PDA

View Full Version : Your favorite pinhole formula...in reverse?



derevaun
03-19-2005, 05:22 PM
I just got a set of pinholes from Wm Christiansen, so now I'm busy sorting out bodies to mount them in. They came with a chart of optimal focal lengths that match the formula that Guillermo Peñate gives (on his recently disappeared site?) Anyway, I notice that the focal lengths that are generarted on http://www.mrpinhole.com/calcpinh.php use a different formula arriving at much shorter focal lengths--by a factor of about 1:1.3. I know I can't get precise enough for it to matter, but for the sake of learning, what are your thoughts about the divergent formulae?

Tom Miller
03-19-2005, 06:33 PM
...They came with a chart of optimal focal lengths that match the formula that Guillermo Peñate gives (on his recently disappeared site?)...
Try this URL. *I think Guillermo moved his site recently. *http://ca.geocities.com/penate@rogers.com/

Tom Miller
03-19-2005, 06:44 PM
Anyway, I notice that the focal lengths that are generarted on http://www.mrpinhole.com/calcpinh.php use a different formula arriving at much shorter focal lengths--by a factor of about 1:1.3. I know I can't get precise enough for it to matter, but for the sake of learning, what are your thoughts about the divergent formulae?
I read somewhere that there were over a hundred formulae developed to determine the optimal pinhole... in the 19th Century. *So the count must be way up there by now. *My advice is: don't worry about it.

One serious reason for not worrying about a 30% difference is that an optimal pinhole size will actually be optimal for only one wavelength of light. *Visible light ranges from 390 nm to 780 nm, which is a factor of 1:2. *

taco
03-19-2005, 07:33 PM
Mr. Pinhole indicates for a focal length of 150 mm a pinhole of O,516 mm,
Guillermo 0,457 mm and
David Balihar 0,546 mm.
For me that's a deviation of ± 10%
So, I just can repeat Tom Miller;
"My advice is: don't worry about it"
taco. ;D
Anyway, even if the factor is 1,3 is nothing, that's less then a half f-stop and easy to repair in B&W neg film or paper negatives and more or less negligable with colour neg films. Only with dia films it could be critical

cnmne
03-23-2005, 01:59 PM
So what is the effect of using a "non-optimal" focal length? Is there a change in sharpness?

Tom Miller
03-23-2005, 02:23 PM
Yes, there is a loss in sharpness. *If you have Eric Renner's book, check p 173 in the 3rd edition or p 129 in the second for examples of how an image "softens" as the f-stops increase (from bigger apertures). *

I think that smaller than optimal apertures can lose sharpness as well because of relatively increased difraction making the images fuzzier. *I've never verified this by experiment. *A couple of times workshop students made pinholes so small that the exposure times were too long for them to make an exposure, at least before the workshop ended.

ImageMaker
03-23-2005, 05:03 PM
You can easily demonstrate the diffraction effect that causes loss of sharpness when the hole gets too small. All you need is a tiny and very round hole, or a straight, clean slit, a strong point light source, and a dark room (the light source should shine only through the hole or slit -- a laser pointer works very well for this, but a bright bulb shining through another small hole will also work well). With the room darkened and the hole or slit some distance from a white surface, you should be able to see the series of progressively weaker lines or circles surrounding the primary projected pinhole image of the light.

The circles or lines aren't a big deal for sharpness -- the 2nd one out from the center is weak enough not to record significantly in a normal exposures -- but the other thing that's harder to see is a big deal: the spreading of the light that forms the primary image into a Gaussian bell curve distribution. When that spreading is larger than the geometric circle of confusion from the pinhole (which, for distant objects, is the same size as the hole) at a high enough power level to record on the film or paper, it reduces sharpness.

The further the hole is from the sensitive surface, the larger the spreading due to diffraction -- and thus the larger the hole can be without adversely affecting image quality; the hole can be (for distant objects) as large as the portion of the Gaussian central "peak" (a term derived from the graph of energy distribution in the projected rings or lines) without losing sharpness compared to the tiniest possible hole. More correctly, the projection of the hole can be that large -- this is what makes it advantageous, for macro images, to make the hole smaller than the usual optimum pinhole calculations suggest, because geometry increases the size of the projection of the hole when significantly diverging rays from a nearby object pass through the hole.

