?

Log in

No account? Create an account

Previous Entry | Next Entry

Worldwide Pinhole Photography Day

Still Life Still Life
Taken for Worldwide Pinhole Photography Day.
Go take a pinhole photograph today, you know you want to. Then submit it to the Worldwide Pinhole Photography Day website, sometime between now and next month.

Jesse and I modified the body cap of my Nikon D70 into a pinhole, using a 0.3mm drill bit, and a gem setting burr, (to cut down on the thickness of the cap.) Then we sealed the back of the hole using part of a broken slide cover and some hot glue. Since it only has a 23.7 x 15.6mm sensor the images from it will always be pretty blurry, even though we used the approximately optimally sized hole for the distance to the sensor. (Trades off resolution vs diffraction.)

[Reference to submittal]

Comments

( 7 comments — Leave a comment )
eub
May. 1st, 2006 01:06 am (UTC)
Well then. I poked a hole in a junk-mail credit card with a pocketknife awl, and a hole in beer-sixpack cardboard (feh, dusty) with a tack.

Really the diffraction was the best part.

katharos
May. 1st, 2006 05:45 am (UTC)
That is a pretty interesting picture, what were you shooting, the sun? It reminds me of looking through my eyelashes. Your livingroom shot is interesting too, because it shows off the shape of your aperture, was that the one you made with an awl? It's pretty distinctly not round.
eub
May. 1st, 2006 05:56 am (UTC)
Yeah, that was the sun. Easiest thing. :)

I've always vaguely wanted to digitally synthesize images recreating what you see peering through your eyelashes, since I am tragically unable to mount a CCD directly on my retina yet.

By the time of the living-room shot I was up to three holes in a tight cluster.
katharos
May. 1st, 2006 06:39 pm (UTC)
Neat! Do you know those cookie guns, like caulk guns, but with little metal plates at the end that extrude cookies in different shapes? Your three pinhole experiment makes me want a set of pinhole plates, you could have the smileyface pinhole, and the triangle pinhole, etc. Really, it would be the most fun if you could replace your lens aperture with one of these plates, then you would get better focus on the subject, and the bokeh would be creatively interesting, rather than a polygon approximating a circle.
eub
May. 2nd, 2006 04:21 am (UTC)
If you print your pinhole on slide film, it could be the Fourier transform of whatever pattern you want your bokeh to turn out as.




I was wondering if it would be useful to do this with a simple gaussian transmission profile -- if the resulting diffraction would have noticeably less spatial splatter than the one you get from a disk aperture.




Fresnel zone plates turned into multipinholes halfway down --
http://www.whizkidtech.redprince.net/zoneplate/
katharos
May. 4th, 2006 01:48 am (UTC)
Busy busy busy.

I guess I was really thinking about the shape of the blur circle rather than the bokeh, which would just come into the image as a convolution. Good bokeh is gaussian, and the Fourier transform of a Gaussian is another Gaussian, so in that case Fourier transforms are sort of involved, but in general with a pinhole wouldn't your bokeh just be a simple convolution of the pinhole shape, since there is no contribution from spherical aberration? I don't see where the Fourier transform would come in.

And, yah, if you gave your pinhole soft edges it would probably cut down on diffraction. Tried it yet?

The fresnel zone plate is pretty interesting, couldn't you make one that was multi colored using slide film? Print the ideal plate for green with magenta, the ideal plate for red light in cyan, and the ideal plate for blue in yellow? I vaguely understand how it would act as a lens, but diffraction still seems pretty esoteric to me. Now I want a multi color starfish pinhole fresnel lens... but I probably wouldn't use it.
eub
May. 4th, 2006 03:56 am (UTC)
Sorry, yeah, I was thinking ahead and got a Fourier transform back where it didn't belong. For diffraction it belongs, I have the impression -- I don't know anything about optics, but I remember in physics class the diffraction pattern of the single slit was a sinc function. So the Gaussian pinhole. I'm not sure it would really be a good thing, though: in window-function terms, the disc window gives you high sidelobe, but it does have a nice narrow mainlobe. I'm used to audio where the ear can easily pick out sidelobe crud buried 50 dB down, but the eye not so much, so I dunno.

I like the multicolor Fresnel zone plate. I wonder if dyes are available that form a neat bank of four or more stopbands, to cut down on aberration even further.
( 7 comments — Leave a comment )