Talk:APEX system

Light value system ?

If I understand things correctly, the APEX system is a development of the Light Value Scale, an exposure system proposed by Friedrich Deckel in the 1950s. The Wikipedia entry for Light Value Scale redirects here, so I would expect to find some information. That seems to be missing completely.

The exposure system we use on cameras today was developed in a few steps. Should this be the page that describes the historical evolution of that system (including the Light Value Scale and APEX)? WoutR (talk) 17:37, 1 December 2016 (UTC)

I think it would be great to cover this. But it seems to be awaiting someone with the knowledge and supporting sources. This issue was mentioned about nine years ago under Talk: Exposure value#Hasselblad gets all the credit?.

The Light Value System/Exposure Value System was apparently developed by Deckel, and adopted by several other German shutter manufacturers. It was also used by Kodak and Polaroid—and perhaps others. Polaroid initially used a scale that was offset by several steps from the current exposure value, but changed to the current scale on later models. APEX—proposed in ASA PH2.5-1960—went a bit further, extending the concept to the right-hand side of the exposure equation. In ASA Z38.2.6-1948, the exposure equation (“calibration formula”) was stated as

$T={\frac {KA^{2}}{BS}}\,.$

ASA PH2.5-1960 rearranged this to the form given in the article; perhaps the rearrangement was influenced by the concept of exposure value. But given my lack of direct knowledge—or citable documentation—I don’t think I could reasonably add this to the article.

There is anecdotal evidence to support all of this. Deckel’s 1958 US patent strongly implies that Deckel developed the concept of light value/exposure value, and several later Deckel patents that cite the 1958 patent seem to suggest the same. A Google search finds a few articles that also support the concept, including this article in Popular Photography and this article in the Arizona Repbulic. Polaroid user manuals that cover use of exposure value can be found on the Web; they’re probably a useful element, but only part of the picture. Doug Kerr wrote a paper discussing the history of the exposure value system; he knows what he’s talking about, but I doubt he’s citable under WP guidelines.

I recently added citation of Kurt Gebele’s patent (assigned to Deckel) to the Exposure value article, hoping that, in combination with the the citation of Sidney Ray’s article, it would suffice to support the claim that Deckel developed the concept of exposure value; though not the most solid citation, I thought it was preferable to nothing. But I don’t know how much further this approach can reasonably be taken here. Perhaps I’m rejecting the good for the perfect, but I think some care is indicated. The bulk of the hits from a Google search for “exposure value” or “light value” are largely wrong, and most—unsurprisingly—provide no sources.

Another issue is where this topic should be covered. A more logical redirect for Light Value Scale might be to Exposure value, because the Deckel system doesn’t appear to have encompassed the additional elements of APEX. Though it doesn’t specifically mention the Deckel Light Value System, the Light value article touches on the topic (item 6 as of this date); again, it’s light on reliable sources.

JeffConrad (talk) 03:26, 2 December 2016 (UTC)

I think Doug Kerr is citable. It's probably the best scholarship available on these topics, even though self published. Dicklyon (talk) 03:48, 2 December 2016 (UTC)

I dunno. I’m not sure Doug is any more citable than you or I, and the community seems to have decided that we are not. I never really maintained that we were; I′d probably consider all of us suitable for inclusion as external links, but I see that as a step below a citable source. One issue with Doug is that he’s not strong on citing sources. In the referenced paper, he has no specific citations, and credits Al Olson of the Large Format photography forum for some of the content. I think Al is pretty well regarded, but I’m not sure of the sources on which he relies—so citation of Doug would in some cases be at least doubly indirect. The topic was discussed on the LF forum in this thread; some of the missing information (e.g., the relationship between Polaroid values and the ones currently used) is covered in the Polaroid manuals (to some extent, Doug addresses this as well). Were I to choose, I’d probably first look to the article in the Arizona Republic and the Polaroid manuals. Though not as comprehensive as Doug’s paper, both sources would probably comply with WP:RS. An additional good point about these sources is that they are contemporaneous rather than retrospective (consider, for example, the differences between Ansel Adams’s early and late accounts of the making of Moonrise, Hernandez, New Mexico). Though synthesizing an answer might border on WP:OR, I’m not sure it would be worse than any other approach that currently seems to be available. I might include Doug’s paper as an external link; we already include a link to his APEX paper.

As I had suggested, I think this might better fit Exposure value. If it could be done properly, it might help clarify how EV is actually used, including the connection between camera settings and lighting conditions. With Polaroid (my dad had a Model 95B, among others, as I recall—alas, I was young and didn’t really pay attention), the process was pretty simple: measure the scene with the meter, and transfer the indicated exposure value to the camera. In essence, the process was the same with other cameras and meters that indicated in EV. In looking at the article several years after doing anything to it, I’m not sure we make that clear—and I’ll even take the bulk of the credit for the failure to do so.

JeffConrad (talk) 08:49, 2 December 2016 (UTC)

I’ve posted some initial ideas about how to cover this in the Exposure value article; it seemed easier to expand on what is already there than repeat everything here. It might also work under Light value, perhaps abbreviated.

JeffConrad (talk) 04:27, 4 December 2016 (UTC)

Thank you for looking into this! WoutR (talk) 19:35, 11 December 2016 (UTC)

Redundant?

Does anyone else find 'APEX system' (Additive system for Photographic EXposure system) redundant?

I think much of the material on exposure value should go in the entry for that topic. Perhaps most of the discussion of the exposure equation should be in a separate entry for that topic.

I removed the reference to "light value" because the term has had so many different meanings that it essentially is meaningless; I added all credible meanings of which I am aware to the entry for light value.

Any discussion of APEX is necessarily historical—although the concept was technically interesting, APEX essentially was stillborn. I'm still not completely sure when the use of past (vs. present) tense is appropriate.

JeffConrad 08:03, 19 July 2006 (UTC)

Edits of 19 October 2006

There appears to be a bit of a conflict in EXIF 2.2 among $K$, $N$, and the stated value of $B$ for Bv = 0; no manufacturer that I know of currently uses $K$ = 10.7. The standard doesn't explicitly state that $N$ = 1/3.125, but the implication is strong. I've speculated that it may be due to rounding, but it may just be sloppy copy editing (it seems strange to state that $B$ is in cd/m^2 yet give the example in footlamberts). The description of the APEX quantities in general is pretty cryptic; it also seems strange not to state or even give guidance on the constants $K$ and $N$.

I've tried to simplify the EXIF section (do we really care whether the data are recorded as rational numbers or unsigned short ints?). I changed the formula to force TeX because the HTML rendering has the annoying habit of appending hyphens to some <math> . . . </math> sequences. I also moved the EXIF link to the References section for consistency with the other references.

The last paragraph in the EXIF section still needs work. JeffConrad 10:16, 19 October 2006 (UTC)

Reading the EXIF 2.2 specification yet again, it would appear that Bv and Sv are quantity symbols that weren't properly typeset, so I have changed them accordingly. Because the Use of APEX values in EXIF section isn't strict homage to APEX, I have followed ISO 31-0 and used upright type for the descriptive subscript v. The seemingly redundant extra curly braces were added to inhibit the dreaded appended hyphen. JeffConrad 02:51, 20 October 2006 (UTC)