Saturday, July 21, 2018

e2v on CCD vs CMOS Sensors

AZO Materials publishes Teledyne e2v article "The Development of CMOS Image Sensors" with a table comparing CCD and CMOS sensors. Although I do not agree with some of the statements in the table, here it is:

Signal from pixelElectron packetVoltage
Signal from chipAnalog VoltageBits (digital)
Readout noiselowLower at equivalent frame rate
Fill factorHighModerate or low
Photo-ResponseModerate to highModerate to high
Dynamic RangeHighModerate to high
UniformityHighSlightly Lower
Power consumptionModerate to highLow to moderate
ShutteringFast, efficientFast, efficient
SpeedModerate to HighHigher
Anti-bloomingHigh to noneHigh, always
Image ArtefactSmearing, charge transfer inefficiencyFPN, Motion (ERS), PLS
Biasing and ClockingMultiple, higher voltageSingle, low-voltage
System ComplexityHighLow
Sensor ComplexityLowHigh
Relative R&D costLowerLower or Higher depending on series


  1. This article (and table) must have been prepared by Teledyne's marketing department. Technically speaking, it has more holes than Swiss cheese and should be filed under "disinformation" which is better known as fake news these days. Get just enough facts right to make the article look legitimate, but insert enough disinformation to confuse the reader and sow doubts. Shame on you Teledyne. You impugn your own good image sensor engineers by publishing such material.

    1. The big thing I noticed is fill factor. BSI CMOS devices have as good or better FF than CCDs.

  2. I bet CMOS-TDI will have much worse antiblooming performance than CCD.

  3. Comparing apples and oranges! Whereas there are a plenty of different apples and oranges! Frame Transfer CCD, Full Frame CCD, Interline Transfer CCD, EMCCD, CMOS with analog output, sCMOS, ... Each of them has it's special features and drawbacks!
    And whoever has implemented a CCD sensor into a camera knows very well about the sensor's complexity!


All comments are moderated to avoid spam and personal attacks.