Dual Pixel AF system - does it buy you anything for still photos

[RussRoc]

Can someone explain what "Dual Pixel AF system " gets you, if your not shooting video? Keep it simple.

 

[Loswr]

Faster focus in Live View.

 

[TLN]

Yep. If you don't use LiveView or shoot video all that modern features like STM-motor or Dual-Pixes absolutely uselss for you.

 

[Mt Spokane Photography]

It has the potential to be very valuable to you for still images, but only in liveview.

1. Fast focus in liveview which many users moving up from P&S bodies prefer. I use it exclusively in my studio, and it could be a marked improvement over the slow AF on my old 40D. Tracking also seems to work well from the few tests I've seen.
2. Once again in liveview, you can now AF with a f/11 lens. This means you can use a 1.4X or a 2X TC on a f/5.6 lens and have fast autofocus. Right now, there is no other Canon camera that has fast AF at f/11. I've had a 1 series camera focus slowly on a f/11 lens combo with taped contacts, but not quickly or without hunting.

 

[bchernicoff]

As others have said the main benefit is faster Live View if we are only talking about the 70D, but your question could be interpreted to apply in a broader sense.

It should allow for a camera that can self-tune its microfocus adjustment by comparing the result of traditional AF phase detection to on-chip phase detection. I haven't seen this mentioned as a feature of the 70D, but maybe the Magic Lantern folks will come to the rescue.

It also allows for much better phase detect AF in mirrorless cameras. Currently, they embed a handful of traditional phase detect AF points in the sensor and then interpolate to fill in the holes in the final image. This allows for much more powerful phase detect AF across at least 80% of the sensor without this downside.

Also, it seems that there may be additional benefits in a pro body like the 1DX. With no need to move a mirror around to focus it could achieve higher FPS in still burst mode while maintaining focus. There may be subject tracking benefits in a sports setting as demonstrated in the 70D's ability to follow a subject in video mode.

 

[Mt Spokane Photography]

As others have said the main benefit is faster Live View if we are only talking about the 70D, but your question could be interpreted to apply in a broader sense.

It should allow for a camera that can self-tune its microfocus adjustment by comparing the result of traditional AF phase detection to on-chip phase detection. I haven't seen this mentioned as a feature of the 70D, but maybe the Magic Lantern folks will come to the rescue.

It also allows for much better phase detect AF in mirrorless cameras. Currently, they embed a handful of traditional phase detect AF points in the sensor and then interpolate to fill in the holes in the final image. This allows for much more powerful phase detect AF across at least 80% of the sensor without this downside.

Also, it seems that there may be additional benefits in a pro body like the 1DX. With no need to move a mirror around to focus it could achieve higher FPS in still burst mode while maintaining focus. There may be subject tracking benefits in a sports setting as demonstrated in the 70D's ability to follow a subject in video mode.

The issue may be that on chip phase detection is not inherently accurate. It likely also has to be calibrated and adjusted. The tiny dimensions involved have tolerances as well, a millionth of inch error could be significant.

Have you seen anything that indicates that the onchip phase detection is inherently accurate? Does Canon have perfect dimensioning with zero tolerances? I've wondered if averaging over many sensors would mean that errors mostly cancel out, but its also possible that if they all have the same error that errors in dimensions would be additive.

Unless someone has found a white paper explaining the accuracy, I assume that onchip AF has errors as well, but I hope not.

Its certainly possible for Canon to have a utility to adjust phase detect AF by comparing it with contrast detect, so there is nothing new that way. Of course, contrast detection can have errors as well, as some have discovered.

 

[Loswr]

Have you seen anything that indicates that the onchip phase detection is inherently accurate? Does Canon have perfect dimensioning with zero tolerances? I've wondered if averaging over many sensors would mean that errors mostly cancel out, but its also possible that if they all have the same error that errors in dimensions would be additive.

Unless someone has found a white paper explaining the accuracy, I assume that onchip AF has errors as well, but I hope not.

I suspect it's probably as accurate and precise (or inaccurate and imprecise) as the dedicated sensor. Having the phase AF on the image sensor eliminates the registration error between sensors, but trades that error for a need of tighter (not achievable in practice) tolerances given the smaller distances involved.

Overall, I suspect a properly microadjusted dedicated sensor will be more accurate than the dual pixel AF from a practical standpoint.

 

[bchernicoff]

The issue may be that on chip phase detection is not inherently accurate. It likely also has to be calibrated and adjusted. The tiny dimensions involved have tolerances as well, a millionth of inch error could be significant.

Have you seen anything that indicates that the onchip phase detection is inherently accurate? Does Canon have perfect dimensioning with zero tolerances? I've wondered if averaging over many sensors would mean that errors mostly cancel out, but its also possible that if they all have the same error that errors in dimensions would be additive.

Unless someone has found a white paper explaining the accuracy, I assume that onchip AF has errors as well, but I hope not.

Its certainly possible for Canon to have a utility to adjust phase detect AF by comparing it with contrast detect, so there is nothing new that way. Of course, contrast detection can have errors as well, as some have discovered.

IIRC, the reason we need to AFMA is that the distance that light takes from the lens to the camera sensor does not exactly match the distance it takes from the lens, through the mirror, and to the AF sensor due to manufacturing tolerance reasons. In other words, the camera's AF system correctly focuses the lens to the AF sensor and then assumes that focus distance also applies to the camera's image sensor. The reality is that sometimes fine-tuning is needed.

Since this new dual-pixel phase detect AF is using the actual image sensor to detect correct focus it is inherently free from needing any fine-tuning. Therefore it should be possible for the camera to self-microadjust the traditional phase detect AF. It would first obtain what it thinks is correct focus the traditional way, then check that focus using the dual-pixel phase detect AF. The difference between the two is the amount to micro-adjust. Even if neither the traditional nor the dual-pixel phase detect AF are perfectly accurate, the correct micro-adjust value can still be determined by making multiple comparisons and using some statistics.

 

[dafrank]

All the previous posts do well to specifically answer your question as to the dual pixel advantages in focusing. However, there are some possibilities that may or may not actually ever be incorporated into later designs, using dual pixels, which could open up a number of options not now available. Somewhat like Fuji in some of their recent sensor designs, Canon could actually use one pixel in each pair for reading highlights and the other for reading shadows, thereby creating a sort of instant HDR exposure, increasing DR by quite a lot. Other arrangements could also be made to use one of each pair to read color information separately and differently from its mate in the same pair, thereby increasing color depth and accuracy. I'm sure that many things that others could think of, would also be possible , but these are some that sound interesting to me. Whether any of this could or will actually happen is another thing entirely. Maybe Canon already has this in mind, maybe not.

Regards,
David