Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
05-16-2008, 09:23 AM   #16
Veteran Member




Join Date: Apr 2008
Location: Minnesota
Photos: Gallery
Posts: 1,530
You do not even have to go that far. Here is some text refereced in the Korean patent:

"In a case where a particular pixel reaches a saturation level, an output voltage of the pixel may not be normally measured depending on the quantity of light, i.e., expressed as the quantity of charges, received in the pixel. To avoid this, the pixel is to be reset. For each pixel, there are two types of reset operation. In a first type of reset operation called a global reset operation, a pixel is reset upon initialization of a new frame. In this case, all pixels are concurrently recharged. In a second type of reset operation called a local reset operation, each pixel is recharged by triggering. In this case, the recharging states of each pixel may vary according to the quantity of light received in each pixel. In other words, some pixels may be recharged and some may not be recharged. In addition, some pixels may be recharged fast and some may be recharged slowly. The number of times the pixel has been reset (simply referred to hereinafter as "the number of resets") and the residual light quantity of the pixel are obtained, in order to measure the quantity of light received in the pixel even in an area where the pixel is saturated."

From the looks of it they are saying that each pixel has a known upper limit. Once a 'local reset operation' is done they will keep appending that number with each subsequent local reset. More from the patent:

"Quantity of Light Received in Pixel=SV*NRC+LIR

[0033]where SV stands for a saturation value representing the quantity of light required for a particular pixel to reach a saturation level, NRC stands for a total number of resets, and LIR stands for a residual light quantity representing the quantity of light, i.e., the quantity of charges, remaining after the pixel is finally reset. These values may be measured for a single pixel during a period of one frame.
"

05-16-2008, 10:01 AM   #17
Veteran Member




Join Date: Jan 2007
Location: Northern California
Posts: 475
Original Poster
QuoteOriginally posted by MrApollinax Quote
You do not even have to go that far. Here is some text refereced in the Korean patent:

"In a case where a particular pixel reaches a saturation level, an output voltage of the pixel may not be normally measured depending on the quantity of light, i.e., expressed as the quantity of charges, received in the pixel. To avoid this, the pixel is to be reset. For each pixel, there are two types of reset operation. In a first type of reset operation called a global reset operation, a pixel is reset upon initialization of a new frame. In this case, all pixels are concurrently recharged. In a second type of reset operation called a local reset operation, each pixel is recharged by triggering. In this case, the recharging states of each pixel may vary according to the quantity of light received in each pixel. In other words, some pixels may be recharged and some may not be recharged. In addition, some pixels may be recharged fast and some may be recharged slowly. The number of times the pixel has been reset (simply referred to hereinafter as "the number of resets") and the residual light quantity of the pixel are obtained, in order to measure the quantity of light received in the pixel even in an area where the pixel is saturated."

From the looks of it they are saying that each pixel has a known upper limit. Once a 'local reset operation' is done they will keep appending that number with each subsequent local reset. More from the patent:

"Quantity of Light Received in Pixel=SV*NRC+LIR

[0033]where SV stands for a saturation value representing the quantity of light required for a particular pixel to reach a saturation level, NRC stands for a total number of resets, and LIR stands for a residual light quantity representing the quantity of light, i.e., the quantity of charges, remaining after the pixel is finally reset. These values may be measured for a single pixel during a period of one frame.
"

I seem to recall mention of prior art in the application, so it is not a new idea. However, this is not going to work with CCD sensors as they cannot be read out pixel by pixel, and the circuitry they talk about cannot be incorporated into the area around CCD pixels, so this is related to a CMOS sensor, or any sensor that can be read out and reset pixel by pixel.

In this case, the trick is that they figured out how to build a trigger into the pixel that is activated when the well is full and resets the pixel and also sets a counter. This then becomes a self-resetting pixel, which is the main part of the patent, I think.

I presume that Samsung knows the full well capacity of the sensors they make and can easily test each one to quantify that even more accurately, so at the end of the day you could have a sensor that is accurately mapped as to the actual full well capacity of each pixel, and a table that is then used to lookup the value to add to the total for each time the pixel was filled up.

