Talk:Exposure value

EV—illuminance conversion
I don't know what was intended by the Ray 1993 reference, and I can't find a match in the Library of Congress database. In any event, the revised values were not in accord with the formulas (which are correct and in agreement with ISO 2720 and many other sources). The previous values were also in agreement with two different Minolta meter manuals, so it's quite unlikely that they're wrong. Without seeing the claimed source, it's tough to determine why it might differ from the previous version of Table 3 in this article, but something is amiss. Regardless of the values in the claimed source, they cannot reasonably be put in the table without reconciling differences with the formulas and other sources. JeffConrad (talk) 02:56, 13 January 2009 (UTC)


 * The ref was confused, mentioning both 1993 and 2002. The Ray 2002 has a bit on this page that might be the basis of different numbers. I haven't quite interpreted it yet. Dicklyon (talk) 03:56, 13 January 2009 (UTC)


 * I think the editor's confusion arose simply from copying and pasting another citation of Ray (2002).


 * Figure 66.2 on p. 592 of Ray (2002) is not easily interpreted ... There's some confusion in the figure caption, which states that the luminance is in apostilbs, or lumen m−2; the latter is a measure of illuminance rather than luminance. But the values appear to be for luminance, measured in apostilbs (asb; 1 asb = 0.3183 cd m−2), using an ISO arithmetic speed of 12.5 and a calibration constant of 12.5 rather than illuminance. Then


 * $$\mathrm{EV} = \log_2 \left( \frac {LS} K \right ) = \log_2 \left ( 0.3183 L \right ) = \log_2 L - 1.65$$


 * Thus EV 5 corresponds to a scene luminance of 100 asb, as is shown in the table. So the table does seem to be for luminance, and inapplicable to the previous edit.


 * Apologies. It was me what done it. My copy of Ray, et al.'s Manual of Photography is the 1993 print of the 1988 edition. I was trying (with humiliating ineptness) to hint that page numbers might be different whilst guessing that the data has not changed in the newer edition ... my bad. Sorry for the confusion. The table of data I referred to is entitled Table 20.3 Photometric units and camera exposure. Approximate corresponding values of various units of luminance and illuminance related to exposure value (EV) for camera determination with a film of speed ISO 100/21° and plots EV (-2 to 18) against cd/m2, stilb, apostilb, cd/ft2, lambert, foot-lambert, lux, phot, and lm/ft2. --Redbobblehat (talk) 23:29, 4 February 2009 (UTC)


 * I think Table 20.3 in Ray (1993) is essentially the same as Table 19.3 in Ray (2000). JeffConrad (talk) 06:02, 5 February 2009 (UTC)


 * I believe the reverted data were taken from a table equivalent to Table 19.3 on p. 323 of Ray (2000). The header for the rightmost column indicates that the illuminance values give an EV for 100% reflectance, and that appears to be the problem. The values in Table 3 differ from those in Ray (2000) by a factor of 6.25; dividing 100 by this gives a reflectance of 16%, very close to the reflectance assumed in relating the luminance and illuminance values in Table 3 (reflectance is given by πK/C; with K = 12.5 and C = 250, reflectance is 15.7%). Consequently, those data are inapplicable to this article. JeffConrad (talk) 06:24, 13 January 2009 (UTC)


 * If we use a similar format for the formulae, defining, comparing and explaining K and C should become a little easier.
 * L = K/S * 2^EV and
 * E = C/S * 2^EV
 * Where L = luminance cd/m2; E = illuminance lm/m2; S = ISO sensitivity. --Redbobblehat (talk) 23:29, 4 February 2009 (UTC)
 * Ray (2000 p315) hints that K = Key-tone: Measurement of a key tone: A subject zone of intermediate luminance, i.e. a midtone, is selected and located at a fixed point on the characteristic curve. Typically an exposure 10 to 16 times more than that for the shadow or speed point is given. This key tone may be a Caucasian flesh tone (about 25 per cent reflectance)... [or] a medium grey card (18 per cent reflectance) is held so as to receive the same illumination as the subject it is replacing... The ‘arrow’ setting of the original Weston meter was calibrated for such a mid-tone, as are most metering systems. ... "exposure 10 to 16 times over speed point" encompasses the K=12.5 value. Redbobblehat (talk) 12:48, 5 February 2009 (UTC)


