Huge Sony sensor advance heralds amazing video features - 6K, and 1080p at up to 16,000fps

[Andrew Reid]

Sony Active Pixel Color Sampling is coming and it is completely different to anything else. It solves the low light performance issues with global shutter sensors, makes ultra high frame rates possible and scraps the traditional bayer RGB filter altogether.

Read the full article here

[maxotics]

Any idea how it "scraps the traditional bayer RGB filter"?  Manufacturers have been trying for decades.  Forgetting everything else, that alone is revolutionary if done without extra noise in one of the color channels.  

 

As for what camera they'll put it in.  Maybe Nikon will surprise us ;)  

[Nikkor]

The color filter shifts in front of the sensor at highspeed. Pretty impressive stuff.

[pablogrollan]

Any idea how it "scraps the traditional bayer RGB filter"?  Manufacturers have been trying for decades.  Forgetting everything else, that alone is revolutionary if done without extra noise in one of the color channels.  

 

As for what camera they'll put it in.  Maybe Nikon will surprise us ;)

 

It seems the technology has been designed with video in mind and the leaked specs refer to a video sensor, so my guess is it would appear on a traditional broadcast camera first with a bigger-than-usual sensor (1.5 inch) and later on in s35 cameras and maybe photo cameras.

We still don't know much -such us the body size requirements to keep heat under control, etc- but moving away from bayer sensors sounds like an option for the future...

[PJ McKenna]

If they are rapidly alternating colour filters on from of the sensor then is there not the risk of colour bleeding across edges when panning?

There was a Canon patent some time ago that kind-of sounds like this, but it was related to flipping colour filters in a 2-filter DSLR light meter.

[MattH]

Pretty vague information. The picture of the document doesn't even show the right side of the text.   I doubt they are moving a filter across the sensor, as in the whole fucking sensor.  It will more likely happen on a pixel level.  The filters will probably be electronic.  I can't see having 4 million miniscule moving parts to be very easy to make or being very reliable.  What is interesting is that it talks about effective pixels, not photosites, and each effective pixel area is about twice as wide as it is tall. It will be interesting when more info comes to light.

[jgharding]

Interesting! Great to see new non-Bayer sensors.

 

I love the image of Sigma Foveon, glad someone else is going for a similar full RGB technology.

[schnorgie]

Thanks for posting, and the blog. It's articles like these that illustrate the sheer folly of "upgrading" to this year's model camera. Why invest thousands of $$$ this year, when something cheaper and logarithmically better will be available next year, or next week? AND, you never know WHAT will happen, after you click "add to cart"..... 

 

 

....p.s., the first line from Andrew's drone post today...."Well, DJI Phantom 2 is toast!". ;-)

[Nick Hughes]

Thanks for posting, and the blog. It's articles like these that illustrate the sheer folly of "upgrading" to this year's model camera. Why invest thousands of $$$ this year, when something cheaper and logarithmically better will be available next year, or next week? AND, you never know WHAT will happen, after you click "add to cart"..... 

 

Yeah that's why I'm still filming on mini-dv. No point to upgrade if something better is just going to be released eventually.

[lafilm]

"Yeah that's why I'm still filming on mini-dv. No point to upgrade if something better is just going to be released eventually".

 

:D  :D  :D

[noirist]

How can the sensor output 6k video (or even 4k video) when it only has 4.44M active pixels?  That's more like 1080p-class resolution. The only way I can see it outputting 4k video is if it upscales its 2166 vertical lines to 3180, which can't help but look nasty.

[Tzedekh]

How can the sensor output 6k video (or even 4k video) when it only has 4.44M active pixels?  That's more like 1080p-class resolution. The only way I can see it outputting 4k video is if it upscales its 2166 vertical lines to 3180, which can't help but look nasty.

The only thing I can think of is that there are three layers of pixels, each of which has 4.44 MP. The 2,052 x 2,166 array is then cropped to 2,048 x 2,160. Because the pixels are twice as wide as they are tall, the aspect ratio is the 1.89:1 DCI standard. The additional horizontal resolution may be achieved through something like pixel-shifting.

[Damphousse]

Thanks for posting, and the blog. It's articles like these that illustrate the sheer folly of "upgrading" to this year's model camera. Why invest thousands of $$$ this year, when something cheaper and logarithmically better will be available next year, or next week? AND, you never know WHAT will happen, after you click "add to cart"..... 

 

 

You can't buy a patent filing.  So many technologies have been patented over the years and promised to revolutionize our world and then nothing came of it... or something much less impressive actually came out of the factory.  I think we will continue to see the same situation we have today for quite some time.  There will be a lot of interesting cameras but under a particular price point you will have to make trade offs.  Someone mentioned Fovean.  It was a great idea and when it was announced people swore it would take over the world.  When was the last time you've actually seen someone use one of their cameras?  I would be shocked if this sensor functions in the way you guys imagine it will.  It's either going to have an outrageous price tag or multiple issues.  That doesn't mean it won't be a worthy addition but let's wait and actually see some video before declaring our gear obsolete.

