0

This question already has an answer here:

For example, the D3300 is advertised to have 24.2 megapixels, but its highest camera resolution in all types of settings including raw and full manual, the output resolution becomes 6000×4000 instead of 6016×4000 like the D3200. I was wondering if those 16 pixels are used for some sort of physical noise reduction mechanism or are just simply cropped off for whatever reason.

I've also talked with Nikon support and they don't know why that happens. They implied that the image is simply cropped, but they were unsure why it does this.

marked as duplicate by mattdm, Hugo, John Cavan Jul 4 '15 at 13:11

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

  • @DrMoishePippik No, DSLRs have dedicated metering sensors, metering via liveview uses regular pixels. Some mirrorless cameras (and a couple of Canon DSLRs) have phase detection AF pixels on the main imaging sensor but those are still used to produce the image. That article is about a very specific sensor design used only by Fuji. – Matt Grum Jul 2 '15 at 7:56
  • BTW, rawdigger can recover the full image even if the reported resolution is less. There are sometimes extra pixels cropped off implicitly. I also recall some other program that woukd reset the raw file. – JDługosz Jul 2 '15 at 17:56
5

Because of the way color is created from single monochromatic luminance values created by each pixel of a CMOS sensor, the pixels on the edge are needed to allow interpolation of the Red, Green, and Blue values for the next row(s) of pixels in. To have enough information to compute RGB values for each pixel on the edge, you would need an additional row of pixels outside of it.

Some sensors also have specific pixels that are "masked" to not allow any light into the well. The readings from these pixels are used to create a noise floor. Any signal recorded from these masked pixels is noise, since no light is allowed into the pixel well to create an electrical signal. The same amount of signal is then subtracted form the rest of the pixels on the sensor to help in noise reduction.

  • Ohhh, that would make much more sense now. I think the D3300 uses the first option since there is no pixel difference when noise reduction is turned off. But how come the D3200 comes with 16 extra pixels instead then? – Muhatashim Jul 2 '15 at 6:11
  • @Muhatashim The edge pixels are permentantly masked, so there will be no difference when turning noise reduction on or off. The masked area is also probably larger than 16 pixels, I suspect the D3300 simply has a marginally larger masked area. 16 pixels on a 6,000 pixel wide image are fairly inconsequential anyway... – Matt Grum Jul 2 '15 at 8:31
  • 4
    "Inconsequential" is an understatement. Print that image on a 20"×30" poster, and that row of pixels is about 0.08" missing. I bet you can't even trim your paper that precisely. – mattdm Jul 2 '15 at 19:32

Not the answer you're looking for? Browse other questions tagged or ask your own question.