I'm a British software developer and photographer living in Vancouver, BC. I mainly photograph landscapes, cityscapes, night scenes, and water.
If you're interested in any...
Interesting. Was the stacking done to preserve the focus of the main subject & bokeh vs just shooting at f/22 (or smaller) and possibly introducing artifacts in the background? Just trying to understand.
@brianl Focus stacking should produce much the same result (in terms of depth of field) as simply using a smaller aperture. There are two main reasons that focus stacking is used:
1. Focus stacking allows you to create a wider depth of field than is possible to be captured with the lens otherwise. In the case of this series of photos, I have a total depth of field (across all the photos) of approximately 10cm, with a subject distance of 60cm and a focal length of 105mm.
If we plug those numbers into a depth of field calculator (this being my favourite: http://www.dofmaster.com/dofjs.html ), we can see that in order to get a 10cm depth of field with a subject distance of 60cm and a focal length of 105mm in a single photo, we would need to use an aperture of f/64. My lens maxes out at f/22, and even proper macro lenses (which I don't own) rarely go much higher than about f/45, so this photograph simply wouldn't have been possible with a single shot. (You may have been able to get a similar shot by moving closer and using a wider angle, but that would have altered the perspective of the photo).
2. Focus stacking allows us to avoid the loss of sharpness that occurs when we use a very narrow aperture. This loss of sharpness is caused by diffraction -- in very simple terms, light beams spread out when they pass through a hole, and the smaller the hole, the more they spread out. When the hole (the iris in the lens) gets too small, the light beams spread out to the point of hitting several sensor sites, rather than just one. This results in a loss of detail -- the image that comes out of the lens is lower resolution than the sensor.
The point at which diffraction starts to affect image sharpness depends on the size of each pixel in the sensor -- smaller sensors on crop bodies are more susceptible, as are sensors with a lot of megapixels. On my camera I can go down to about f/16 without seeing any significant loss of sharpness, on a modern crop-frame sensor or a high-megapixel full-frame sensor, as on the D800, you may start to see a loss of detail above f/11. The aperture value that this occurs at is called the 'diffraction-limited aperture' (DLA).
By stacking the photos, because each individual photograph is below the DLA for the camera, the result is a pin-sharp image with an extremely deep depth of field.
By moving the mouse over the image, you can switch between a photo taken at f/4 and one taken at f/16, with a good lens. This is on a crop-frame camera, but still only an 18 megapixel camera. Unfortunately the chart only goes up to f/16, but you can imagine the additional loss of sharpness at smaller apertures, and/or with a higher-megapixel camera.
Obviously focus stacking has limitations, the most obvious being that it only works on non-moving subjects, but if you are looking for the ultimate in sharpness, and/or to cover a wider depth of field than your lens can natively produce, it's an excellent option to consider.
Great explanation and much more than I expected. Yes I see how this technique preserves the DOF while also making the main subject in full focus...like it was shot at something like f/22. I've used Zerene Stacker before for extreme macro and water refraction shots. I highly recommend it if you haven’t tried it.
I'm curious to know when doing the multiple shots with your Canon, does your camera have a feature that assists? My Sony has a feature called “peaking" which essentially shows you the area that is in focus by highlighting it in a color such as yellow. It's very helpful. As you refocus the area changes via live view. It makes getting the multiple shots done fairly easy. One bonus for Sony
@brianl I just use the function built into Photoshop. I don't do this kind of shot very much so I find it sufficient. I suspect that once you get more advanced, the Photoshop function might be a bit limiting, as it is for panoramic photos and HDR.
Canon don't offer focus peaking, which is slightly odd given their (pretty successful) push to get the 5D2 and 5D3 seen as a viable option for filmmakers. That hasn't stopped people from finding workarounds, of course. The Magic Lantern firmware available for many Canon cameras adds focus peaking as well as many many other features, such as fully-automated focus stacking (simply set the start and end point and the camera will take the required number of intermediate shots, adjusting the focus between each photo). Similarly, the app I have on my phone that controls my DSLR also offers a similar fully-automated stacking method. However, for this shot I simply moved the focus ring by the smallest amount possible in between photos -- not the most efficient method, but easy enough for one shot.
