The GetDPI Photography Forum
Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!
Focus Stacking Question
- Thread starter Greg Haag
- Start date
dchew
Well-known member
Greg,
I went down the same rathole a year or two ago. My question was more about whether these edge interface issues would be any different if you moved the lens vs moving the back when focusing. For example, If I use the helical on the Alpa mount, I am moving the lens when focusing. If I use the Actus DB, I am moving the sensor. I "thought" using the Actus would give better results because if the lens does not move, perspective is not changing; only magnification is changing. Using the Alpa helical changes both perspective and magnification.
I had a setup very similar to yours, and I could not really say one approach was better than the other. Didn't matter if it was Zerene, Helicon or Photoshop. As Gerald points out, there is no way to "see" behind the close subjects because it is magnified slightly larger at the different focus settings. At least that is how I think of it.
Dave
I went down the same rathole a year or two ago. My question was more about whether these edge interface issues would be any different if you moved the lens vs moving the back when focusing. For example, If I use the helical on the Alpa mount, I am moving the lens when focusing. If I use the Actus DB, I am moving the sensor. I "thought" using the Actus would give better results because if the lens does not move, perspective is not changing; only magnification is changing. Using the Alpa helical changes both perspective and magnification.
I had a setup very similar to yours, and I could not really say one approach was better than the other. Didn't matter if it was Zerene, Helicon or Photoshop. As Gerald points out, there is no way to "see" behind the close subjects because it is magnified slightly larger at the different focus settings. At least that is how I think of it.
Dave
pegelli
Well-known member
One more effect that I observe when using a macro rail to move the whole camera (without refocusing the lens) from the front to the back of the subject is that a longer focal length works better than a short focal length because the perspective is changing less between the closest and the farthest shot and thereby the "rims" around closer objects become smaller.
I haven't tried the same by refocussing the lens from front to back of the subject to see what the effect of focal length is in that case. I guess similar but the proof of the pudding is in the testing (which is on my "to do list, but hasn't been done yet).
Obviously when using a longer focal length lens the perspective of the whole scene changes which might lead to not getting the composition you want, so it's not a solution for all problems when items in the front of the scene shield a bit of the background.
I have no technical camera so these are the only two methods I can use for focus stacking.
I haven't tried the same by refocussing the lens from front to back of the subject to see what the effect of focal length is in that case. I guess similar but the proof of the pudding is in the testing (which is on my "to do list, but hasn't been done yet).
Obviously when using a longer focal length lens the perspective of the whole scene changes which might lead to not getting the composition you want, so it's not a solution for all problems when items in the front of the scene shield a bit of the background.
I have no technical camera so these are the only two methods I can use for focus stacking.
Greg Haag
Well-known member
I adjusted my composition and tried a version with 150mm without the extension tube. At least on this image it seemed better, but still present at close inspection. I think all of these would be improved by shooting more images, currently only shooting about 40 images.
pegelli
Well-known member
I just remembered we have a dedicated thread for stacking discussion and results in the Image Processing forum. You might be able to pick up some details there as well.
And as gerald.d said and rdeloe showed earlier in this thread I don't think making more images to stack won't resolve the "halo" issue, since these halo's are caused by parts in the background the camera can't see when it's focussed there because it's then shielded by a (out of focus) foreground subject.
I posted another "shielding halo" example in the above linked thread, this time by leaving the camera in place and focussing "through" the scene by turning the lens. Looks very similar to the effect seen by moving the camera without touching the lens distance ring
And as gerald.d said and rdeloe showed earlier in this thread I don't think making more images to stack won't resolve the "halo" issue, since these halo's are caused by parts in the background the camera can't see when it's focussed there because it's then shielded by a (out of focus) foreground subject.
