Herrbill
Forum Enthusiast
I am experimenting with focus stacking but can't quite get my head around the differential settings.
Can anyone give me a simple explanation.
Can anyone give me a simple explanation.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Focus stacking -Differential
- Thread starter Herrbill
- Start date
mikero
Senior Member
If you want a simple answer, then just stick to a differential of 3. Going much higher is liable to result in gaps.
Mike
Mike
Pedagydusz
Veteran Member
The differential depends on the initial focus distance and the f value (which means the DoF). The result is the differential when focusing at 1 m is entirely different from the one when focusing at 10 cm.
I strongly advise you to try for yourself. Set your system to focus at the middle point of a ruler angled at 45º, and at the approximate distance you plan to work. Then try step 3 (as the previous poster suggested), look at the result, and move to step 6, and then step 9 (all at same aperture). Then repeat at two other apertures (all likely to be used, i.e. f/3, f6, and f9, for example).
Then think about the result, and work on the basis of that knowledge.
I strongly advise you to try for yourself. Set your system to focus at the middle point of a ruler angled at 45º, and at the approximate distance you plan to work. Then try step 3 (as the previous poster suggested), look at the result, and move to step 6, and then step 9 (all at same aperture). Then repeat at two other apertures (all likely to be used, i.e. f/3, f6, and f9, for example).
Then think about the result, and work on the basis of that knowledge.
Photongraphy
Senior Member
- Messages
- 2,230
- Solutions
- 4
- Reaction score
- 3,147
A while ago, Chris McGinnis made and posted this sheet for the 60mm macro. If you're using this lens, you may be as happy as I am with it.
Trevor Carpenter
Forum Pro
you may find this thread useful https://www.dpreview.com/forums/thread/4712398
Your link is pointing to this current thread - he'd end up in an endless loop if he clicks on it
Would the tables in this message be of help? Richard Turton: New Focus Stacking Tables
Trevor Carpenter
Forum Pro
Apologies, it should have been this one https://www.dpreview.com/forums/post/66682171Your link is pointing to this current thread - he'd end up in an endless loop if he clicks on it
Would the tables in this message be of help? Richard Turton: New Focus Stacking Tables
Richard Turton
Forum Enthusiast
- Messages
- 401
- Solutions
- 2
- Reaction score
- 592
Those tables from 7 years ago are not very accurate and not practical for real world use. I have recently redone the measurements after acquiring the new 90mm macro. This graph shows the focus differential (step size) as a function of focus distance for different aperture values. It is plotted on a base 2 logarithmic scale because the low end is unreadable on a linear scale.Your link is pointing to this current thread - he'd end up in an endless loop if he clicks on it
Would the tables in this message be of help? Richard Turton: New Focus Stacking Tables
This graph shows the depth of field (DoF) calculated using the conventional formula. The depth of field standards are very old and not adequate for modern digital cameras. The usual Circle of Confusion (CoC) for m4/3 of 0.015 mm corresponds to 4.5 pixels on a 20 MP sensor. The depth of field has therefore been calculated for a smaller CoC of 2.8 pixels (0.093 mm), corresponding to the resolution limit for the green channel of a Bayer sensor. The relationship is very close to linear on a log/log plot.
The third graph shows the result of dividing the depth of field by the focus step size. The curves for the different aperture values all lie on top of each other.
It appears that the camera adjusts the focus step size to maintain an approximately constant ratio between step size and depth of field for all apertures for a focus distance range of about 0.3 to 2.0 metres. This relationship does not hold for distant subjects or at true macro (1:1) distances.
Selecting a focus differential therefore requires determining how much variation in sharpness between focus steps is acceptable to you. You can then use this same value to get consistent results for all apertures and focus distances within the applicable range. Step sizes of 3 or less are almost always safe.
Herrbill
Forum Enthusiast
Thanks one and all for your feedback. It is informative and very welcome.
To date, I have used differential 3 for my macros and varied the DOF depending on the depth of the subject. The results are satisfactory.
Recently, I have tried focus stacking for landscape where front to back focus was a priority. Again, I choose the differential setting of 3 in conjunction with aperture f/5.6. The results have been very satisfactory.
I still don’t quite understand the relationship between aperture and differential despite getting satisfactory results.
I shall try out Antonio’s method and see how I get on.
Thanks again.
Bill
To date, I have used differential 3 for my macros and varied the DOF depending on the depth of the subject. The results are satisfactory.
Recently, I have tried focus stacking for landscape where front to back focus was a priority. Again, I choose the differential setting of 3 in conjunction with aperture f/5.6. The results have been very satisfactory.
I still don’t quite understand the relationship between aperture and differential despite getting satisfactory results.
I shall try out Antonio’s method and see how I get on.
Thanks again.
