Tuesday, August 21, 2012

Sony Announces "Exmor RS" Stacked Products

Sony announces the commercialization of "Exmor RS", the CMOS image sensor incorporating a unique, newly-developed "stacked structure". Shipments will commence in October. Sony is introducing three models of the “Exmor RS,” stacked CMOS image sensor with 1.12um pixels, for use in smartphones and tablets with three corresponding camera modules incorporating these sensors:

Above: Camera modules (left to right): the IU135F3-Z, IU134F9-Z and IUS014F-Z
Below: "Exmor RS" stacked CMOS sensors (left to right): IMX135, IMX134 and ISX014

IMX135, the 13.13MP (eff) 1/3.06-inch sensor model and IMX134, the 8.08MP (eff) 1/4-inch sensor feature "RGBW coding" and "HDR movie" function. The "RGBW coding" has W (white) pixels in addition to conventional RGB pixels, and leveraging Sony’s proprietary device technology and signal processing to improve low-light sensitivity without compromising its high resolution. "HDR movie" function enables two different exposure conditions to be configured within a single screen when shooting, and seamlessly performs appropriate image processing to generate optimal images with a wide dynamic range.

The other “Exmor RS” sensor is the ISX014, a 1/4-inch model with 8.08MP (eff), which has a built-in camera signal processing function that, apparently, does not support "RGBW coding" and HDR.

Sony will also bring to market three compact AF camera modules equipped with newly-designed lens optimized for the 1.12mm pixels: the IU135F3-Z, IU134F9-Z and IUS014F-Z. The IU135F3-Z module incorporates fast F2.2 lens. The IU134F9-Z (W:8.5 x D:8.5 x H:4.2mm, excluding FlexPCB) is thin and compact.

The release shipment schedules and prices are in the table below:


Sample image captured at 10 lux illumination


A brief stacked sensors spec comparison:


The new camera module comparison:

15 comments:

  1. Well... according to previous iPhone5 camera rumor, the spec of IMX 135 looks almost the same in KGI researcher's report published in early June. Would it imply that iPhone5 adapted the new Sony stacked CIS with up to 13MP camera? I can't wait if Apple really do so!!!

    ReplyDelete
  2. What does "sensor saturation signal" mean? Max voltage swing at FD?

    ReplyDelete
  3. Judging from core Vdd of 1.05V, N65 would be used for circuit.
    I'd like to see N28 stacked circuit.

    Which technology becomes de facto standard for future CIS, SoC or 3DIC?
    (^^)

    ReplyDelete
  4. I am afraid it's too late for Iphone 5 whose production is well underway...

    ReplyDelete
  5. sony may have made the 1/4 inch 8mp being made available in october to other customers available to apple a few months earlier as part of a special deal so apple could be first to market with sony's cutting edge stacked bsi product. what a way for sony to debut its new product in the most sought after smartphone. good for apple and sony. funny how we used to refer to women as being stacked, and now we use the term for cmos image sensors. lol

    ReplyDelete
    Replies
    1. I doubt so. Apple needs millions of CMOS chips for its first wave iphone 5 launch, which started in July already! It's just not possible for Sony to ramp up such a huge volume several month before the official launch of stacked CMOS.

      It seems Sony has missed a critical step.

      Delete
    2. I believe it is almost certain that the apple will have the stacked sensor. This is standard practice for Apple with technology partners as history shows us clearly, that is, to be the first product to have significant new technology from suppliers. This is part of Apples core product strategy.

      Delete
  6. What is stacked about these 'stacked' sensors? Is it just marketing lingo for BSI?

    ReplyDelete
    Replies
    1. Not just a lingo. It's a sensor chip stacked with an processing one, see figure in the Sony PR link.

      Delete
    2. It does not seem to offer significant advantages over current BSI because the pixel structure has not changed. It is definitely not as revolutionary as BSI, more like a gimmick to me.

      Delete
    3. Moving out transistors from the pixel is a huge advantage! My doubt is: Is the via between the 2 silicon layers a part of the FD capacitance? How can they maintain a low PRNU?

      Delete
    4. See http://spie.org/x648.html?product_id=873610 for an example of what stacking can do.

      Delete
    5. dark current 1e-/s?

      Delete
  7. Does anyone know how to read Sony's "sensitivity" and "sensor saturation signal" values?

    Alternatively, would anyone like to guess QE, electrons of readout noise when gain is set so that max DN is somewhere around full well, and electrons in that full well? I've never been able to get these numbers from Sony.

    ReplyDelete
    Replies
    1. Hello Iain McClatchie,

      Just to reinforce the importance of your question, I have never understood how to interpret the "Saturation Signal" and "Sensitivity" data provided by Sony in all imaging sensors datasheets. They characterize these parameters in mV.
      In my opinion it doesn't make sense. They should use "electrons" (for saturation signal) and "uV/electron" (for sensitivity).

      Sony has just shared the result of saturation in mV but they don't tell us the ammount of uV generated by one electron. Of course that this signal is amplified.

      So, if someone has an additional information, please share with us!

      Thank you!

      Delete

All comments are moderated to avoid spam.