Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

16:8:8 Color Sampling?

Sony put up a website for the F65. Perhaps everyone can have a better understanding now.

http://pro.sony.com/bbsccms/ext/digitalcinematography/f65.html

For the little information I see there I'm pretty excited about this product I may or may not ever have a chance of actually using I'm pretty excited about this camera, I watched the video too I think I'm drinking the kool aid on this one.

I see that they have 16 bit RAW support now so I would be interested in the workflow for that and if it will be compatible with the new yet to be released Final Cut Pro.
 
There's zero mention of the funny chroma sampling ratio at the Sony display - the video looked version much like our early Red One demo reels.

Graeme
 
So, say they have 4096x2160 green pixels, and therefore as they're bayer (although rotated) they have 4096x2160/2 reds and 4096x2160/2 blues, right? That gives us 17694720 in total, which seems right for their sensor. And we know a bayer pattern is just a bayer pattern, so even if they did a full demosaic, which they won't, but let's assume, then that's equivalent to a 5792 x 3054 array, which if treated as a proper bayer pattern would give a measured resolution of about 4.6k - pretty good, but a fair bit short of their claimed 8k.

Now let's think about what they're actually doing - they're leaving the greens as is, as they're diagonal. They basically have a 4096x216 array of greens that we can think of as a single colour chip (with some holes for red and blue) that can measure no more than 4k resolution or else it will show the false detail of aliasing, or it may measure less, depending on how they set the OLPF. Either way, the 8k claim is not supported by the math or the sensor.

What I want to see is it's zone plate performance.....

Graeme
 
measured resolution of about 4.6k - pretty good, but a fair bit short of their claimed 8k.


What I want to see is it's zone plate performance.....

Graeme

Well to be fair they are calling it a 4K camera to suit the hundreds of 4K projectors they have in theaters.

---------

At what point does zone plate performance really matter to anyone but pixel peepers? Doesn't the contrast of an image have a lot more to do with sharpness than actual resolution itself? Or does this all change when you wanna project on IMAX? How small do you think pixels can be before you start to reduce the quality of the image?? Their must be a optimum pixel size and something tells me this is why ARRI designed a 2K camera for a 35mm sized sensor... 4K would suit a 65mm sized sensor
 
I sat down with my little understanding of this all to try and wrap my brain around this and the best I could come up with to spin an 8k sensor was to say, okay 4k of green across plus 4k of red + blue across = 8k of horizontal resolution.
An 8k sensor is apparently much more than 20 mega pixels but what do I know and for that matter what do I care.
Apparently the NHK thing pulled off 8k by taking 4 ccds each 3840 × 2048 2 Green, 1 blue and 1 red so it could be a similar idea to that except they've crammed the pixels onto 1 chip instead of 4.
 
NHK make a pseudo-bayer from 4 * 4K chips. It's pretty clever, if a rather wild way to do it! Again what we see is NHK are validating that higher resolution is the way forwards.

Sony are indeed calling their 4K camera a 4K camera - which is most probably correct and will be confirmed once we can measure one. However, the 8K sensor comment I read from their press is a little too much for me.

Graeme
 
I don't see how the 4k can be over-sampled when there's only 4k worth of green to begin with.... Methinks they need to check their math.

Graeme
 
Back
Top