 * I don't see any such implication. The key tone is a simplified application of the Zone system. The calibration constant K was empirically determined to give the “best” exposure from measurements made with wide-area averaging meters. JeffConrad (talk) 21:17, 5 February 2009 (UTC)


 * (Incidentally, it may be worth noting that Ray, Fujifilm and others use log10(H) for illuminance rather than log10(E). My guess is E is the proper notation for illuminance, and H has crept into photography from mis-translating Hurter-Driffield's "HD curve" as a plot of H (for illuminance) against D (for density) :) --Redbobblehat (talk) 23:29, 4 February 2009 (UTC)


 * I'd be surprised to see a plot of D vs. log(illuminance), because it wouldn't make much sense. I've never seen other than D vs. log(exposure). JeffConrad (talk) 06:02, 5 February 2009 (UTC)


 * Yes I see my error. Exposure = Illuminance * Time. So if H = exposure and E = Illuminance, ... The Illustration (H&D_curve.png) at Hurter–Driffield_curve shows Exposure as "logEC.M.S" ... any idea what C.M.S stands for? Cock-ups per Moron per Second? --Redbobblehat (talk) 12:48, 5 February 2009 (UTC)


 * Pretty close. Probably candle-meter-seconds, where a "candle meter" is a bastard illuminance unit analogous to a "footcandle", I think – that article says "Some sources do however note that the "lux" can be thought of as a "metre-candle" (i.e. the illuminance cast on a surface by a one-candela source one meter away).".  So think of it as Lux seconds. Dicklyon (talk) 20:42, 5 February 2009 (UTC)


 * I won't get into the unit's parentage, but C.M.S. is a mighty unusual way to state it; far more common was meter-candle seconds, or mcs—long ago, Kodak film characteristic curves gave exposure in mcs (strictly, they gave log(exposure) in log(mcs)). As Dick mentioned, it's the same as lux seconds. JeffConrad (talk) 21:17, 5 February 2009 (UTC)


 * After reading http://www.natmus.dk/cons/tp/lightmtr/luxmtr1.htm I concluded that Ray's data correlates with C=πK and K=12.5. The C=πK proof is worth pursuing not only as a really easy formula, but more importantly to determine the "application caveats" for valid luminance values deduced from illuminance readings and vice versa. An article by Schlyter (http://stjarnhimlen.se/comp/radfaq.html#6) appears to suggest that converting cd/m2 to π lm/m2 encompasses spherical geometry and lambertian reflection. Although I don't fully understand the maths, I will attempt to summarise. SI unit definitions state: 4π lumens = 1 candela = 1 lumen/steradian. Reflection at a surface occurs only into a hemisphere (so 2π lm/cd) and then Lamberts Cosine Law kicks in (as a reflectance averaging?) to arrive at (π lm/cd) C=πK. In trying (unsuccessfully) to understand the lambertian step, I found the average value of the cosine of all angles between 0° and 90° tends towards 2/π = 0.6366 which might be related to the 6.25 factor? I must emphasise that maths is not my thing and I am out of my geometrical depth, so any corrections and clarifications are most welcome.--Redbobblehat (talk) 23:29, 4 February 2009 (UTC)


 * C = πK applies only to an object of 100% reflectance. The more general relationship, using consistent units (such as SI) is


