To stack or not to stack

John K wrote:

mawyatt2002 wrote:

DoF is unfortunately not a precise well defined "scientific parameter" based upon an arbitrary circle of confusion (good name!) and up to lots of interpretation and "confusion"!

Outside macro work it's usually defined based upon the acceptable in-focus longest position minus the closest position acceptable in-focus regions. Calculation involves the Hyperfocal distance and such which you can find with Google.

At macro and below it's usually best to consider a different approach to DoF and use Numerical Aperture, or NA.

From my notes:

NA= (M/(M+1))*(1/(2F))

Physics based "wave optics" results in a relatively simple definition for a quarter wavefront error:

DoF = lambda/NA^2,

where lambda is wavelength of light, usually assumed to be 550nm ( Green).

Another definition which augments the wavefront definition to include pixel size, not sensor size though, for small NA:

DoF = lambda/NA^2 + e/(M*NA)

where e is the pixel size.

So DoF does depend somewhat on pixel size.

For example with a NA of 0.28 at 5X DoF = ~7um, using a pixel of 4um results in DoF of ~9.9um. At 5X NA of 0.038 (F11, Feff of 66), DoF = ~ 383um, and considering a 4um pixel DoF = ~404um. At 2X with a NA of 0.055 results in DoF of ~182um and ~218um when 4um pixel is considered.

Best,

...and this is where we get into a debate over the math/physics and how it applies. The difference in the circle of confusion, pixels size, etc. between sensors is so small that it really doesn't make much of a difference in actual depth of field, or in how depth is perceived. I can get more apparent depth by controlling the angle between the sensor and the subject than you'll ever get by shooting with a different sensor...

Tech Specs: Canon 80D (F11, 1/250, ISO 100) + a Canon MP-E 65mm macro lens (over 3x) + a diffused MT-24EX (both flash heads on the Canon flash mount, E-TTL metering).

Experience has taught me that although a lot of the equations look like they should make a difference they really don't...

There is no debate, these facts are directly from optical physics. The only confusion you have for debate is the CoC, which is up to interpretation. Pixel size does make a difference, and even more so with the new Cell Phone cameras, where pixel size is ~1um.

You can continue to bury your head in the sand and ignore the real world with your "hand held, single image, no crop, F66 diffraction doesn't matter because I say so BS", meanwhile the rest of us will move forward.

Anyway, this is another thread that deviated from the OP intent, To stack or Not to stack is the original question I believe.

Best,