• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Sensor type discussion

Harvey Moore

New member
Over at the fm site this discussion http://www.fredmiranda.com/forum/topic/415203 is going on, 61 pages at last count.

The discussion is covering the virtues of cmos, ccd, foveon, anti-alias v no anti-alias. Some of the posts have sparked my curiosity.

Does anyone have links or facts that cover the current State of the Art re ccd, cmos, foveon et al ?

thanks,

harvey
 

Dierk Haasis

pro member
There's a promo site by Canon, telling a bit about the current state of (their) CMOS development.

For current state of technological development info Wikipedia is not such a bad source.

Personally, although admittedly all the arguments in favour of one sensor of the others are valid, I find current sensor designs are so good that the discussions are largely academic. Just look at Nikon, starting out with CCD on the D1-series, then changing to a CMOS-like design with their LBCAST, changing to CMOS for the D2x ... Obviously they are still improving hardware and software. Looking at D1x, Canon G2 and D2x images shows at best only subtle changes.
 

Asher Kelman

OPF Owner/Editor-in-Chief
Harvey Moore said:
Over at the fm site this discussion http://www.fredmiranda.com/forum/topic/415203 is going on, 61 pages at last count.

The discussion is covering the virtues of cmos, ccd, foveon, anti-alias v no anti-alias. Some of the posts have sparked my curiosity.

Does anyone have links or facts that cover the current State of the Art re ccd, cmos, foveon et al ?

thanks,

harvey

Harvey,

I have not as yet read the 60 pages you referenced.

I do know that there is a consortium, including Canon (an I believe HP) which is involved in exploiting a an advanced CMOS chip.

The main features, as of 3-5 years ago, are the following:

1. Each pixel is independently addressed and so in effect is a single cell camera.

2. Adjacent pixels can be binned to increase sensitivity

3. Each pixel has a Foveon-like three color sensitivity

Since this work is already some 3-5 years old, I would expect to see revolutionary cameras in the next several years. I wouldn't be surprised if it would appear first in intelligent security cameras as well as smart digicams.

Asher
 

Harvey Moore

New member
Asher,

About 6 years ago I was working with a company based in Israel that developed a camera based system to detect flaws in the form of chips on the edges of carbide insert cutting tools. The camera and optics were based on Obsolete Israeli surveillance systems, and could detect (automaticaly) a 10-15 micron chip.

The system was not purchased because of mechanical problems, not optical or programming problems.

This said, it would not surprise me if sensors and decoding programming similar to what you describe are being used by governments right now.

Also, Canon is working with Microsoft, see this: http://www.eweek.com/article2/0,1895,1858595,00.asp

I wonder what they are really doing beyond what has been announced.

I think that what we publicly know now is only the tip of the iceberg.

harvey
 

Tom Yi

New member
From what I've heard, mainly about CCD's and CMOS, CCD's came out first then the CMOS. The main diff between the two is that CCD has it's circuit and whatnot out side of the sensor itself while CMOS has it's supproting circuits and whatnot actually in the sensor itself. Theorethically, CCD's are hence to be less noisy and CMOS cheaper and less power consuming.

In reality CMOS seems to hold all the cards as they are less noisy, perhaps cheaper to make, and use less energy. For some reason, most CCD sensor cameras seem to take hugh RAW files, e.g. the Nikon D200 has a 10.2MP but it's RAW files are about 15MB's while a 8MP or a 12MP CMOS sensor cameras have RAW images right around 8 and 12MB's respectively.
 
Tom Yi said:
For some reason, most CCD sensor cameras seem to take hugh RAW files, e.g. the Nikon D200 has a 10.2MP but it's RAW files are about 15MB's while a 8MP or a 12MP CMOS sensor cameras have RAW images right around 8 and 12MB's respectively.

There are additional things like noise reduction (sometimes destructive to image detail) and lossy/lossless image compression going on as well. That will also affect the file size.

Bart
 
Not a Sensor Technology Issue

Tom Yi said:
In reality CMOS seems to hold all the cards as they are less noisy, perhaps cheaper to make, and use less energy. For some reason, most CCD sensor cameras seem to take hugh RAW files, e.g. the Nikon D200 has a 10.2MP but it's RAW files are about 15MB's while a 8MP or a 12MP CMOS sensor cameras have RAW images right around 8 and 12MB's respectively.

Canon's RAW files are losslessly compressed while IIRC Nikon's NEFs are uncompressed. This is unrelated to sensor technology and is more an aspect of Canon's image processing ASICs (i.e., DIGIC and DIGIC II) compressing the data before saving the file.
 
Top