Olympus in-camera focus bracketing with 10x inf objective
Moderators: rjlittlefield, ChrisR, Chris S., Pau
No I have no formula, only self experience.zzffnn wrote: Once your first bracketed slice stops, how do you determine where to start for the second slice? Do you look at the last image of the first slice, then start from that focus plane? Or do you go pass that focus plane slightly (how much do you go pass, if you have formula or way to determine), then start again?
I work with the Olympus capture software.
I use the focus adjustment delivered by default with the software, all the amount drivers in the middle of the rule.
Calibration of the system.
First of all at the beginning of a new stack, I press several times on the ">>>" to force the lens to focus at infinity.
I make focus on a precise target like micrometer calibration slide.
Then I click 8 times on the "<<<" button. this number "8" was chosen by empirical way, the slice is not too narrow and ZS manage well perspective changes. At 20x, I click only 6 times on the "<<<" button.
Then on the focus bracketing menu, I always choose 150 images and the focus differential adapted for the objective in use (for 10x in example)
The focus bracketing begins after pressing the shutter. It always go from your focus point to infinity. Normally, the last image of the stack is in focus. I press again 8 times on "<<<" and then, I move the object to find again the focus. This movement is made with the micrometer adjustment of an old microscope so I could know how deep was the stack. By security, I keep only 75% of this measure to have some overlapping between slices.
No, It is true when you work on the side of depth of field (object side) but not when you work on the side of depth of focus (image side).zzffnn wrote: Also, I though step size for 20x would be smaller than that of 10x, due to higher NA of 20x.
see the table 1 in this website, very instructive.
https://www.microscopyu.com/microscopy- ... h-of-focus
By experimentation.zzffnn wrote: How did you determine step size 8 for 20x and step size 5 for 10x?
Here it a 100% crop at actual pixel size but due to the maximum image size allowed on the forum I needed to compress the jpeg to only 80% .Lou Jost wrote: I have never done such high m as Jojm. It is very interesting to see how well that works. Jojm, could we see a 100% crop of the sharpest of these?
Lou Jost wrote:Jean-Marc, how is the image quality without the teleconverter?
The image quality without the teleconverter was better in the center but it suffers a lot of vignetting.Lou Jost wrote:Jean-Marc, does the zoom vignette much if you don't use the teleconverter?
As example, with the NIKON CFI 10x, this table shows you that even at 150mm I saw some little vignetting.
- rjlittlefield
- Site Admin
- Posts: 23608
- Joined: Tue Aug 01, 2006 8:34 am
- Location: Richland, Washington State, USA
- Contact:
To clarify... The reason is effective aperture, as seen by the camera. Because the NA of the 20X is not double the NA of the 10X, the 20X provides a narrower effective aperture, that is, a larger effective f-number. It is that larger effective f-number which provides a correspondingly larger DOF in the image.jojm wrote:No, It is true when you work on the side of depth of field (object side) but not when you work on the side of depth of focus (image side).zzffnn wrote: Also, I though step size for 20x would be smaller than that of 10x, due to higher NA of 20x.
--Rik
Thank you very much, Jean-Marc , Lou and Rik.
Rik's clarification helped me understood it. I am slow here as I am not good with physics or optical math.
Jean-Marc,
When you said you "press again 8 times on "<<<" and then, I move the object to find again the focus."
How do you "find the focus"? Do you move your sample image plane in focus to the focused plane in the last image, then back off around 25% (to obtain some overlaps)?
I did not know we can use manual focus in focus bracketing with Olympus capture software.
I did focus bracketing without computer control (with camera alone) and cannot get bracketing to run under manual focus (MF), with a stereo scope focus drive (which is too imprecise). MF always gave me one frame (with focus bracketing turned on). I don't know why.
So I used single auto focus, which is unpredictable and sometimes won't start, unless I moved camera/lens to around middle depth of my sample.
I should reconfigure my focus drive system, after I fully digest comments here.
Lou,
How do you move your camera or subject? I am guessing at high magnification, stereo scope focus drive may not be precise enough.
Rik's clarification helped me understood it. I am slow here as I am not good with physics or optical math.
Jean-Marc,
When you said you "press again 8 times on "<<<" and then, I move the object to find again the focus."
How do you "find the focus"? Do you move your sample image plane in focus to the focused plane in the last image, then back off around 25% (to obtain some overlaps)?
I did not know we can use manual focus in focus bracketing with Olympus capture software.
I did focus bracketing without computer control (with camera alone) and cannot get bracketing to run under manual focus (MF), with a stereo scope focus drive (which is too imprecise). MF always gave me one frame (with focus bracketing turned on). I don't know why.
So I used single auto focus, which is unpredictable and sometimes won't start, unless I moved camera/lens to around middle depth of my sample.
I should reconfigure my focus drive system, after I fully digest comments here.
Lou,
How do you move your camera or subject? I am guessing at high magnification, stereo scope focus drive may not be precise enough.
Last edited by zzffnn on Tue Feb 06, 2018 6:09 pm, edited 1 time in total.
Selling my Canon FD 200mm F/2.8 lens
Jean-Marc, thanks very much for that information. Your 100% crop looks very good, no color fringing. The vignetting without the teleconverter is worrisome, thanks for letting me know about that. I am trying to decide whether to buy that zoom, and that informatin helps a lot.
