Friday, 14 June 2013

Overview of Samyang 8mm/2.8 and NEX-7

I'm calling this an overview rather than a review because I haven't taken any test images to give a systematic look at the performance of the lens+camera combination. I am going to give my impressions of using the Samyang 8mm/f.2.8 fisheye lens with the NEX-7 over the last few months. Over that time, I've travelled a bit and taken several panoramas which is my main use of this lens.

The lens itself is small, light, and complements the NEX-7. The aperture and focus rings have a nice feel and damping. The front element is quite a bit smaller than the Samyang 8mm/3.5 fisheye which I have in Alpha mount. The built-in lens hood protects the front element from inadvertent knocks. A clip-on lens hood and pouch completes the package.

The first thing to note is that the lens does have issues with the sensor on the NEX-7. The position of the exit pupil means that the angle of the light rays hitting the sensor is too far from the vertical leading to colour shading in the corners of the image. In this case, it is a magenta cast which can be quite noticeable on images which would otherwise be neutral. In some very strongly coloured circumstances, for instance twilight, you may not see the colour shift. Luckily this colour shading is correctable using software and a reference image. This will require converting your RAW files to DNG first before passing it through corrections tools. CornerFix is a free program which can build a correction profile from a reference image which can then be used to remap the DNG data. A similar process is used by Adobe Lightroom DNG Flat-Field plugin. These can reduce the colour shading to an acceptable level for most applications, panoramas included.

If the colour shading issue doesn't put you off, then the rest of the performance of the lens+camera combination is well worth it. Even at maximum aperture, the image is reasonably sharp, but at f5.6 (optimal for centre sharpness) the lens is stunning giving incredible amount of detail across the frame. One thing to note is that the field of focus is not flat but can be quite irregular especially at the corners. Even with the very wide depth of field inherent with using a fisheye lens, to gain maximum sharpness requires fully magnifying liveview and adjusting the manual focus. However, the effective hyperfocal range is approximately 50cm to infinity at f/5.6 when focussed around the 2m mark, and 20cm to infinity at f/8 when focussed around the 1m mark. I find that although depth of field increases at f/8, diffraction leads to a slight softening compared to f/5.6.

Overall, if you are into panoramic photography this lens should be high on your list of lenses to consider. The price is right, optically it performs very well (colour shading not-withstanding), and it's a compact and light addition to your bag.

Student Evaluations: Campbell's Law in Action

"The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor" - Cambell's Law.

I am reminded of this and Goodhart's Law by an article reporting the results of psychology research into how students rated their lecturers and how they performed in their immediate and future exams. A summary of the results is, "We find that less experienced and less qualified professors produce students who perform significantly better in the contemporaneous course being taught, whereas more experienced and highly qualified professors produce students who perform better in the follow-on related curriculum." and "that
student evaluations reward professors who increase achievement in the contemporaneous course being taught, not those who increase deep learning."

One potential explanation mooted was, "the less experienced professors may adhere more strictly to the regimented curriculum being tested, whereas the more experienced professors broaden the curriculum and produce students with a deeper understanding of the material. This deeper understanding results in better achievement in the follow-on courses."

This is a demonstration of the folly of short term drivers and the law of unintended consequences. Teaching to the test becomes the optimum strategy for teachers to satisfy student evaluation scores, yet this is not optimal for long term performance. As a junior lecturer, one does not have the luxury of being idealistic and teaching what you think the students should learn but what will satisfy exam pass rates and evaluation forms.

Unfortunately, these same pressures have been at work throughout various education systems, from primary, secondary and through to tertiary levels. By subjecting schools and universities mercilessly to the drivers of tests scores and league tables, these become ends in themselves. It leads to a "culture of low expectations", where students are not pushed to fulfil their potential.

Education is not about passing exams, nor is it a popularity contest. It is about the betterment of the individual and consequently society.

Monday, 14 January 2013

Refreshing Honesty and Integrity

Here's a story which shows that one can retain dignity in failure. Short version is that a paper in Physical Review Letters has flaws pointed out to the author, twice. The first flaw isn't fatal but the second is uncorrectable, "I hereby retract my paper [1] due to a fatal error I explained in [2]. All my attempts to patch the error have failed."

Additionally, the author writes a completely new paper explaining why the approach in the original one fails, "Since the error from the original paper [1]—which makes the proposed setup unfeasible—proved to escape immediate recognition by the physicists who considered the details of the paper I think that it would be of service to the community to comment on the approaches that can and canot be taken in attempts to reach the aforementioned goal or to prove it unreachable."

Everyone makes mistakes, but not everyone has the integrity to properly own up to them and to go the extra mile to help others making the same ones. Science is about being honest, with each other and to oneself, but especially with Nature. Simply wishing something was so does not make it true, something which is lost on denialists and fundamentalists. Admitting you were wrong gracefully is a sign of strength, not one of weakness.