the area for all 3 colors channels.
 
  • Like
Likes   Reactions: jim mcnamara and sophiecentaur