 * $$C = \pi \frac K R$$,


 * where R is the reflectance. With R ≈ 15.7%, C = 20K, consistent with C = 250 and K = 12.5 used in the article. Bear in mind that C = 250 is for a flat (cosine-responding) receptor rather than a hemispherical (cardioid-responding) receptor usually used to meter a three-dimensional subject, so you can't meaningfully compare the illuminance values in Table 3 with Table 2. The main value of the relationship of EV to luminance and illuminance is for making approximate photometric measurements with a photographic exposure meter. There is no need to normalize values in either table. The relationship between C and K is discussed in detail in the Light meter article, and in much greater detail in the paper Exposure Metering: Relating Subject Lighting to Film Exposure linked at the end of that article. JeffConrad (talk) 06:02, 5 February 2009 (UTC)


 * As far as I can tell, C=πK makes no allowance for arbitrary average reflectance (16%, 18% etc) so it should be used to reference a white surface (eg snow). If so, this is definitely noteworthy (as an "application caveat"). Luminance and Illuminance values in table 3 should be normalised with table 2 values (or vice versa :). It is not currently explicit, but I would guess that table 2 values represent average scene luminance ≈ 18% reflectance. --Redbobblehat (talk) 23:29, 4 February 2009 (UTC)
 * Stop press! the Luminance value is already reflected and so L effectively includes the 2 reflectance factors: the 15+/-3%avg and the cosine diffusion. To derive C from K, both of these need to be included: so C=πK/p or K=Cp/π, where p is average reflectance due to surface colour. Phew. My diffusion has lifted and my formulae are beginning to look distinctly Cyrillic ... :) --Redbobblehat (talk) 03:56, 5 February 2009 (UTC)


 * The reflectance p isn't necessarily related to surface color—it applies to a neutral surface (such as a test card) as well. Whether or not you need the factor π depends on the units of luminance; you need it with SI units like cd/m2, but not with comparative units like the lambert or footlambert. The easiest way to get the relationship between C and K is from the meter calibration equation. JeffConrad (talk) 21:17, 5 February 2009 (UTC)


 * Table 2 gives recommended camera settings for the best picture; most derive from average luminance of sunlit outdoor scenes, but some (e.g., a cityscape at night) simply derive from experience. Incidentally, typical snow reflectance is only about 90%, so it's not well represented by an object of 100% reflectance. JeffConrad (talk) 06:02, 5 February 2009 (UTC)
 * I have been using Magnesium Carbonate blocks for my tests. The have a p=98.4%. Kq6up (talk) 18:53, 11 June 2013 (UTC)


 * Incidentally, I trust you noticed the oblique sans-serif f in Ray's Figure 66.2 ... JeffConrad (talk) 06:24, 13 January 2009 (UTC)


 * Didn't notice. GBS won't show me that page, and the book is back at the office.  And you know I'm blind to such things. Dicklyon (talk) 06:56, 13 January 2009 (UTC)


 * This is your lucky day ... the same figure is on p. 319 of The Manual of Photography (Ray 2000), replete with oblique sans-serif f. The caption even appears to be correct, and includes an explanation of the rectangle in the center of the figure. Yeah, I know, the MoP is probably also at the office. JeffConrad (talk) 09:30, 13 January 2009 (UTC)


 * Page 62 in the same book says "italic f", which I would interpret in terms of Italic type. The figure you pointed out says "f-number", does not show an "f/#" construct.  Opinions are pretty split as to whether to set the f in f-number the same as the f in f/#; I don't like to italicize it in f-number, but Ray does, apparently, though in the figure the oblique f was not a true italic (see Italic_type).  Dicklyon (talk) 22:52, 13 January 2009 (UTC)

Confusion about use of tabulated values with speeds other than ISO 100
Dick, I'd be surprised if the edit you reverted on 13 October 2009 was vandalism—vandals usually aren't that carefully consistent. But it certainly was boneheaded and boorish for an IP to make such an edit without at least some discussion, so perhaps there's little practical difference. In any event, the way in which the ISO speed conversion works is obviously confusing people (hence my comment in the Exposure compensation article). Offhand, I don't have any great ideas for fixing this. A glance at the ANSI standards offers no help whatsoever; they're probably far more difficult to apply than the simple table based on EV100. I've made a couple of minor tweaks, but I don't think they will really be of much help. Perhaps we could replace greater with higher or “faster”, though the physical implication, especially with the latter (shades of “shutter speed” ...) is somewhat of a misnomer. And I'm not even sure such a change would help. JeffConrad (talk) 01:54, 14 October 2009 (UTC)


