Who says close-up says limited depth of field. So, first solution:
With the more than 10 Mpx recent sensors, one can still crop the picture at post-production.
Yes, the smaller the sensor, the less important is the depth of field problem. Naturally, this smallness is paid for by a lower image quality, especially in low light. Conversely, the phenomenon is more important with a “full-format” SLR (i.e. in 24 × 36 format).
When the diaphragm is closed, the depth of field, and therefore the sharpness, is increased. But if one exceeds f-number 11 or 16, a parasitic phenomenon degrades this sharpness: diffraction. It is therefore better to be satisfied with f/5.6 to f/11. But then, the depth of field remains very limited (a few centimetres to the best).
This so-called focus stacking method can only be used with a camera with the possibility of manual focusing, and mounted on a tripod.
You take several photos, modifying focusing distance between each. A series of pictures is thus obtained, the focusing plane (sharpness) of which varies progressively from front to back or vice versa. The software will automatically search for the optimal sharpness of each image and merge all of them into a single image.
It sounds simple, but it is not so much: between each photo, there may be small shifts of framing (even, and it is mandatory, if you work on tripod), because, when you change the focus, the image magnification changes a bit. In addition, there may be small variations in colour and exposure.
With this process, it is possible to photograph three-quarters a ten coach train, hence about 3 m long in HO, sharp from one end to the other!
A search on the Internet with the keyword “focus stacking” shows that there are many image stacking software to increase depth of field. See a test unfortunately rather old (2011) by Helisud on eos-numerique.com. Among this, we find:
See also an article where the principle of focus stacking is well explained, as well as the other methods to get the best depth of field possible.
I use or have simply tested the following software: CombineZP, Helicon Focus and Zerene Stacker.
CombineZP is a totally free software written by an English academic
who use (used?) it for micrography
(go and see his insect photo
galleries). But of a very special ergonomic! Operation is very slow.
The results are often correct, sometimes very bad (it depends partly on
the macro — method — chosen, but because of the slowness, it does not
incite to test several ones). So, rather difficult to use, but has the
merit of being free!
Helicon Focus is a software that works correctly, although the help is (was) not very explicit about the different working methods possible. I find the latest version interface less user-friendly than before, making handling less easy. Helicon Focus offers a retouching tool, useful when the software does not find the sharpest area among all the files in the stack. Moreover, it allows to automate the process of multiple shooting with certain cameras (Nikon and Canon, as by chance!): it is sufficient to program the minimum and maximum distances and the number of views; the software calculates the distance between each view and drives the camera via the lens focus motor!
I recently tested this software, but it did not look better than Helicon Focus, or even CombineZP (but I could not try on very contrasted images, something that completely disrupts the latter). Zerene Stacker is much slower than Helicon Focus. Its interface is more (too much?) sober and some functions are rather unclear: it is necessary to consult the help, which is clear. Finally, this software has a retouching tool, like Helicon Focus, but more rudimentary.
I own a Pentax SLR (K10D and now K3) mounted on a tripod. I use the standard zoom, not excellent but sufficient. It is interesting to note that this lens, acquired with the K10, gives better results with the K3, not only in terms of image acutance, but also with regard to chromatic aberration. I made a white mark on the focus ring. After configuring the camera in manual focus mode, I search through the viewfinder for the minimum and maximum focus distances. I add a safety margin, then I take the series of photos at f/8 or f/11 by shifting the focusing ring about 1 mm between each shot, starting from the maximum distance — locating the positions relative to the zoom ring corrugations.
Tip: use a cable release or wired or wireless remote. It can be found at less than €5 on the Internet (at this cost, one can wonder about quality…).
Here is a result, using CombineZP:
Old Jouef rolling stock: 141 R 1264 ref. 8272 lightly
and Est metallized cars aka “Romilly” ref. 5102, 03, 04.
Click the image to enlarge it.
If you look at the enlarged picture above, you’ll see that the front of the loco buffer beam (left buffer and SNCF logo) is blurry: it is because I approached the camera below the minimum focusing distance (0.25 m).
Note: the cloudy effect in the background is not voluntary. The software did this by chance, combining the different blurry backgrounds!