In addition to distance from the recording surface, the smaller the hole, the more diffraction effect is present, because the larger fraction of the light passes close enough to the edge of the hole to diffract (an effect that can be though of as occurring when the light wave partially overlaps an edge). This is why sharpness is lost with a smaller hole, even when distance to the film doesn't change.

So, it's a balancing act -- make the hole too big, and geometry reduces sharpness; make it too small, and the wave nature of light does the same. It's (relatively) easy to calculate both of these effects for a given combination of hole size and distance to the film, as long as we assume "distant" objects with effectively parallel light. It gets much more complex when the objects are close enough that diverging light rays increase the projected size of the hole -- an effect that takes place when the distance from object to pinhole is too small relative to pinhole to film distance and pinhole size. In practice, the effect can be seen in photographs take from 2-3 times the pinhole to film distance, when optimum pinholes are used; in theory it should be more pronounced with shorter focal lengths as they use a relatively larger hole.

taco
03-23-2005, 07:26 PM
Yes, there is a loss in sharpness. If you have Eric Renner's book, check p 173 in the 3rd edition or p 129 in the second for examples of how an image "softens" as the f-stops increase (from bigger apertures).
I'm still with the first edition of Renner, I suppose you refer to the 8 photos of a girl/lady which are in the first edition on p.120 ? Which what did Renner fill nearly 50 pages between the second and the third edition, additional text or just new photos ? Is it worth to buy a new editition ?
taco

Daryl
03-23-2005, 07:51 PM
Don't buy any more books, Taco.
Spend the money on film and developing.

Tom Miller
03-23-2005, 08:41 PM
I'm still with the first edition of Renner, I suppose you refer to the 8 photos of a girl/lady which are in the first edition on p.120 ? Which what did Renner fill nearly 50 pages between the second and the third edition, additional text or just new photos *? Is it worth to buy a new editition ?
taco

I was talking about that set of 8 photos. *They're the same in the 2nd and 3rd edition. *Personally, I feel that buying the third edition is worth it. *The 3rd edition is greatly expanded. *Just about every page of the book has been re-written. *There is a lot more on zone plate, slits and camera obscura, more photos and artists. *The 3rd has color throughout and has a list of suppliers and manufacturers and even mentions Worldwide Pinhole Photography Day. *

To Daryl's point. a person doesn't have to buy a new book. *The first edition is fine as an intro and reference. *I enjoy books and seeing the new photos; the section on Michael Wesely's work was especially interesting. *I also enjoy teaching pinhole and look at as many sources as I can to learn new tricks and techniques. *So, I balance the price of the book against film... and usually splurge on both.

Daryl
03-24-2005, 09:10 PM
I like the calculator at
http://www.stanford.edu/~cpatton/phcalc3.htm for pinhole size.
It will take into consideration magnification and wavelength which is really cool for stuff like infrared.

But I still want to see pictures of Belgium ??)
and Taco is already there and has all those cameras. :)

Somebody needs to shoot more film. >:(

taco
03-25-2005, 11:56 AM
But I still want to see pictures of Belgium ??)
and Taco is already there and has all those cameras. :)
Somebody needs to shoot more film. >:(
Daryl,
have a look on this link, not many pictures but all my cameras
http://taco.thoma.be/gallery/
taco

Daryl
03-26-2005, 11:38 PM
You have a wonderful collection of cameras.

GO SHOOT SOME FILM! with 'em.
Let's see some pics.

willem
03-27-2005, 03:12 AM
Yea more pinhols more film or paper the amera is not importend I can bay all the cameraas on one day but its the question to make a good pinhole to day I cane do nothing with all this books I cane make my pinhole the hole is importend and I hope to see more and more differend pinhols its the only way HUNT and go on Taco go to school and learn to make a pinhole You tekst is some indistinctness its like a hole please send your pinhols all your pinhols are wellcome and we learn from it to show it !!!!! :K)WIIIIM

taco
03-27-2005, 11:24 AM
You have a wonderful collection of cameras.

GO SHOOT SOME FILM! with 'em..

That's my problem, Daryl. With the exception of the 120's, they are all for paper ;D
and without a changing bag it's taking them on a walk, shoot one foto with every can, go home, see most of them are not what I wanted, but then it's already to dark, waiting for the next weekend, it rains etc.
As you can see I have quite a lot never used :'(
taco

ImageMaker
03-27-2005, 07:16 PM
Well...

First, get a changing bag. Really.

Second, make more cameras for film or Polaroids so you can do multiple shots on a single outing.

