8K x 2K

Time for some seriously, non-official, pre-announcement, highly speculative math excercises. Many astute readers have been emailing or posting the big question, “How can the new Sony CineAlta Camera be 16:9 and greater than 4K when its pixel count is 8K x 2K? 8 x 2 is like a super wide-screen 4:1 format, isn’t it?” No.

After spending time with some leading non-Sony academic and technical gurus over whiteboards, computers and Sake in Asakasa, here are some thoughts until the mysteries are revealed at NAB.

The sensor is 27 mm diagonal: slightly wider and a bit shorter than the F3 sensor. We don’t have the dimensions yet. It’s similar in shape to the F35.  (The F35 is CCD, not CMOS.)

The CMOS sensor of the new camera is 20.4 Megapixels, 8K x 2K. Output is 16 bit RAW, 16:8:8. Normal frame rate is 1-72 fps and High frame rate is 1-120 fps. There will probably be a rotating mechanical shutter to handle this. The pixels on the new CineAlta Camera are arranged diagonally, diamond shaped, and have a unit cell size perhaps 1/5th the size of the F3 pixels.

Native data rate was estimated at 20 Gbps, and it can be compressed 4:1 for the 5 Gbps sustained data rate of the SR Memory Cards. A 1 TB SR Memory Card will store about 1 hour of “footage” at 24 fps.

How can they do this? How is 8K x 2K possible?  What if the horizontal pixels were sliced, divided or multisampled? In other words, if the pixels were rectangular, not square…taller than wide…

This math exercise was done mainly to try to explain that 8K x 2K is the imager resolution, not the format size. Anyway, I’m more interested in how good it looked–wide latitude, gorgeous color, rich blacks, high ISO. A few people are calling this camera the F65. Others are calling it 8x2K. I was calling it Altissima. We’ll have to wait until NAB to learn more…


Leave a Comment