I posted another "shielding halo" example in the above linked thread, this time by leaving the camera in place and focussing "through" the scene by turning the lens. Looks very similar to the effect seen by moving the camera without touching the lens distance ring
Last edited:
dchew
Well-known member
I don't think there is a way to make it look good at 100%. But, I do think if you err on masking so the foreground subject has sharp edges and let the subject behind go soft at the interface, it looks a bit better.
Dave
Considering the evident expertise in this thread and forum generally, I’m a little diffident to post my simplistic approach to mitigating the problem - but here goes!
Your experience Greg motivated me test out something that I thought I had seen, but never confirmed. So I set up a small scene that has a passing resemblance to yours and took a series of shots with the XF in Auto Stacking mode and the SK 120mm - at a range of apertures from f4 to f16. The number of frames (as calculated by the XF) ranged from 101 at the largest aperture to 26 at the smallest. These seemed to work well enough for near to far sharpness so I then processed them in Zerene Stacker (just because I’m more familiar with it than Helicon) via C1.
I used my normal settings which involve stacking with both PMax & DMap (a standard option in the menu) and retouched edges in the final DMap from the PMax result. All learnt from Zerene tutorials telling me that DMap produces superior colours and PMax handles edges better.
The results? Poor edges (and at crossovers) at f4 and very acceptable ones at f16.
Of course there was a tad less sharpness through diffraction at f16 but that’s easily corrected in C1 if needed. I usually prefer not to because the final stacked images sometimes look over sharp to me.
My conclusions? The 120mm SK is great for Close-ups (and normal shots as well actually) and XF Auto Stacking works very well. The smaller apertures reduced the number of shots needed significantly and improved the quality of the final result noticeably. That’s what I thought I knew but my tests have now proved it to myself at least!
I think that it’s perfectly logical that a setting with an inherently deeper DOF will reduce the edge and overlap ghosting and would be interested if others have found the same.
I realise that the IQ150 may be another story altogether but I can only relate what I found with my mere 100MP
.
I’ve made some sort of attempt to show the difference in the 100% crops below but all the downscaling required has made it rather less obvious. But here they are anyway.
f4
f16
And this little exercise revealed the sorry state of my cutting mat, which has to go!
Your experience Greg motivated me test out something that I thought I had seen, but never confirmed. So I set up a small scene that has a passing resemblance to yours and took a series of shots with the XF in Auto Stacking mode and the SK 120mm - at a range of apertures from f4 to f16. The number of frames (as calculated by the XF) ranged from 101 at the largest aperture to 26 at the smallest. These seemed to work well enough for near to far sharpness so I then processed them in Zerene Stacker (just because I’m more familiar with it than Helicon) via C1.
I used my normal settings which involve stacking with both PMax & DMap (a standard option in the menu) and retouched edges in the final DMap from the PMax result. All learnt from Zerene tutorials telling me that DMap produces superior colours and PMax handles edges better.
The results? Poor edges (and at crossovers) at f4 and very acceptable ones at f16.
Of course there was a tad less sharpness through diffraction at f16 but that’s easily corrected in C1 if needed. I usually prefer not to because the final stacked images sometimes look over sharp to me.
My conclusions? The 120mm SK is great for Close-ups (and normal shots as well actually) and XF Auto Stacking works very well. The smaller apertures reduced the number of shots needed significantly and improved the quality of the final result noticeably. That’s what I thought I knew but my tests have now proved it to myself at least!
I think that it’s perfectly logical that a setting with an inherently deeper DOF will reduce the edge and overlap ghosting and would be interested if others have found the same.
I realise that the IQ150 may be another story altogether but I can only relate what I found with my mere 100MP
.I’ve made some sort of attempt to show the difference in the 100% crops below but all the downscaling required has made it rather less obvious. But here they are anyway.
f4
f16
And this little exercise revealed the sorry state of my cutting mat, which has to go!
