The Tests:
Test results are given below, but before examining them, please note that the interpretation of the images depends on the angle of coverage achieved. Extremely wide-angle lenses, i.e., the Inon UWL-100 with its dome-lens fitted, cannot be expected to produce the same apparent corner sharpness as lesser lenses, the reason being due to curvature of field (the zone of sharpest focus is not flat). Also be aware that a low magnification ratio corresponds to a large angle of coverage, i.e., a 0.56x lens is wider than a 0.58x lens.

In each case, a photograph was taken with the camera zoom-lens set at full-wide (35mm equivalent of a 35mm lens), with the aperture set at f/8. Numerical values for geometric distortion and chromatic aberration were then obtained by correcting for these effects using the Panorama Tools radial correction function. In this way, the manufacturer's claims regarding correction for chromatic aberration can be verified, and the lenses can be ranked in order of "fish-eye-ness". Note that the radial corrections obtained are not meant to be definitive (more time spent adjusting the coefficients might have produced slightly better results), but they are good enough for the present purpose, and probably good enough for the majority of people who own a C-5050 camera. Similar results, incidentally, are to be expected using the C-5050 in any housing with a flat lens port (e.g., Ikelite), a small difference in optical path length having no significant effect.

Discussion:
The C-5060 zoom lens set at maximum-wide gives a small amount of barrel distortion in air, but this distortion is almost completely corrected underwater by the pincushion distortion introduced by the flat port. This results in a very small third-order radial correction coefficient ( b=0.005) to remove the residual pincushion distortion. This happy coincidence, incidentally, is almost certainly accidental. Compact camera zoom lenses are not designed with underwater ports in mind, and other cameras will not necessarily turn out to be rectilinear underwater. The flat port moreover, does introduce some chromatic aberration (as it must), as indicated by the shift in the first-order coefficient (Dd) for blue.

When correcting a supplementary wide-angle lens for chromatic aberration, the starting point is something of a moving target because it depends on the degree of correction of the camera lens. When the supplementary lens is designed with no particular camera in mind, the only logical approach is to try to reverse the chromatic aberration introduced by the flat air-water boundary; and so we should expect to see less chromatic aberration than for the camera and port alone (if the lens maker has used a correction scheme), provided that the camera is reasonably well corrected. In this case, the camera lens is very well corrected, and the d-coefficient shifts (Dd values) indicate that all of the supplementary lenses produce an improvement, even though only the Inon lens is claimed to be an achromat. In fact, we might have guessed that all of the lenses tested would be achromats, since all have more elements than they need for the degree of barrel correction achieved. We should say however, that the Inon UWL-100 with its dome lens fitted, achieves a remarkably high degree of correction in view of its 128° nominal coverage.

If we leave the Inon lens with the dome aside for a moment; there is little difference in performance between the four remaining lenses. The Sea & Sea lens shows the best rectilinear correction, as indicated by the quantity a+b+c, but this benefit is offset somewhat by the fact that this lens is no longer manufactured (unavailability is a serious practical disadvantage). The UN PWC-01 has slightly less chromatic aberration than the others in numerical terms, but the overall aberration when used with the C-5050 is of the more noticeable red-cyan variety (see corner detail images above), and the other lenses offer a greater angle of coverage. The Epoque DCL-20 and Inon UWL-100 are practically identical in behaviour, except that the Epoque lens gives slightly greater coverage (and is the cheapest of the bunch). The indications are therefore that there is no great reason to favour one lens over any other (especially if the user intends to carry out radial correction); except on the basis of price or (in the case of the Inon lens) the ability to accept an add-on dome lens in order to achieve super-wide-angle coverage. In this latter respect, the UWL-100 with dome is a thoroughly remarkable device, and it is important to understand how different it is from the other lenses in order to appreciate its merits. The apparent low resolution of the corner detail is due to curvature of field, which is actually exhibited by all of the lenses*, but exaggerated in the ultra-wide photograph because, although the test-card looks distant, the camera was only a few inches away from it. With the lens pulled back to photograph a very large object, this curvature of field effect will be compensated by the depth of field. Hence, while the Epoque DCL-20 is the best choice (amongst M67 lenses) for a general-purpose wide-angle adapter, the Inon UWL-100 and dome port is the lens of choice for photographing such subjects as giant marine animals and whole shipwrecks.

* The effect of using a wide lens aperture is shown in a set of supplementary lens tests which were carried out using an Olympus m410 camera in an Ikelite housing. These pictures were taken using an aperture of f/3.1 and show a pronounced softening of focus at the edges in close-up wide-angle pictures.

In closing, it should be pointed out that none of the lens setups tested actually requires radial correction for normal purposes. All are good enough to produce large (A4) prints, and the fish-eye effect is an accepted feature of underwater photographs. Radial correction however, can be used to remove fisheye distortion if so desired; and it may also be used selectively to remove only the chromatic aberration.