There may be some technical issues or limitations with this idea, but if it even increases highlight DR of the sensor by 50% it would blow the socks off anything else on the market.

Ray
05-16-2008, 10:25 AM   #18
Veteran Member




Join Date: Feb 2008
Location: Chicagoland
Posts: 338
QuoteOriginally posted by Ray Pulley Quote
There may be some technical issues or limitations with this idea, but if it even increases highlight DR of the sensor by 50% it would blow the socks off anything else on the market.

Ray
No kidding. It would effectively give you unlimited DR (dependent on the number of bits allocated for the counter) You could just expose for the darkest object you want to grab and the highlights would be there no matter what. Great for still tripod shots where you can let it have all the time it needs. Even for high shutter speeds it would allow them to be much more aggressive with exposing to the right since it would be effectively impossible to blow a highlight.
05-16-2008, 11:19 AM   #19
Veteran Member




Join Date: Apr 2008
Location: Minnesota
Photos: Gallery
Posts: 1,530
This has quickly turned into the nerdiest thread I have seen on this board .... I now have an overwhleming urge to go program functions into my old graphing calculator....

05-16-2008, 02:00 PM   #20
Veteran Member




Join Date: Jan 2007
Location: Northern California
Posts: 475
Original Poster
QuoteOriginally posted by MrApollinax Quote
This has quickly turned into the nerdiest thread I have seen on this board .... I now have an overwhleming urge to go program functions into my old graphing calculator....
Careful!

You might grow hair on your palms doing that...

Ray
05-16-2008, 03:26 PM   #21
Inactive Account




Join Date: Jul 2007
Location: London
Posts: 1,067
Hi Geekybiker

Re your assessment:

QuoteQuote:
It would effectively give you unlimited DR......since it would be effectively impossible to blow a highlight.
Could THIS perhaps be the "bomb" that *Benjikan was so often alluding to (NDA's not withstanding) in some of his previous posts.....??

Best regards
Richard
05-16-2008, 03:51 PM   #22
Veteran Member




Join Date: Feb 2008
Location: Chicagoland
Posts: 338
QuoteOriginally posted by Confused Quote
Hi Geekybiker

Re your assessment:



Could THIS perhaps be the "bomb" that *Benjikan was so often alluding to (NDA's not withstanding) in some of his previous posts.....??

Best regards
Richard
We can hope. I know alot of people would jump for pentax/samsung if that tech works.
05-16-2008, 06:48 PM   #23
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by Ray Pulley Quote
I presume that Samsung knows the full well capacity of the sensors they make and can easily test each one to quantify that even more accurately, so at the end of the day you could have a sensor that is accurately
The full well capacity at ISO100 is 16000 or 32000 electrons (for a K20D sensor cell, if I remember correctly).

Actually, what you do here is to create a virtual full well capacity of an unlimited size. If you, at the same time, reduce the physical full well capacity to one, you end up with a device counting individual photons (well, nearly if quantum efficiency is high).

All these ideas are well known and have been patented, obviously by multiple parties already. What remains to be solved is its implementation, in terms of size and heat dissipation. The patent says nothing about this.

As soon as the real problems are solved (rather than lawyers playing their games), I am sure every manufacturer will jump onto the bandwagon and Samsung won't stay out.

Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
application, counter, patent, pentax news, pentax rumors, pixel, range, samsung, sensor
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
HOYA Patent Application Numbers for the Past 8 Years Miserere Pentax News and Rumors 11 05-27-2010 10:32 AM
Live view - useful application xGene Pentax DSLR Discussion 7 09-12-2008 10:22 AM
Note taking application clawhammer Digital Processing, Software, and Printing 4 08-07-2008 08:31 PM
DA 10-17: First Application smile-123 Pentax SLR Lens Discussion 5 02-25-2008 11:32 AM



All times are GMT -7. The time now is 10:59 PM. | See also: NikonForums.com, part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top