 * Sometimes with an uncommented anonymous edit inverts the sense of something, like changing true to false or plus to minus, it's simple vandalism; I was giving him the benefit of the doubt, rather than assert that he was just confused and wrong. Hopefully the small clarification I added will help; thanks for fixing my spelling typo. Dicklyon (talk) 03:50, 14 October 2009 (UTC)


 * I guess we can see how long it is before this section is “corrected” again. I have no additional ideas. With the revised wording and the examples, I can't think of what else to say. Perhaps we should consider the expanded explanation SOP whenever discussing EV conversions.


 * There may be some folks who don't know how to calculate base-2 logarithms. But I suppose this article isn't a basic math tutorial, and the folks who really don't get this section probably won't be dealing with logarithms anyway. In any event, I'd rather stick with exact expressions using base 2 than use base-10 logs with magic conversion factors. JeffConrad (talk) 04:10, 14 October 2009 (UTC)

The EV article is in fact extremely confusing because it is simply incorrect about basics. Specifically, the chart itself is OK, but the heading of the EV chart (Table 1) is very misleading because the heading ends with "(ISO 100)" as if that applied somehow. But it does not. That heading is ignorant of EV basics. It was a hugely incorrect assumption by whoever added it. The EV chart is of course valid as is for each and any ISO (specifically, whichever ISO is currently being used). An ISO change of course changes to another table row (to another EV) since it obviously changes the exposure settings, but the chart itself is valid for any ISO, as written. That certainly seems a major fundamental of EV. If you double ISO, your EV value increases by 1, to new settings one stop greater, and one row in the table, but in the same one chart. The one chart is correct for any ISO. If you use any ISO, a light meter will tell you the exposure EV (or settings), which you then find in the one EV chart. There is only one EV chart (including EV 0 = f/0 at one second) for any and all ISO, but a different ISO will choose some different exposure and EV value. That "(ISO 100)" note needs to instead read "(for any ISO being used)". That would clear up most confusion about EV. ISO 100 is just a number, in no way special, other than it is popular. — Preceding unsigned comment added by WF222 (talk • contribs) 18:34, 9 September 2022 (UTC)

Sign errors
This uncommented diff introduced a bunch of sign errors. I notice when a partial fix was done today. I'm pretty sure it was correct before, so I'll fix it. Dicklyon (talk) 22:38, 24 May 2013 (UTC)


 * as one who regularly photographs on film using an external light meter in EV mode, I tell you that the current ISO-conversions and examples in the article have the wrong sign. Consider, for example, the following equation:


 * $$\mathrm{EV}_{50} = \mathrm{EV}_{100} + 1 \,.$$


 * In order to use a film with an ISO sensitivity of 50 instead of 100, one has either to open the aperture one step or to double the time of exposure. Either adjustment corresponds to decreasing EV by one, not increasing it. The correct relations are:


 * $$\mathrm{EV}_{50} = \mathrm{EV}_{100} - 1 \,.$$
 * $$\mathrm{EV}_{400} = \mathrm{EV}_{100} + 2 \,.$$


 * Please, amend the error in the article. Ant 222 (talk) 22:32, 26 September 2019 (UTC)


 * Yes. The error came back on 05 August 2019.  I just had to revert the most recent edit. Dicklyon (talk) 01:15, 27 September 2019 (UTC)

