New MFA method

[digital paradise]

Looks pretty good as it uses LV which is accurate and consistent as the base.

http://www.fredmiranda.com/forum/topic/1187247/0

[J.R.]

Looks pretty good as it uses LV which is accurate and consistent as the base.

http://www.fredmiranda.com/forum/topic/1187247/0

Interesting ... worth a try probably, not too much time involved in the steps

[helpful]

Pure genius.

This is something so simple that even the camera manufacturers should be able to program it into their firmware.

Now wouldn't that be nice... just point the camera at a target at the distance where you wanted AF fine tune to be optimized, and press a button which would do this procedure instantly and electronically. You could even re-tune your cameras for a new distance (i.e., when shooting from the back row vs. front row) in real time on the job.

I'm definitely trying it. Why didn't we think of this before? It's so obvious.

[neuroanatomist]

Now wouldn't that be nice... just point the camera at a target at the distance where you wanted AF fine tune to be optimized, and press a button which would do this procedure instantly and electronically. You could even re-tune your cameras for a new distance (i.e., when shooting from the back row vs. front row) in real time on the job.

I'm definitely trying it. Why didn't we think of this before? It's so obvious.

Something similar has been suggested:

In theory, it should be possible to achieve an ‘accurate’ focus using contrast detection in Live View, where no alignment is needed because the image sensor is used to determine best focus, then compare that to the phase detection AF and correct accordingly. In practice, this is something that’s difficult to do (because the act of moving the focus ring to see if you were really optimally focused changes the focus, and you lose the ‘zero point’). However, this is an idea that Canon could implement as a semi-automated routine, i.e. set up and align target (Canon could sell one, and suitably overcharge for it as they do for other small pieces of plastic *cough*), then the camera automatically determines the optimal adjustment. File that one under ‘gee, wouldn’t it be nice…’

Using AF Confirmation for manual focus to avoid having to move the focus ring is a nice solution!

[Dylan777]

Interesting...thanks for sharing

[Apop]

looks really nice , i am going to try it tomorrow!

[TrumpetPower!]

That is a really interesting solution.

However...the great thing about FoCal isn't just the automation, but the fact that it's automated.

I wonder, though, if this might inspire the FoCal crew to a faster way to do their magic. Is focus confirmation without focus available through the SDK?

b&

[helpful]

What's even cooler about this method is that the sensor is used only once, at the very beginning. The imaging sensor isn't even used when performing the phase detection AF afterwards.

In contrast, using live view continuously as with FoCal, etc., slowly warms up the sensor and increases the noise and decreases the live view focusing accuracy.

So this method should give maximum accuracy, although it all is dependent on not bumping or vibrating the lens when flipping the switch to manual focus after the first live view photograph is taken.

[RS2021]

However...the great thing about FoCal isn't just the automation, but the fact that it's automated.

I use and like FoCal...but it is not fully automated with either 1DX or 5D3.  "Change to -20, or +20, +10, -10, etc etc" is not exactly automation...and FoCal puts it on Cannon and I am sure Canon has no interest in making it easy for FoCal. Still, it is a great program.

But the approach in this post looks interesting.

[comsense]

Now wouldn't that be nice... just point the camera at a target at the distance where you wanted AF fine tune to be optimized, and press a button which would do this procedure instantly and electronically. You could even re-tune your cameras for a new distance (i.e., when shooting from the back row vs. front row) in real time on the job.

I'm definitely trying it. Why didn't we think of this before? It's so obvious.

Something similar has been suggested:

In theory, it should be possible to achieve an ‘accurate’ focus using contrast detection in Live View, where no alignment is needed because the image sensor is used to determine best focus, then compare that to the phase detection AF and correct accordingly. In practice, this is something that’s difficult to do (because the act of moving the focus ring to see if you were really optimally focused changes the focus, and you lose the ‘zero point’). However, this is an idea that Canon could implement as a semi-automated routine, i.e. set up and align target (Canon could sell one, and suitably overcharge for it as they do for other small pieces of plastic *cough*), then the camera automatically determines the optimal adjustment. File that one under ‘gee, wouldn’t it be nice…’

Using AF Confirmation for manual focus to avoid having to move the focus ring is a nice solution!

C'mon don't give half-hearted compliment. It's little more than similar. The crux here is to find the spread of focus distribution with MFA adjustment around the best focus achieved by live view. Its smart. Rather, what you quoted is very straightforward.

[Viggo]

This could maybe have solved the problem with my Zeiss 28 f2, that hit spot on from infinity down towards mfd, but the other way it missed, have something in between would have helped.

[TrumpetPower!]

What's even cooler about this method is that the sensor is used only once, at the very beginning. The imaging sensor isn't even used when performing the phase detection AF afterwards.

In contrast, using live view continuously as with FoCal, etc., slowly warms up the sensor and increases the noise and decreases the live view focusing accuracy.

So this method should give maximum accuracy, although it all is dependent on not bumping or vibrating the lens when flipping the switch to manual focus after the first live view photograph is taken.

That's nothing a bit of gaffer's tape won't solve....

b&

[Mantanuska]

I dont know much about AFMA since I've never owned a body that has the feature, but after finding the acceptable range with AF confirmation wouldn't you want to set it to a value that is on the first 3rd of the lower end instead of the midpoint, since the focal plane extends 1/3 in front of the subject and 2/3 behind?

[neuroanatomist]

The 1/3-2/3 distribution only applies at short focal lengths, especially with close subjects.

[digital paradise]

It will be interesting to get some feedback on how it works. I have always believed that an MFA system should be based on LV. As suggested it could be incorporated in the camera system.