[K(s)evin]

So, if a digital full frame sensor is 36mm x 24mm, an APS-C crop is 2/3 of that, or 24mm x 16mm. Then how come on the K-1 36mp sensor when you use the crop setting you only get ~15mp, instead of 24mp?

[Adam]

So, if a digital full frame sensor is 36mm x 24mm, an APS-C crop is 2/3 of that, or 24mm x 16mm. Then how come on the K-1 36mp sensor when you use the crop setting you only get ~15mp, instead of 24mp?

The APS-C size (that you listed above) is just under half the area of full frame, hence you get just under half the megapixels.

[TER-OR]

Area is tricky - think of the size difference between a 12" and 16" pizza.

(volume is worse for me, I'm terrible at estimating that)

[sjwoodworth]

Area of a 2cm by 2cm square is 4cm^2. Area of a 4cm by 4cm square is 16cm^2. Pixels are like cm^2, not cm.

[Bob 256]

The pixel count is proportional to the square of the sensor size (assuming compared sensors have the same proportions and pixel density). To increase the sensor size by a factor of two requires 4 times as many pixels. Another way to look at it is the the sensor size is proportional to the square root of the number of pixels. Doubling the number of pixels will not get you a sensor with double the linear dimensions.

The best comparison is the K-1 and the K-01, the latter of which is APS-C (the former being FF). These have the same sensor density so it's almost like you took a cookie cutter to the K-1 sensor to get the smaller K-01 sensor (these are 36.4 Mp versus 16.28 Mp). If you divide 16.28 by 36.4 and take the square root, you get 0.669 which is pretty close to 2/3. You can shoot a pic on the K-1 and K-01 using the same lens, and if the K-1 pic is cropped to the APS-C size, you almost can't tell the difference pixel-wise (same number of pixels in each).

[K(s)evin]

The pixel count is proportional to the square of the sensor size (assuming compared sensors have the same proportions and pixel density). To increase the sensor size by a factor of two requires 4 times as many pixels. Another way to look at it is the the sensor size is proportional to the square root of the number of pixels. Doubling the number of pixels will not get you a sensor with double the linear dimensions.

The best comparison is the K-1 and the K-01, the latter of which is APS-C (the former being FF). These have the same sensor density so it's almost like you took a cookie cutter to the K-1 sensor to get the smaller K-01 sensor (these are 36.4 Mp versus 16.28 Mp). If you divide 16.28 by 36.4 and take the square root, you get 0.669 which is pretty close to 2/3. You can shoot a pic on the K-1 and K-01 using the same lens, and if the K-1 pic is cropped to the APS-C size, you almost can't tell the difference pixel-wise (same number of pixels in each).

I think that I think I know what you're talking about. I may need to reread it a few times.

Math is not, and never was, my strong suit. Thankful that there are folks out there that can figure it out, though!

[ewig]

So, if a digital full frame sensor is 36mm x 24mm, an APS-C crop is 2/3 of that, or 24mm x 16mm. Then how come on the K-1 36mp sensor when you use the crop setting you only get ~15mp, instead of 24mp?

its not the matter about the wide from the sensor to produce the picture , but the algorithym also must divided by half factor

[rdenney]

24 x 36 = 864 square millimeters

15 x 24 = 360 square millimeters

36mp x 360/864 = 15mp

Rick “calculate the area” Denney

[aslyfox]

math

[jhaji]

(16mm x 24mm)/(24mm x 36mm) * 36MP = 16MP

[rogerstg]

So, if a digital full frame sensor is 36mm x 24mm, an APS-C crop is 2/3 of that, or 24mm x 16mm. Then how come on the K-1 36mp sensor when you use the crop setting you only get ~15mp, instead of 24mp?

24 x 36 = 864 square millimeters

15 x 24 = 360 square millimeters

36mp x 360/864 = 15mp

Rick “calculate the area” Denney

To take Rick's calculation one step further, the APS-C area is only ~42% of the full frame sensor, not the 2/3 (67%) as you thought.

[rdenney]

Or, square that 2/3, since we are talking about area.

2/3x2/3 = 4/9, or 44%.

Rick “area squares, volume cubes” Denney