Table 3 Confusion
I have been trying to understand this page completely for some time now. I have a strong math/physics background and the math is no problem. The concepts are very confusing, so I could be off base here. If I am, I will consider any corrections a teachable moment. The bright sunlight provides a measurement (I have performed myself) of approximately 10,000 footcandles. I have also read that in several places that is the correct value. EV15@ISO100 would be the correct settings on the camera for a good exposure of an average scene under these conditions. As you know -- it is the sunny 16 rule. However, Table 3 has the illuminance of EV15@100 as 7611fc. This is 1/3 of a stop away from the value I expect to see here. Comments? Kq6up (talk) 17:21, 11 June 2013 (UTC)


 * One-third of a stop is generally considered a negligible difference in photography rules of thumb. Dicklyon (talk) 17:25, 11 June 2013 (UTC)


 * Granted, however the table should represent scientific data derived from calculations or empirical measurements. 33% error would indicate that either an instrument is crude, miscalibrated, or broken.  If the data was derived, then there is a calculation error some where.  This could lead to someone trying to get a deep gut understanding of how all the pieces fit together to bang there head needlessly.  Back to the salt & pepper notebook and head scratching Kq6up (talk) 18:21, 11 June 2013 (UTC)


 * 7611 fc is only 24% lower than 10000; if the rule of thumb gives the right exposure for 7611, it will be slightly over on your typical sunlit scene, which is exactly the direction that you want your rule of thumb to be a little off. Maybe it's that simple?  Dicklyon (talk) 19:15, 11 June 2013 (UTC)


 * The concern I raised was not addressed by the last comment. Moreover, the table is also ambiguous as it stands.  The illuminance is very straight forward since there is no consideration of placement of tones photographically.  It is simply luminous flux per unit area.  However, there is no clue as to what shade of grey the luminance values would be rendered as photographically (what zone are we talking here? -- zone V?).  The ratio of illuminance to luminance is 20:1 for the metric side of table.  I can understand a factor of 4*pi for the units to line up, and then there is the whole lambertian defuse reflector cosine factor.  When this is integrated over a hemisphere throws in another factor of pi some how.  All very confusing even for a huge science nerd as myself.  This is hard enough to comprehend without the ambiguities in the table.  A paragraph explaining the luminance values especially.  Sorry for the rant.  I just hope my critique will make for a clearer article.  Thanks Kq6up (talk) 20:29, 11 June 2013 (UTC)


 * Well, it's all rules of thumb. As the article says of luminance metering, "Values for the reflected-light calibration constant K vary slightly among manufacturers."  The problem is that scene luminance, or illuminance, is not enough to determine a "best" exposure.  Both reflected and incident light meters, and the sunny 16 rule, give you a guideline, but you can adjust it based on what tonality you're trying to achieve.  The history of standards around metering is complicate; ask Jeff Conrad, iirc he's the expert on this.  Dicklyon (talk) 00:04, 12 June 2013 (UTC)


 * I understand it is all a big ol' mess. I am just trying to get my head around the photometry involved.  Once that makes good sense, I can sort it all out later.  That being said if one sticks to the modern formulas and definitions, it should all be internally consistent.  I doubt the standards themselves are er.  Just that most people don't really understand them.  Just look at some of the replies on Yahoo Answers.  They are horrendous.  One final question.  How do you get notified that there has been a change made to this page?  You responded very quickly.  Also how would I get in touch with Jeff Conrad? Kq6up (talk) 01:39, 12 June 2013 (UTC)
 * Unfortunately, Jeff seems to have left Wikipedia quite some while ago: User:JeffConrad. You could try to email him, perhaps he'll answer. --Matthiaspaul (talk) 01:54, 12 June 2013 (UTC)
 * Indeed, I miss him. Here are some of his relevant edits to Light meter. Dicklyon (talk) 03:37, 12 June 2013 (UTC)
 * I don't get notified except by checking my watchlist, which has a few thousand articles on it. Dicklyon (talk) 03:37, 12 June 2013 (UTC)

Sign confusion
I had an email from an editor concerned about the plus sign in the formula for adapting the table numbers in Exposure value to different ISO values, pointing out that dpreview and others go the other direction as at. I have just done an edit to resolve that confusion, showing that the equation with the minus sign goes the other direction. I added an example showing how it relates to what dpreview says. Both were correct, but now the relationship should be more clear. Dicklyon (talk) 23:19, 19 January 2014 (UTC)

§§ Thank you Dick for looking at this; however, the algebraic sign shown is still wrong - please try and prove this to yourself using the following formulae:



Once you've done the mathematical proof, I hope you'll correct the article as I'll leave that you :)

Thank you!

