Monday, February 6, 2017

So just how much difference does it make?

There was a lot of interest in my comparison of brass and aluminum pinholes last week, with lots of comments on Facebook expressing definite opinions of one sort or another, and one especially got to me where a rookie pinholer wondered how to avoid "blurriness."  There were several responses that seemed to imply that the only way was to buy a commercial pinhole.

So I felt it was probably the responsible thing to do to actually take pictures with a variety of pinholes to see exactly how much and what kind of difference it does make.

My original plan was to make the best pinhole in the same size out of several materials, but I quickly found out that's pretty difficult to do with hand-drilled pinholes, and some materials are harder to work with than others.  Again, I used my standard technique of drilling through with a number 10 sewing needle, in this case held in a cork from a bottle of Scotch, against a hard table with a matchbook cover underneath the material so the needle could just penetrate enough to make the size hole I wanted, in this case .3mm, and then sanding the burr off the opposite side, and in one case, sanding the drilled-into side to make it smoother. I thought about getting a micro-drill at this size, but that would have been a mail order purchase and I could envision breaking something that small the first time I touched it. (Curious how you work with those - something like a tiny hand drill?)

So here are the six pinholes, this time rendered at the same scale. They range from about .29 to about .33mm.  I mounted them on the outside of the shutter of a 60mm 6x6 camera so I could change the pinhole between each exposure.  (Mr. Pinhole says .32 is optimum at this distance) 



They are from left to right:

1. Sheet brass .05 mm thick from a roll I bought in 1992 that has had thousands of pinholes made from it. It's pretty close to round but not quite.  It's hard to spin the Scotch cork around - maybe I should use a pencil or an exacto handle I could twirl more easily instead. 

2. Very thin brass foil.  I only had about two square inches of this stuff.  I kept making slightly oval pinholes and it was difficult to get a really perfect edge.  If I tried to smooth out the edge, I'd end up enlarging it, so this was the best one I could do.

3. A piece of beverage can aluminum (a beer can, Justin). This is the initial state which one commenter described as "craters of the moon."

4. The beverage can aluminum with the drilled-into side sanded to remove the crater rim. I was using 400 grit sandpaper, so maybe I could do better with a finer grit, but I checked at the hardware store and decided not to put $6.00 into it just for this project.

5.  One commenter mentioned that "cooking" aluminum was a fine material.  In retrospect, I supposed they meant a disposable pie pan or something like that, but I used regular aluminum foil. I tried several times to sand the burr off the back side but kept tearing it, so when I got this relatively circular one about the right size, I quit while I was ahead. From it's appearance it should be a really awful pinhole.

6. A Gilder Electron Microscope Aperture, which is in copper that's probably about .025mm. Maybe not a premium priced laser drilled pinhole, but the best I could do, and I can't image how those expensive pinholes can be rounder or smoother.

So here are the pictures. They are all half hour exposures about three feet under a couple flourescent fixtures, using Arista.edu 100 developed in Caffenol C-M.  They vary a little in density and it's impossible to get the scanner software not to do a little preprocessing, but I tried to make them as unmanipulated as possible.

1. The .05mm thick sheet brass.


2. The brass foil.


3. The Aluminum with the crater rim.


4. Aluminum with the crater rim sanded off.


5. The Aluminum foil. Notice that the bottom of the frame is slightly lower contrast than the top, something I didn't expect from diffraction at this scale.


6. The Gilder electron microscope aperture.


There's a little difference in contrast, but that may have to do with variation in exposure based on the slight variations in size.  Except for that diffraction caused fogging at the bottom of No. 5, it's a little hard to tell them apart when looking at them at this scale.

But the argument about this sort of thing was about sharpness, so let's look at some details. These are actual pixel reproductions. I scanned them at 2400 dpi, so assuming you're looking at them at "actual size" in your browser and based on the native screen resolution of my MacBook Air of 128dpi, that means these are like looking at a 42 inch print!

This is the text below that high speed camera shot on the upper left hand page of the encyclopedia (The bold text in the first line actually says "High speed camera."  Because of the variations in density, I adjusted the levels on this detail to where I thought you got the best look at the text.


The diagram of the pinhole camera on the right hand page of the encyclopeda.


The Cheerios shutter on the camera to the right.


The shutter speed dial on the Canon F-1.  In this one you can clearly see the reduced contrast in No.5.


And lastly, the 529 Graphite designation on the shutter of the Chaneloflex.


I guess there's a few conclusions that can be drawn.  

First it's a pretty robust process.  Even the ragged, three dimensional hole in the aluminum foil created a reasonable image, and with a little care you can easily drill a pinhole that most people won't be able to tell the difference from an expensive laser drilled aperture.

To that rookie pinholer that everyone wanted to buy a premium pinhole, I think your problem was a pinhole that was too big.

One thing that I sort of knew already from some experiments I did a long time ago is that diffraction is going to reduce contrast as much as it's going to reduce sharpness.

A surprise was that the aluminum pinhole with the crater rim looks like it performed a hair better than the one where I sanded it off. Go figure.

And of course, the Gilder aperture wins in sharpness, if you're into that sort of thing, and I bet if I repeated the experiment with a .15 aperture on 35mm film it would make more difference. 

But there are other considerations. Drilling your own pinhole is just more pinholey than buying one. That's kind of a snotty attitude, but it's about as close to religion as I get. And of course concept, subject matter, lighting, creative use of motion and composition can elevate an image more than simple sharpness.


2 comments:

  1. Nick, I really enjoyed reading and going through your experiment.
    This post goes in my bookmark folder, under test-research.
    If you are curious about the comparison I made, a while ago, take a look at: http://pinholemoustache.com/uncategorized/various-pinholes-a-test/

    ReplyDelete
    Replies
    1. Interesting. I think what it shows is that you can get color fringes on highlights and overall reduction in contrast, bit it doesn't necessarily reduce resolution as much as people expect.

      Delete