Posted by Noah Aboussafy on
I just got one of thos 10X macro filters from my friend Diego (more correctly it’s a supplementary lenses, and it’s +10 diopters not 10X.) I did a little testing with it today. It does cause pincushion distortion, and some color fringing at contrast boundaries as expected. However it did do something I really didn’t expect.
I tried it out with my Nikon AF-S DX Nikkor 18-55mm f/3.5-5.6G VR; a lens with a 0.31× reproduction ratio, and a claimed minimum focusing distance of 28cm.
This first sample (shown below) was taken at ~26.25cm; and I wasn’t quite as close as I could get. I’d bet I could have gone right down to 26cm. That would be 7% closer than listed. All of these samples were taken at f/8.
For those of you that don’t know the hansom fellow is Sir Wilfrid Laurier, the 7th Prime Minister of Canada (as pictured on the Canadian Five Dollar Bill.) He was a pretty cool guy, worth reading a bit about.
I screwed on the 10× supplimentary lens and moved in as close as it would focus. This is the full frame: (notice that the bill is in focus, and face of the quarter is not, how’s that for narrow DoF! Next time I’ll remember to stop down haha.)
And this is a 50% crop.
I didn’t post a 1∶1 crop for several reasons, the first is the bill just doesn’t have enough detail to bother, it just becomes an inky mess. The second is because all context was lost, it would be hard to tell what you were seeing. The third is because the result is dependent on the resolution of the camera, making the filter seem to produce very impressive images, when in reality it’s just a large file. The fourth is because it wasn’t razor sharp; but certainly I didn’t expect it to be as good as it was though! It wasn’t as good as you’d get with a high quality macro lens, but hey I don’t have $500+ right now.
The Nikon D90’s sensor is 23.6mm across, with 4288px effective pixels, though I have no idea what the effective horizontal dimension is. Thus 1” horizontally should take up ≈3298px, in my quick and dirty test it took ≈3202px. That’s 97% of the expected size, and that’s that’s close enough to say 1∶1 in my books.
Caveat lector, I used the filter it on top of a collapsable rubber lens hood which acts as a filter stand-off, and may have effected the results very slightly. I did test the filter directly attached to the lens, the only difference I could detect was that the assembly’s minimum focusing distance was slightly closer to the subject without the standoff (as mesured from the sensor.) Another point of interest is that the lens isn’t cemented into the mount perfectly on-axis, and as a result the feild of focus is slightly diagonal.