WA focussing: advice please

[mwilliamsphotography]

Advertisement (gone after registration)

Hmmm, I've always used a slight lateral side shift, parallel to the subject plane for focus-recompose.

 

Far less chance of forward-backward sway. Feet in a "T" stance, slight bend at the knees, keep your back arched to hold the center of balance steady over your hips.

 

Learned that one from my years as a Karate instructor (long gone along with my knee cartilage : -)... holding your stance/balance while dodging a punch to the face is pretty good motivation ... LOL!

 

Practice makes perfect.

 

-Marc

[dbraid]

However, if you are focussing on a part that is 1.60 away, you'd better not lean at all. After you recompose, you still want to have focussed at 1.60 as long as the relative positions have not changed.

 

This is a good discussion, because it makes me think about what I do.

 

I am puzzled, however by the suggestion quoted above, given the original example of a wall at 1.5 meters and then focussing on a detail that is off center and at 1.6 meters when the rangefinder patch is aimed at this point to focus. When you recompose, if you do not lean or compensate, is not the camera then focussed at an object plane 1.6 meters from the camera? If the lens is ideal and has a flat pane of focus, then both wall and detail will be out of focus. Am I missing something?

 

Thanks

 

Donald

[giordano]

Am I missing something?

 

Not a thing, except the l in "plane of focus".

[Guest mc_k]

The formula is Pythagoras's, so I worked out the details for myself...

 

...was expecting a formula in terms of the original distance and lens angle-of-view. This is interesting, so can you give all of your details for the 21mm lens calculation?

[giordano]

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

The blue dot is the camera. The dotted lines show the approximate angle of view. The object of interest is half-way between the centre and the edge of the frame.

[Guest mc_k]

thanks for the diagram! How did you get the 136mm? Does the hypotenuse represent 1.5m + the error?

 

Am also wondering what you used for the angle of view?

[Guest mc_k]

Advertisement (gone after registration)

here is a formula that keeps popping up on rangefinderforum:

 

error = d X (1 - cos(t/4)), where d is the original distance and t is the angle of view;

 

 

I'm sure this is mistaken though.

[giordano]

thanks for the diagram! How did you get the 136mm? Does the hypotenuse represent 1.5m + the error?

 

Am also wondering what you used for the angle of view?

 

I think the 136mm was a mistake, or rather two mistakes (hitting 1,3,6, when I intended 1,2,5): should have been 125mm, i.e. 1.625m minus 1.500m.

 

The hypotenuse is simply the distance from the camera to the object of interest. What the diagram is meant to illustrate is that the focus error when you recompose will be the difference between the base of the triangle (distance from camera to subject plane, i.e. the wall) and the hypotenuse (distance from camera to object of interest).

 

The angle of view isn't important in this example. According to the Barnack calculator (which I checked with some rough geometry) the field of view of a 21mm lens focused at 1.5m is approximately 2.5m wide. So an object halfway between the centre of the frame and the edge would be 0.625m off the central axis.

 

If you want a formula, how about this:

Focus error = r(1 − cos θ)

where r is the distance from the camera to the subject plane and θ is the angle you turn the camera through when you recompose.

[Guest mc_k]

...

The angle of view isn't important in this example. According to the Barnack calculator (which I checked with some rough geometry) the field of view of a 21mm lens focused at 1.5m is approximately 2.5m wide. So an object halfway between the centre of the frame and the edge would be 0.625m off the central axis.

...

 

The angle of view was important because it determined one of the sides of the triangle, and no way to check your work without it.

 

The hypotenuse in the diagram is indeed the 1.5m + error.

 

Thanks for the good info.