Bill
Atho
Senior Member
- Messages
- 2,378
- Solutions
- 1
- Reaction score
- 1,350
And this one is forApologies, it should have been this one https://www.dpreview.com/forums/post/66682171Your link is pointing to this current thread - he'd end up in an endless loop if he clicks on it
Would the tables in this message be of help? Richard Turton: New Focus Stacking Tables
“How many fps is enough for wildlife (bif)?”
3rd time lucky?
Trevor Carpenter
Forum Pro
Thanks!
btw, I think it would be brilliant if you'd post a final topic that collects all your focus stacking findings before the forum is turned into an archive. This reply that you just posted, along with your previous threads Focus Stacking Focus Point and A Feature of Olympus Focus Stacking contain a wealth of information that should be especially useful to people who are joining the system for its macro capabilities.
btw, I think it would be brilliant if you'd post a final topic that collects all your focus stacking findings before the forum is turned into an archive. This reply that you just posted, along with your previous threads Focus Stacking Focus Point and A Feature of Olympus Focus Stacking contain a wealth of information that should be especially useful to people who are joining the system for its macro capabilities.
I've seen charts like these before, and they often go out to many meters distance, and as you point out, the relationship does not hold at near 1:1 distances. When people are trying to focus stack 20, 40, or 100 framse, it's not at many meters distance, it's close up. The trick to focus differential is when subject distance is measured in millimeters, not in meters.
It would be incredibly helpful so see an analysis and chart devoted to distance of less than 30 centimeters, especially with the 60mm macro and 90mm macro lenses, and explaining how focus differential, depth of field, and subject distance interact for true macro shooting.
Peadingle
Senior Member
Presumably we need an A3 printer as the actual size won't fit on A4?A while ago, Chris McGinnis made and posted this sheet for the 60mm macro. If you're using this lens, you may be as happy as I am with it.
--
"https://www.flickr.com/photos/6pix/"
Richard Turton
Forum Enthusiast
- Messages
- 401
- Solutions
- 2
- Reaction score
- 592
Richard Turton
Forum Enthusiast
- Messages
- 401
- Solutions
- 2
- Reaction score
- 592
This is just speculation on my part, but it appears that the focus bracketing algorithm only works over a limited range because of the way distance is determined. The camera does not have any means to measure the actual focusing distance and must estimate it from the only information is has, the lens focusing position. The problem is that the relationship between EXIF value Focus Step Count and focusing distance is highly non-linear. These graphs show the relationship for the 60mm and 90mm macro lenses.
The estimated focus distance in the EXIF is quite accurate for the 90mm, but a crude straight line approximation for the 60mm. Perhaps the newer lens has a more powerful CPU that can implement a more complex function when converting from step count to focus distance. The resolution of the EXIF focusing distance is 5 mm, which may give an indication of its expected accuracy.
The slope of the curve approaches infinity for long distances and zero for very short distances. It is therefore possible to accurately estimate focusing distance only in the transition range between these extremes. Outside this restricted range a fixed step size proportional to aperture is used and the relationship between step size and depth of field is lost. This may explain why most of my attempts at extreme macro produce a very disappointing result.
1) Your viewing/printing software should be able to print it to any size sheet.
2) It's only 813x1053 pixels in size. That's less than a postcard when printed at 300 dots per inch.
KarstenG
Member
Richard - I'm not quite sure how to read these tables.
DoF vs Focus Distance is clear; I'm able to reproduce this using a DoF calculator.
But how to read Step Size vs. Focus Distance?
For example:
I'm looking at a subject in 1m distance. Using 60mm @ f2.8 DoF is ~14mm.
The corresponding step size (light blue curve) is ~8mm.
Does that mean step size is 8mm for differential=1, 16mm for diff=2, 24mm for diff=3 ??
DoF vs Focus Distance is clear; I'm able to reproduce this using a DoF calculator.
But how to read Step Size vs. Focus Distance?
For example:
I'm looking at a subject in 1m distance. Using 60mm @ f2.8 DoF is ~14mm.
The corresponding step size (light blue curve) is ~8mm.
Does that mean step size is 8mm for differential=1, 16mm for diff=2, 24mm for diff=3 ??
David5833
Senior Member
There is no simple explanation. You are stuck with trial and error. What is needed is an algorithm that allows the user to set the near focus point and the far focus point and have the camera calculate the focus differential and the number of steps at the selected aperture to cover the near and far points. There are 3rd party apps that can do it for Nikon and Canon cameras. The brain trust at OMDS should be able to figure it out.
I don't know if the OM-x cameras don't have the processing power to figure it out, maybe that's the reason. Regardless, it ought to be available in some form.
Richard Turton
Forum Enthusiast
- Messages
- 401
- Solutions
- 2
- Reaction score
- 592
You are correct. I should have specified that these graphs are calculated for Focus Differential=1, and step size is linearly proportional to Focus Differential.
Similar threads
Keyboard shortcuts
- f
- Forum