New Canon video discussing the EOS C300 Mark III dual gain output sensor

[RMac]

This popped up yesterday without much fanfare and provides some clarifying info...

I wonder if/when we'll see this sort of technology filter down into non-cinema cameras.

 

[SecureGSM]

This popped up yesterday without much fanfare and provides some clarifying info...
I wonder if/when we'll see this sort of technology filter down into non-cinema cameras.

like DPRAW in 5D4 available since 2016?

 

[koenkooi]

like DPRAW in 5D4 available since 2016?

DPRAW uses the same gain on both diodes, this uses a different gain on each diode.

 

[koenkooi]

DPRAW uses the same gain on both diodes, this uses a different gain on each diode.

About halfway into the linked video this is actually explained: each single diode is read simultaneously with 2 difference gains. So in theory you could do the DPRAW trick on top of DGO.

 

[Joules]

like DPRAW in 5D4 available since 2016?

It doesn't sound like this has anything to do with the Dual Pixel design. The guy even says that it is only similar in name.

The two sub pixels in a DPRAW image can cause slight artifact due to parallax. Using the DPRAW approach for HDR would have required them to work around that. To me it sounds like they are actually using new additional hardware in the amplification process to achieve what they have done here. Meaning it is also unlikely to come down to stills soon, since it probably comes with additional costs?

 

[koenkooi]

It doesn't sound like this has anything to do with the Dual Pixel design. The guy even says that it is only similar in name.

The two sub pixels in a DPRAW image can cause slight artifact due to parallax. Using the DPRAW approach for HDR would have required them to work around that. To me it sounds like they are actually using new additional hardware in the amplification process to achieve what they have done here. Meaning it is also unlikely to come down to stills soon, since it probably comes with additional costs?

It would require double the circuits per diode for readout and additional circuits for what Canon calls 'blending'.

 

[SecureGSM]

It doesn't sound like this has anything to do with the Dual Pixel design. The guy even says that it is only similar in name.

The two sub pixels in a DPRAW image can cause slight artifact due to parallax. Using the DPRAW approach for HDR would have required them to work around that. To me it sounds like they are actually using new additional hardware in the amplification process to achieve what they have done here. Meaning it is also unlikely to come down to stills soon, since it probably comes with additional costs?

the guy says it is only similar in name to dual pixel AUTO FOCUS. not DPRAW

however this:

Will the EOS R5 get a Dual Gain Output Imaging Sensor?

Will the EOS R5 get a Dual Gain Output Imaging Sensor? — The-Digital-Picture.com

www.the-digital-picture.com

 

[koenkooi]

the guy says it is only similar in name to dual pixel AUTO FOCUS. not DPRAW

however this:

Will the EOS R5 get a Dual Gain Output Imaging Sensor?

Will the EOS R5 get a Dual Gain Output Imaging Sensor? — The-Digital-Picture.com

www.the-digital-picture.com

Weird, in that interview he makes it seem that in a dual pixel the 2 diodes get a different, single gain, but in the youtube clip in the OP he says each diode (not pixel!) gets read out with difference gains.
I'm eagerly awaiting the white paper

 

[BeenThere]

If the R5 had DGO, this fact would have likely been leaked with the other info we have. But Canon could be saving it for the R1x.

 

[Joules]

the guy says it is only similar in name to dual pixel AUTO FOCUS. not DPRAW

however this:

http://[URL]https://www.the-digital-picture.com/News/News-Post.aspx?News=34631

[/URL]

You're right, upon reading the quote from a Canon Marketing guy on that page it seems the tech is based on the Dual Pixel design.

"Now for Dual Pixel autofocus, it's using those two frames for phase detection. However, the Canon engineers realized that they can use those two different frames and value them at different gains in order to expand the dynamic range. And so what is happening here is that off of diode A you have one image that is low gain and low noise. Off of diode B you're getting a frame that is of higher gain"

All the other marketing materials from them is written less clearly though. Would be nice if they could clear it up properly.

 

[Mt Spokane Photography]

I suspect that it will be a feature reserved for the eventual high end R series. Since the R5 will have a new sensor, its entirely possible but unlikely. We've seen patents for this sort of thing in the past, it was a matter of the ability to manufacture the sensor with the dual gain feature and still use the dual pixel for autofocus. That sounds pretty complex. There is little doubt that the dual pixel sensors are going to allow new features that are unique to a dual pixel design. The design also brings some issues, Canon has done very well at keeping side effects unnder control and they are of little consequence for most of us.

 

[miguelchain]

Guys, I´m new here. There is any news about a new C200? I´m thinking about buying one, but It´s a camera that has been released a few years ago...