# Is it true that Canon sensors have half the number of colors than Nikon, Fujifilm or Sony sensors?

Is it true that Canon sensors have half the number of colors than Nikon, Fujifilm or Sony sensors? I read this on Quora.

• Abraham Lincoln once said: Do not trust everything you read on the internet Commented Aug 24, 2021 at 5:13
• Each bit multiplies the number of colors by 8 (2^3, because there are 3 colors). Since most people are still using 8-bit (or lower) displays, 10-14-bits is more than enough. Technology is always advancing, and companies tend to play leapfrog with their competitors. So what was true yesterday isn't necessarily true today. Commented Aug 24, 2021 at 5:37
• @xiota You comment doesn't answer the question. Do you mean each pixel instead of each bit? Commented Aug 24, 2021 at 6:03
• @xiota means bits. With N bits you can encode 2^N values, in other words measure 2^N values in a "sensel". Sensors are between 12 to 16 bits. With 12 bits you can encode 4096 levels (2¹²), with 14, 16384 (2¹⁴), and with 16, 65536 (2¹⁶). So a sensor with more bits per "sensel" can theoretically give more information. However with more bits the additional bits are mostly in the noise area, so cost you processing power without adding much information. BTW if you add 1 bit to the sensor, you multiply the number of final colors by 8 (because you have doubled each of the three channels), not by 2. Commented Aug 24, 2021 at 6:58
• Colors are represented in bits. Each extra bit doubles the number of values that can be represented. Adding a bit to each color adds three bits total because there are three colors. An 8-bit color representation supports 256 values for each color primary. About 16.7M colors total. 10-bits = 1.1B colors. 12-bits = 68B colors. 14-bits = 4.4T colors. If you're using JPG, you're stuck with 8-bits. If you're using RAW with any reasonably modern camera, you have more than enough colors, regardless of the camera manufacturer. Commented Aug 24, 2021 at 6:58