And one more piece of advice, more for me than anyone else -- *STOP PROCRASTINATING!*

taco
03-28-2005, 03:33 PM
Second, make more cameras for film or Polaroids so you can do multiple shots on a single outing.
And one more piece of advice, more for me than anyone else -- *STOP PROCRASTINATING!*
Promised, but also promised that I'll put only THE VERY GOOD ONES on this forum, the rest will end up in my own gallery >:(
taco

JoeVanCleave
04-01-2005, 06:29 PM
Aside from the gaussian spread of light caused by diffraction, there's a strictly geometrical effect causing additional image spread, for subjects that are rather close to the pinhole.

I've notice this effect on my little 2" square format brass camera, when shooting close up images of toy soldiers (as was pointed out in a different thread). As I recall, the distance from subject to pinhole in that shot was about 1.5 inches, about the same as the camera's focal length.

So, my personal BKM (Best Known Method) is when designing a camera that I know will be used often for close-up subjects, to make the pinhole "somewhat smaller" than that calculated by the Rayleigh criteria.

This implies that our beloved pinhole cameras really don't have true "infinite" depth of focus; rather, they have "pretty darn good" depth of focus. But you should optimize them for either landscape or close-in shooting.

taco
04-01-2005, 07:08 PM
This implies that our beloved pinhole cameras really don't have true "infinite" depth of focus; rather, they have "pretty darn good" depth of focus. But you should optimize them for either landscape or close-in shooting.
What's "optimized" for landscape :o
taco

JoeVanCleave
04-02-2005, 11:24 AM
Conventional formalae for calculating optimal pinhole diameter seem to be optimized for subjects near infinity, such as in landscape images - at least it appears that way to me. If I were building a camera intended to mainly shoot landscapes, or other "far-off" subjects, diffraction is the limiting factor to image sharpness, and the formulae we already use do suffice.

But for close-in subjects, there's the additional image spread caused by geometrical effects not related to diffraction. The closer the subject is to the pinhole, the more these geometrical effects dominate, to the point that diffraction effects are secondary. To build a camera optimized for close-in subjects would require a pinhole smaller than that normally considered "optimal".

I'm not necessarily talking about a pinhole microscope, but if one were to build such a camera, a very, very small pinhole would be necessary. This implies more issues with reciprocity, which implies the use of careful calibrations with fast film, and bright lighting on the subject.

The close-in shooting I've done using outdoor daylight suggests that the camera's shadow becomes a big problem with proper illumination onto the subject. I usually try to side-light the subject in this case. But for indoor sets or dioramas, bright artificial lights and extended exposure times would be necessary.

ImageMaker
04-02-2005, 02:00 PM
Macro photographers using lenses, who also have problems with flash coverage and camera shadow, have often used a "ring light" -- a xenon flash tube arranged in a circle around the lens to case an even, if very flat light on a subject close to the camera. The effect could be approximated with multiple small flashes arrayed around the front of a camera, on converging angles if necessary to ensure coverage of the subject region.

JoeVanCleave
04-03-2005, 04:36 PM
I generally dislike the math-intensive approach that some use with pinhole, as I enjoy the process flow when its more intuitive.

But to violate my own principles, here's a sketch of a simple ray-tracing layout for a point source of light (S) at the object. This sketch involves the geometrical effects of pinhole only, not diffraction effects.

"F" is the camera's focal length, and "D" is the distance from the object to the pinhole. "P" is the diameter of the pinhole itself.

The size of this point source of light at the film plane is "i", as indicated by the formula i=P*(F/D+1).

There are three conditions of image size as its related to object distance (D):

1) Where D approaches infinity (ie faroff landscape subjects): the image size "i" approaches the pinhole size "P". This is the condition of the pinhole camera optimized by the Rayleigh criteria, and used for distant subject matter.

2) Where D equals F, the camera's focal length. At this distance, the image "spread" of a point source of light from the subject will be rendered as 2*P in size at the film plane. There will be no detail in the image smaller than twice the size of the pinhole. Objects at this distance will begin to be noticably less sharp than those further away.

3) Where D is less than F. At these close-in distances, image spread becomes even larger, being greater than 2*P. To render these close-in subjects as sharp as those further away, a smaller pinhole diameter "P" has to be used. Attached files http://f295.f295.org/uploads/drawing005_3968.jpg (http://f295.f295.org/uploads/drawing005_3968.jpg)