Fan, at 10x I have no problem moving my subject and camera manually. Beyond that I use a motorized rail and/or a home-made vertical stage made from a camera lens. The focusing makes the platform rise or fall. Very precise.
Fan, at 10x I have no problem moving my subject and camera manually. Beyond that I use a motorized rail and/or a home-made vertical stage made from a camera lens. The focusing makes the platform rise or fall. Very precise.
Yes, at the end of the stack, the last image is in focus because at this moment the zoom is focused to infinity just like you've done your focus at the very beginning of the calibration session. By moving the zoom focus by pressing 8 times "<<<" you defocus again. By moving the subject (or all your optical system if you can) you will find again the focus, the length of this movement is equal to the depth of you slice And yes after I go back around 25% to obtain some good overlaps.zzffnn wrote: When you said you "press again 8 times on "<<<" and then, I move the object to find again the focus."
How do you "find the focus"? Do you move your sample image plane in focus to the focused plane in the last image, then back off around 25% (to obtain some overlaps)?
This is done only once for the calibration, after I always use the same operating mode: 8 "<<<", move the object by x, shoot, 8 "<<<", move the object by x mm, shoot, ... and so on until I have reach the deeper point of my subject.
Yes it is very surprising but I think this is something caused by the construction of this zoom with constant length. On the other side, this constant length is just wonderful when you use this system on a stand!Lou Jost wrote: The vignetting without the teleconverter is worrisome, thanks for letting me know about that. I am trying to decide whether to buy that zoom, and that information helps a lot.
Fan, yes, the camera is all you need, no tethering is necessary.
Jean-Marc, yes, the constant length is the most attractive thing about it, because it would be so easy to compose! I hate having to change tube lenses for even the most minor change in FOV.
If one did not want to use focus bracketing, maybe a FF Nikon 80-200 ED f/2.8 constant-length zoom would give good results. Minor vignetting in the aerial image would not be noticed by the little MFT sensor. I got one of these zooms very cheap because the autofocus was broke:
https://www.ebay.com/itm/NIKON-ED-AF-S- ... 2749.l2649
Will soon try it out. But it would be so nice to have the focus-bracketing!
Jean-Marc, yes, the constant length is the most attractive thing about it, because it would be so easy to compose! I hate having to change tube lenses for even the most minor change in FOV.
If one did not want to use focus bracketing, maybe a FF Nikon 80-200 ED f/2.8 constant-length zoom would give good results. Minor vignetting in the aerial image would not be noticed by the little MFT sensor. I got one of these zooms very cheap because the autofocus was broke:
https://www.ebay.com/itm/NIKON-ED-AF-S- ... 2749.l2649
Will soon try it out. But it would be so nice to have the focus-bracketing!
Two more related question, if you can kindly bear with me:
Sometimes I could see a sharply focused plane in camera's view finder, camera could focus but won't start focus bracketing. I am guessing that is because its lens was already focused to infinity (as I forgot to move focus away from infinity)? Or would there be another reason?
And how much should I move away from infinity (I am guessing too much away from infinity may not be good, since a tube lens for infinity objective should ideally focused to infinity and high NA [0.40] demands better adherence to that rule)?
I don't see that "<<<" or ">>>" button on my camera. So I cannot do 8x "<<<", as Jean-Marc did. I would like to focus to a closer subject, say at XYZ meters away. But how do I determine that XYZ value?
I have two suitable zoom lens, a Panasonic 100-300mm F/4.0-5.6 version I (whose closest focus point is 1.5m) and a Olympus 40-150mm F/4.0-5.6 (whose closest focus point is 0.9m).
Sometimes I could see a sharply focused plane in camera's view finder, camera could focus but won't start focus bracketing. I am guessing that is because its lens was already focused to infinity (as I forgot to move focus away from infinity)? Or would there be another reason?
And how much should I move away from infinity (I am guessing too much away from infinity may not be good, since a tube lens for infinity objective should ideally focused to infinity and high NA [0.40] demands better adherence to that rule)?
I don't see that "<<<" or ">>>" button on my camera. So I cannot do 8x "<<<", as Jean-Marc did. I would like to focus to a closer subject, say at XYZ meters away. But how do I determine that XYZ value?
I have two suitable zoom lens, a Panasonic 100-300mm F/4.0-5.6 version I (whose closest focus point is 1.5m) and a Olympus 40-150mm F/4.0-5.6 (whose closest focus point is 0.9m).
Last edited by zzffnn on Fri Feb 02, 2018 5:41 pm, edited 2 times in total.
Thanks Lou,Lou Jost wrote:Fan, yes, the camera is all you need, no tethering is necessary.
The thing that I do not understand is that MF, without computer tethering, never runs focus bracketing for me; after I set up focus bkt, the camera would take one frame and stop. Only S-AF (single auto focus) can get focus bkt run continuous.
But Jean-Marc apparently used computer tethering / software to run focus bkt in MF mode and could tell camera exactly which plane to start bracketing, instead of having auto focus select one. In my case, sometimes, my camera's S-AF focus bkt started from a plane different than my ideal plane (which is usually the plane closest to camera).
That was why I wondered if I am losing that MF bkt control by not having computer tethering.