 

And by the way that's the reason I bought my BMPCC for $500 instead of $1,000.  It doesn't matter what comes out next year.  I will still be happy with my purchase.

[GMaximus]

If they are rapidly alternating colour filters on from of the sensor then is there not the risk of colour bleeding across edges when panning?

There was a Canon patent some time ago that kind-of sounds like this, but it was related to flipping colour filters in a 2-filter DSLR light meter.

Probably the only way to avoid that is a really fast filter, like 16kHz one.

 

 

How can the sensor output 6k video (or even 4k video) when it only has 4.44M active pixels?  That's more like 1080p-class resolution. The only way I can see it outputting 4k video is if it upscales its 2166 vertical lines to 3180, which can't help but look nasty.

Right, it doesn't make sense. You don't simply capture 6k pixels of horizontal resolution from a 2k sensor.

[Inazuma]

Right, it doesn't make sense. You don't simply capture 6k pixels of horizontal resolution from a 2k sensor.

 

[jgharding]

My guess is that where Foveon uses stacked sensors, this effectively uses  RGB pixels side by side, hence the pixels being wide.

 

The filter will work at an extremely high speed and the three images will be blended together to create full RGB image. 

[jgharding]

How can the sensor output 6k video (or even 4k video) when it only has 4.44M active pixels?  That's more like 1080p-class resolution. The only way I can see it outputting 4k video is if it upscales its 2166 vertical lines to 3180, which can't help but look nasty.

 

assuming Sony use their RGBW system, creating a red, green, blue and white pass using an electronic filter:

 

4.44M * 4 = 17.76M = 18MP

 

18 megapixels is about 5760 * 3240, roughly 6K

 

So the maths points to a very high colour depth indeed, combining full colour and a monochrome channel. Of course, they're then having to stretch this out to 6K final spacial resolution, so like Bayer, there's trickery involved.

 

The only way to get true 6K (as with any resolution) is either user three 18MP sensors stacked, split by prisms, or use one Bayer sensor with four times the pixels. Still, it could be the best picture ever... ;)

[Xavier Plagaro Mussard]

I really really want to read what it lacks from the naysayers...... ;-DD

[maxotics]

When was the last time you've actually seen someone use one of their cameras?  

 

I use them. I hate everything about them except the final image.  If I have good light, and time, and need medium format quality still in my pocket, they are the ONLY game in town.  The same for your BMPCC if you want 1080 RAW video in your pocket.

 

My gut feeling is this new Sony sensor is more about semantics, than any true pixel-level RGB sampling (like Foveon).  

 

The problem is something along the lines of Heisenberg Uncertainty Principle.  In our world, you can either know a light beam's color or intensity, but not both at any given instant.  All sensors are intensity only.  Whether Bayer or Foveon, filters are put in front of the sensor to estimate color.  If you think there is NO problem in this then why does Leica have a monochrome (no color filter) camera that sells for $6,000?  (It probably cost them $50 to make it, but no matter  ;) ).  If you're a B/W purist, those filters degrade your black and white image.  

 

So there are only two ways around this problem (unless Sony has discovered an electrically conductive material that can read color), you can stack color filtered sensors on top of each other (like Foveon) or next to each other (like Bayer).  With Foveon, you get true color pixels with little color distortion (if in strong light); with Bayer, you get high sensitivity color pixels, but when you combine them horizontally you get aliasing/moire problems.  

 

Theoretically, you could take a grid of color filters, RGB, and vibrate them across the sensor so that the sensor could take three readings for each color.  So if you had a global shutter that ran at 72 frames a second, it could take the red at 1/3rd a 24th of a second, then the blue, then green.  Perhaps they use a seriously precise stepper-motor to do this.  

 

If the sensels are rectangle, maybe they use that to capture all three colors at any instance, but the vibration changes the pixel-center focus color and the pixel just averages them all together.  

 

It's all very interesting stuff, to me at least, but my guess is that though it may make for a good video application, it won't be good enough for still photography (at least professional or enthusiast).  The reason is that Foveon doesn't work because of PHYSICS, it isn't a failure of Sigma.  They simply can't find a substance that will take the color value of light and send enough light to the sensel below it, then the next one below.  If Sony can change a color filter over the sensel it could eliminate color moire problems in a still subject, but if the subject moves then the color may change between filter changes (in the 1/3rd of 24th of a second).  Color problems are back!

 

I believe, understanding this stuff makes one a better photographer, or filmmaker, even if it has no immediate practical use on set.  For example, if you noticed a lot of moire in the background of your shot you might go out and buy all kinds of blur filters.  Every time you got rid of the moire you might find the image not sharp enough.  If you knew this stuff about sensors, however, you'd open the aperture up a bit and increase the blur in the background while keeping your subject in focus.

[Nikkor]

The sensor can do 16000 frames per second, so that color filter is faster than that.