Pixels and Resolution are not the same thing.

Before the large sensor resolution most professional video cameras used 3 sensors, one each for red, green and blue. And each of those sensors normally had as many pixels as the resolution of the recording format. So you had enough pixels in each colour for full resolution in each colour.

Then along came large sensor cameras where the only way to make it work was by using a single sensor (the optical prism would be too big to accomodate any existing lens system). So now you have to have all your pixels on one sensor divided up between red, green and blue.

Almost all of camera manufacturers ignored the inconvenient truth that a colour sensor with 4K of pixels won’t deliver 4K of resolution.  We were sold these new 4K cameras. But the 4K doesn’t mean 4K resolution, it means 4K of pixels. To be fair to the manufactures, they didn’t claim 4K resolution, but they were also quite happy to let end users think that that’s what the 4K meant.

My reason for writing about this topic again is because I just had someone on my facebook feed discussing how wonderful it was to be shooting at 6K with a new camera as this would give lots of space for reframing for 4K. 

The nature of what he wrote – “shooting at 6K” –  implies shooting at 6K resolution. But he isn’t, his 6K sensor is probably delivering around 4K resolution and he won’t have any room for reframing if he wants to end up with a 4K resolution final image. Now again, in the name of fairness, shooting with 6K of pixels is going to be better than shooting with 4K of pixels if you do choose to reframe. But we really, really need to be careful about how we use terms like 4K or 6K. What do we really mean, what are we really talking about. Because the more we muddle pixels with resolution the less clear it will be what we are actually recording. Eventually no one will really understand that the two are different and the differences really do matter.

4 thoughts on “Pixels and Resolution are not the same thing.”

  1. Even that misses that resolution as a system includes a lens, and an appropriate measurement metric, tv lines of resolution exists.

  2. “Then along came large sensor cameras where the only way to make it work was by using a single sensor (the optical prism would be too big to accomodate any existing lens system). ”

    That’s not the reason. You can easily cram 37.5K of sensors into a full frame sensor (12.5K per colour which is what you need for a true 3 colour 4K sensor in traditional 4 x 3 format) – the problem isn’t size, it’s can you process that image in real time @ 4K and is the trade off in noise, heat and reduced frame rates worth it. I think we’re still answering that one as 8K cameras have 48 megapixels.

    1. Size was an issue as you can’t have a 3 chip system at super 35mm. And at the time there were no 48 megapixel video sensors.

      Yes you are correct that you can cram in a lot of pixels on that single sensor, but you sacrifice dynamic range, noise, etc etc. And the manufacturers would still market it as an 8K camera, which it is not. It is a camera with 8K of pixels.

      You don’t need 12.5K of pixels per colour for 4K resolution. 6K per colour would be comfortably oversampled.

  3. These people don’t understand Bayer pattern resolution loss. Some of these guys are the same ones that say; “Eww….8k 4:2:0 sub sampled?…oh no, that’s terrible for chroma keying” LOL

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.