Greg Haag
Well-known member
Paul, thank you so much for taking the time to do this, it is very helpful! There was a significant difference between f4 and f16. BTW, while my IQ4 150 has been great so far, my IQ3 100 was my favorite all around back I have ever owned.Considering the evident expertise in this thread and forum generally, I’m a little diffident to post my simplistic approach to mitigating the problem - but here goes!
Your experience Greg motivated me test out something that I thought I had seen, but never confirmed. So I set up a small scene that has a passing resemblance to yours and took a series of shots with the XF in Auto Stacking mode and the SK 120mm - at a range of apertures from f4 to f16. The number of frames (as calculated by the XF) ranged from 101 at the largest aperture to 26 at the smallest. These seemed to work well enough for near to far sharpness so I then processed them in Zerene Stacker (just because I’m more familiar with it than Helicon) via C1.
I used my normal settings which involve stacking with both PMax & DMap (a standard option in the menu) and retouched edges in the final DMap from the PMax result. All learnt from Zerene tutorials telling me that DMap produces superior colours and PMax handles edges better.
The results? Poor edges (and at crossovers) at f4 and very acceptable ones at f16.
Of course there was a tad less sharpness through diffraction at f16 but that’s easily corrected in C1 if needed. I usually prefer not to because the final stacked images sometimes look over sharp to me.
My conclusions? The 120mm SK is great for Close-ups (and normal shots as well actually) and XF Auto Stacking works very well. The smaller apertures reduced the number of shots needed significantly and improved the quality of the final result noticeably. That’s what I thought I knew but my tests have now proved it to myself at least!
I think that it’s perfectly logical that a setting with an inherently deeper DOF will reduce the edge and overlap ghosting and would be interested if others have found the same.
I realise that the IQ150 may be another story altogether but I can only relate what I found with my mere 100MP
I’ve made some sort of attempt to show the difference in the 100% crops below but all the downscaling required has made it rather less obvious. But here they are anyway.
f4
View attachment 179260
f16
View attachment 179261
And this little exercise revealed the sorry state of my cutting mat, which has to go!
bab
Active member
Greg that chart is the Holly Grail you cant use a rail and get results that are better that internal focusing your lens, hence WHY I have asked Hasselblad to give Integrated Step Focusing to the camera or to Phocus. The biggest issue I see here (learned the hard way from shooting over 30k images of jewelry) is as the focus point move down and around a curve the stepping falls off differently than you think. A way to get your mind around this is think of a drop of water sliding down and you keeping the top of the drop and the sides in critical focus.
Just a suggestion that might help with the image you show shoot the table and the tool box separately. Now comes the decision...for every 1 mm you need 1 image minimum depending on Large your final image would be printed. ex if you shooting the hammer and the three metal objects in one shot your going way beyond the limits of stacking possibilities for critical focus from the front to back without doing a major amount of work. The easier solution would be to shoot the objects separately and then combine them in PS. (also PS stacking is not as good as the other mentioned programs basically because your cleanup would be endless.)
F16 is fine no issues, however the camera stand, tripod, air movement, floor vibrations, vibrations from the shutter after taking an image, fan on strobe lights and forced air condition are all factors that WILL complicate the stacking efforts.
Stacking is not for Sissy's its a dedicated grunt job that doesn't have many rewards.
Just a suggestion that might help with the image you show shoot the table and the tool box separately. Now comes the decision...for every 1 mm you need 1 image minimum depending on Large your final image would be printed. ex if you shooting the hammer and the three metal objects in one shot your going way beyond the limits of stacking possibilities for critical focus from the front to back without doing a major amount of work. The easier solution would be to shoot the objects separately and then combine them in PS. (also PS stacking is not as good as the other mentioned programs basically because your cleanup would be endless.)
F16 is fine no issues, however the camera stand, tripod, air movement, floor vibrations, vibrations from the shutter after taking an image, fan on strobe lights and forced air condition are all factors that WILL complicate the stacking efforts.
Stacking is not for Sissy's its a dedicated grunt job that doesn't have many rewards.