DJI needs competition to push them, especially on sensors and processors, at the same price points.
Pixel binning IS exactly sub sampling. So, subsampling and pixel binning are both correct statements. In other words, "sub sampling" is a generic term for "pixel binning".
Pixel binning IS exactly sub sampling. So, subsampling and pixel binning are both correct statements. In other words, "sub sampling" is a generic term for "pixel binning".
Even line skipping falls inside the "sub sampling" catagory. (Its the worst form of sub sampling)
What we want is full pixel read out (oversampled 5k readout, 5k image assembly and then final 5k downscale to 4k)
Sony is the full pixel readout master! Panasonic does it too. (By way of using Sony IMX272 and other sensors)
We know the M2P has the Sony full pixel readout mode "available"....its just turned off today. (Because of processing or heat or Phantom5 protection)
Cliff Totten
You article is good. But as I said earlier, their example of "sub sampling" is the "line skipping" technique. (As I said before, worst kind of sub sampling)Same category yes, but not the same thing. Here is a good illustration for those interested:
Sometimes less means more – Subsampling, binning, or scaler - IDS Imaging Development Systems GmbH
And yes, I am hoping whatever the next iteration is, we get proper full sensor readout and maybe even 4K/60P. The Sony RX10 IV, for example, can do 30 minutes of 4K/30P with full sensor readout with a similar 1" sensor - no reason DJI shouldn't be able to do at least that (or longer) with a bigger, actively cooled package to boot. Even if they reserve the 60P for the Phantom line I'll take that compromise if they can give us proper sensor readout. Based on how easily it's being done in smaller, passively cooled packages, I'm guessing it's being limited artificially in the M2P for Phantom protection, but it's impossible to say for sure.
We use essential cookies to make this site work, and optional cookies to enhance your experience.