| Author | Thread |
|
|
01/17/2006 07:45:15 PM · #1 |
Has anyone read about the new sensor that Koreans have developed?
A company called Planet82 has come up with a new, highly sensitive sensor that is for still and video cameras.
It's called SMPD, Single Module Photo Detector.
It's based on nano-technology, enableing an increase in pixels, meaning an increase in density.
What does this mumble-jumble mean to the photographer?
A camera that does not need a flash, that can capture a clear, color image when there is no light.
Apparently, this chip can capture an image when the light level is at 0.1 lux.
In case no one knows what a lux is, 1 lux is equal to the brightness of a candle one meter away from a lens. This chip can capture light at 0.1 lux.
Amazing.
SMPD
Message edited by author 2006-01-18 00:47:17.
|
|
|
|
01/17/2006 08:24:25 PM · #2 |
Cool stuff. I believe we are on the verge of true digital revolution with photograhpy where technology is going to start improving at a stagering pace. Let's just hope memory card and hard drive capacities can keep up.
At 0.1 lux I wonder what ISO that would equate to... |
|
|
|
01/17/2006 08:38:36 PM · #3 |
| First quarter '06 huh? When's the 30D (or whatever) coming out? |
|
|
|
01/17/2006 09:02:05 PM · #4 |
That's pretty cool. Wonder if it'll really be as good as it sounds, considering life span and power consumption and real world performance with a lot of pixels on the chip.
|
|
|
|
01/17/2006 09:08:03 PM · #5 |
It seemed a rather vague article.
And what does "...and is half the size of current CCD sensors." supposed to mean?
|
|
|
|
01/17/2006 09:19:15 PM · #6 |
| as always, these sort of scientific progress could only benefit us :) |
|
|
|
01/18/2006 04:44:23 AM · #7 |
|
|
|
01/18/2006 04:57:10 AM · #8 |
I know the question that's on EVERYONE's mind about this issue:
What does Kirbic think??
;-) |
|
|
|
01/18/2006 05:27:26 AM · #9 |
Originally posted by nickp37:
First quarter '06 huh? When's the 30D (or whatever) coming out? |
I heard from a canon rep that it was in march.
|
|
|
|
01/18/2006 05:30:14 AM · #10 |
The answer is, I don't think, until I've had my coffee. ;-)
It's a given that, at some point, sensors based on quantum mechanical effects will hit the market. This implementation seems targeted to very compact sensors for video and camera phones, indicating that the image quality is low. Bear in mind, though, that CMOS was perceived as "only for low-end" just a few years ago, and it now dominates the high end.
Speaking in generalities, there's a limit to the sensitivity improvement we can expect, and we're closer to that limit than most folks think. Beyond about ISO 6400 or so, just the statistics of photon arrival will introduce very significant noise, even if the sensor and electronics add little. The only other avenue of improvement will be to do away with color filters (since using them implies that we throw away 67% of incident photons) and sense photon energy (wavelength) directly. That, IMO, is quite a ways off. Though it is easily done in the lab, constructing megapixel arrays with the ability to both count individual photon arrivals and track the energy distributions is pretty far in the future.
For reference, the Foveon technology actually does skirt some of the color filter issue. Using the penetration depth to differentiate means that you effectively can use all the photons. The results to date, however, have been underwhelming in some respects.
All these advances will, eventually, result in improved sensing for high-end imaging, to the benefit of us all. The next ten years will be a very interesting time to be in photography. We had all better keep an adequate budget set aside for body upgrades :-P
|
|
|
|
01/18/2006 06:02:12 AM · #11 |
Originally posted by American_Horse:
Apparently, this chip can capture an image when the light level is at 0.1 lux. |
What's that supposed to mean? Any camera can take images of any source of light, no matter how dark it is. This specification only makes sense when it is related to exposure time, doesn't it?
|
|
|
|
01/18/2006 10:19:11 AM · #12 |
| I think it's refering to instantaneous capture. No exposure time needed. Kind of like capturing exactly what your eyes see exactly at that moment (the most complicated sensor of them all!) so potentially you could capture a runner at night with no flash... (or that's my guess to this technology at least). |
|
|
|
01/18/2006 10:28:45 AM · #13 |
Originally posted by kirbic:
The answer is, I don't think, until I've had my coffee. ;-)
It's a given that, at some point, sensors based on quantum mechanical effects will hit the market. This implementation seems targeted to very compact sensors for video and camera phones, indicating that the image quality is low. Bear in mind, though, that CMOS was perceived as "only for low-end" just a few years ago, and it now dominates the high end.
Speaking in generalities, there's a limit to the sensitivity improvement we can expect, and we're closer to that limit than most folks think. Beyond about ISO 6400 or so, just the statistics of photon arrival will introduce very significant noise, even if the sensor and electronics add little. The only other avenue of improvement will be to do away with color filters (since using them implies that we throw away 67% of incident photons) and sense photon energy (wavelength) directly. That, IMO, is quite a ways off. Though it is easily done in the lab, constructing megapixel arrays with the ability to both count individual photon arrivals and track the energy distributions is pretty far in the future.
For reference, the Foveon technology actually does skirt some of the color filter issue. Using the penetration depth to differentiate means that you effectively can use all the photons. The results to date, however, have been underwhelming in some respects.
All these advances will, eventually, result in improved sensing for high-end imaging, to the benefit of us all. The next ten years will be a very interesting time to be in photography. We had all better keep an adequate budget set aside for body upgrades :-P |
I wonder how many of us actually understand that explanation?
R.
edit to add: FINE explanation, but way up there in the stratosphere. The scary thing, to me, is that I actually DO understand it...
Message edited by author 2006-01-18 15:29:31. |
|
|
|
01/18/2006 10:37:59 AM · #14 |
Apparently this new technology results in a sensitivity about 2000 times greater than current cmos and ccd technologies. As Kirbic indicated, the noise due to thermal electrons and random photons will increase significantly.
A few cameras have attemted to reduce the photon loss caused by RGGB Bayer masks by using CMYG masks. It seems that RGB masks use two layers of CYM colors to make up the RGB colors, resulting greater light loss than CYM masks alone. My old Canon Pro 90 IS uses a CYMG mask and is more sensitive than my other cameras, also has very low noise and excellent image quality. Too bad it is only 2.7 megapixels! |
|
|
|
01/18/2006 10:40:52 AM · #15 |
I understood this part OK:
Originally posted by kirbic:
The answer is, I don't think, until I've had my coffee. ;-) ... |
|
|
Home -
Challenges -
Community -
League -
Photos -
Cameras -
Lenses -
Learn -
Help -
Terms of Use -
Privacy -
Top ^
DPChallenge, and website content and design, Copyright © 2001-2025 Challenging Technologies, LLC.
All digital photo copyrights belong to the photographers and may not be used without permission.
Current Server Time: 04/11/2025 07:31:53 AM EDT.
