|
|
Post by Count Zero on Jul 19, 2005 3:15:09 GMT -4
When an object in a photograph is overexposed, it appears as a blob who's apparent edges extend beyond the actual surface of the object. As an extreme example, photographs of distant stars can look like circles many thousands of times larger than the actual stellar disk. Is this phenomenon the same thing as "emulsion bleeding," as discussed in the cross-hairs section? Someone asked me about the so-called "masked cross-hairs". I explained to him that an overexposed object looks more like a blob. The astronauts set their cameras to the correct exposure settings for the lunar surface, which is dark grey. This meant that white objects in the field of view were slightly overexposed, and the associated blotchiness obscured the cross-hairs. Was this correct? |
|
|
|
Post by JayUtah on Jul 19, 2005 9:14:25 GMT -4
Overexposure and emulsion bleeding are not exactly the same thing, Overexposure is a more general term that simply describes allowing too much light to strike the film. Emulsion bleeding is one possible outcome of overexposure.
Your explanation is correct in that reducing the exposure would correctly render the fiducials.
The most blatant symptom of overexposure is simply saturation; the film reaches the limit of its ability to respond to light, and so you lose some of the subtle grades between colors. This means a loss of detail and contour.
Some films are susceptible to emulsion bleeding, in which adjacent emulsion grains also become excited when nearby grains are saturated. This is likely what causes the fiducials to become faint. The fiducials are very tiny. Even properly exposed, they cannot easily be seen on the transparency with the naked eye.
Bleeding can also occur as a result of the optical properties of the lens. All lenses scatter, although good lenses scatter less than bad lenses. So when a small point of light like a star is exposed, and the exposure time is lengthy, some light is guided to adjacent portions of the film because it scattered somewhat as it passed through the lens and did not precisely focus. That is the cause of "rays" and similar effects around bright (fully overexposed) spots on film.
|
|
|
|
Post by Kiwi on Jul 19, 2005 9:37:30 GMT -4
I no longer have the technical books I did when a professional photographer, but I think the term that applies here is "halation." Here's one link that illustrates it (page down to "Anti-halation") : www.sapiensman.com/photo/Photographic%20tems.htmFilms of the 60s and 70s most likely didn't have anti-halation layers that are as good as modern ones, and there's also a possibility that transparency film may have behaved differently to negative film. Generally, halo diffusion isn't particularly noticeable, but in the lunar surface photos it is the size of the fiducials' stems, one-tenth of a millimetre, that make it noticeable. Light only has to produce halo diffusion extending one-fortieth to one-twentieth of a millimetre from both sides of a fiducial stem to produce a noticeable effect. Edited to add: This site shows how to produce the effect yourself: www3.telus.net/summa/moonshot/index.htm |
|
|
|
Post by JayUtah on Jul 19, 2005 18:39:24 GMT -4
Diffusion halation is certainly possible, but that's not affected by anti-halation coatings or backings. I can produce degraded fiducials with modern professional E-6 film that has excellent anti-halation properties, at least against reflection halation. If reflection halation had been an issue, the fiducials near the image edges would be more affected than those near the center, because reflection halation increases away from the optical axis.
The fact that we can reproduce the results with CCD-based image sensors suggests there is a lens behavior component to this too. Scatter from the optics is less noticeable in normal photography because fine details are not usually so dramatically contrasted. But it would also help explain the problem.
|
|
|
|
Post by Joe Durnavich on Jul 19, 2005 22:25:55 GMT -4
The fact that we can reproduce the results with CCD-based image sensors suggests there is a lens behavior component to this too.
With CCDs, isn't it the case that electrons in overexposed pixels spill over into adjacent pixels? If so, then there is spreading within the CCD chip itself.
|
|
|
|
Post by craiglamson on Jul 19, 2005 23:44:03 GMT -4
The fact that we can reproduce the results with CCD-based image sensors suggests there is a lens behavior component to this too. With CCDs, isn't it the case that electrons in overexposed pixels spill over into adjacent pixels? If so, then there is spreading within the CCD chip itself. Are you talking about sensor blooming Joe? If so the lens plays a big part in reducing or eliminating blooming. Not all lenses are created equal and wides seem to suffer blooming to a greater degree than the rest. I'm a really wide kind of guy because my normal subject matter requires it. Even decent glass like the Canon 17-40L blloms like crazy, while a third party Sigma 12-24 is almost bloom free. Now I'm a cmos user and not a ccd user but in just plain overexposure I'm not seeing a lot of spill to other pixels, just total loss of detail (blown highlights) in overexposed pixels. |
|
|
|
Post by Joe Durnavich on Jul 20, 2005 0:16:52 GMT -4
Are you talking about sensor blooming Joe?
I don't know what I am talking about, Craig! I thought or assumed that some sort of spill effect happened in the CCD sensor, but I don't know enough about them to be sure.
|
|