Nice exposition of the technique. I have really been treated lately with my new Nikkor f/4 70-200 constant zoom for nice sharpness even at crazy aperture settings of f/32, so I haven't played the focus stacking routine yet. But it will be on my list of things to try one day.
1. Focus stacking allows you to create a wider depth of field than is possible to be captured with the lens otherwise. In the case of this series of photos, I have a total depth of field (across all the photos) of approximately 10cm, with a subject distance of 60cm and a focal length of 105mm.
If we plug those numbers into a depth of field calculator (this being my favourite: http://www.dofmaster.com/dofjs.html ), we can see that in order to get a 10cm depth of field with a subject distance of 60cm and a focal length of 105mm in a single photo, we would need to use an aperture of f/64. My lens maxes out at f/22, and even proper macro lenses (which I don't own) rarely go much higher than about f/45, so this photograph simply wouldn't have been possible with a single shot. (You may have been able to get a similar shot by moving closer and using a wider angle, but that would have altered the perspective of the photo).
2. Focus stacking allows us to avoid the loss of sharpness that occurs when we use a very narrow aperture. This loss of sharpness is caused by diffraction -- in very simple terms, light beams spread out when they pass through a hole, and the smaller the hole, the more they spread out. When the hole (the iris in the lens) gets too small, the light beams spread out to the point of hitting several sensor sites, rather than just one. This results in a loss of detail -- the image that comes out of the lens is lower resolution than the sensor.
The point at which diffraction starts to affect image sharpness depends on the size of each pixel in the sensor -- smaller sensors on crop bodies are more susceptible, as are sensors with a lot of megapixels. On my camera I can go down to about f/16 without seeing any significant loss of sharpness, on a modern crop-frame sensor or a high-megapixel full-frame sensor, as on the D800, you may start to see a loss of detail above f/11. The aperture value that this occurs at is called the 'diffraction-limited aperture' (DLA).
By stacking the photos, because each individual photograph is below the DLA for the camera, the result is a pin-sharp image with an extremely deep depth of field.
You can see the effects of DLA at a link such as this (you'll probably need to view it on a computer, not a phone/tablet):
http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=458&Camera=808&Sample=0&FLI=0&API=3&LensComp=458&CameraComp=808&SampleComp=0&FLIComp=0&APIComp=7
By moving the mouse over the image, you can switch between a photo taken at f/4 and one taken at f/16, with a good lens. This is on a crop-frame camera, but still only an 18 megapixel camera. Unfortunately the chart only goes up to f/16, but you can imagine the additional loss of sharpness at smaller apertures, and/or with a higher-megapixel camera.
Obviously focus stacking has limitations, the most obvious being that it only works on non-moving subjects, but if you are looking for the ultimate in sharpness, and/or to cover a wider depth of field than your lens can natively produce, it's an excellent option to consider.
Questions, as always, welcome :)
I'm curious to know when doing the multiple shots with your Canon, does your camera have a feature that assists? My Sony has a feature called “peaking" which essentially shows you the area that is in focus by highlighting it in a color such as yellow. It's very helpful. As you refocus the area changes via live view. It makes getting the multiple shots done fairly easy. One bonus for Sony
Canon don't offer focus peaking, which is slightly odd given their (pretty successful) push to get the 5D2 and 5D3 seen as a viable option for filmmakers. That hasn't stopped people from finding workarounds, of course. The Magic Lantern firmware available for many Canon cameras adds focus peaking as well as many many other features, such as fully-automated focus stacking (simply set the start and end point and the camera will take the required number of intermediate shots, adjusting the focus between each photo). Similarly, the app I have on my phone that controls my DSLR also offers a similar fully-automated stacking method. However, for this shot I simply moved the focus ring by the smallest amount possible in between photos -- not the most efficient method, but easy enough for one shot.