Kelly

Kbellis (talk) 01:26, 21 January 2014 (UTC)


 * Kelly, your formula with the minus sign (which is not justified anywhere that I can see) shows that if you switch to a higher ISO then you should use a lower EV, which corresponds to opening up or using a longer exposure. That's backwards.  That's not what the dpreview calculator is showing, as I explained.  It is correct, but it is not calculating what we mean by the EV setting to use at a different ISO.  It is calculating the EV100 meter reading (an LV as their footnote explains) that you can use with a particular setting of shutter and aperture.  The article has example of both directions; do you think either example is giving a wrong result?  Dicklyon (talk) 04:58, 21 January 2014 (UTC)


 * Also, the formula you use for EV100 is actually the formula for EV, which depends only and N and t, while EV100 depends only on light level (scene luminance). The formula for a recommended EV based on metering, that takes speed S and light level L and metering constant K into account is in a later section:
 * $$\mathrm {EV} = \log_2 {\frac {L \cdot S} {K} } \,.$$
 * which clearly implies a plus sign for the effect of S. You got things the wrong way around.  Dicklyon (talk) 05:43, 21 January 2014 (UTC)

Relationship of EV to lighting conditions: units wrong?
In the section "Relationship of EV to lighting conditions" EV15 is stated to be correspond to a luminance of approx 4,100 *lux*.

Yet in "Table 3. Exposure value vs. luminance (ISO 100, K = 12.5) and illuminance (ISO 100, C = 250)" EV15 is given as 4096 *cd/m^2*

From the Wikipedia article on luminance, cd/m^2 seems to be the correct unit for its measurement.

Are the units given under "Relationship of EV to lighting conditions" correct? If so an explanation might be added as to why.

IanWorthington (talk) 15:57, 6 August 2016 (UTC)


 * I have changed the units for luminance to the correct ones (cd/ft2 and cd/m2 ). I changed the illuminance relationship to use the common value (C = 250) for a flat (cosine-responding) sensor, which is the only way illuminance can be meaningfully measured (and which the manuals for most incident-light meters indicate); I added a sentence to that effect. Though I did not mention it, the relationship of reflected-light measurements to luminance is meaningful only when measuring a single, uniform luminance. Accordingly, I wonder if giving these values here is actually useful, especially since the topic is already covered in EV as a measure of luminance and illuminance. I also think reference to the camera settings adds nothing but confusion. I would recommend dropping the material from this section. JeffConrad (talk) 05:14, 23 November 2016 (UTC)

Indication of photometric quantities with subscript v
All too often, edit comments don’t leave enough space for a proper explanation. I removed the subscripted v from the equation for luminous exposure for two reasons:


 * The sources we cite do not use the subscripts; for whatever reason, the same is true for most photographic texts. I doubt it’s a matter of currency; the subscript v has been used for decades to indicate photometric quantities. The difference may not be fatal, but it may still confuse the reader who bothers to check the source.


 * Because we also refer to the original APEX symbol Ev, the nearly identical symbol Ev is asking for confusion; the implication that the quantities are logarithmic arises only in this context. The note indicating that this is not the exposure value may confuse as much as it clarifies; perhaps a more expansive explanation would address this, but I think the greater problem is needing an explanation at all.

The astute reader can probably sort this all out, but the reader should not need to do so.

To me, there would seem little chance of confusing radiometric and photometric units in the context of this or most photographic articles; perhaps that’s why most photographic texts omit the subscript. NIST appear to accept either using or omitting the subscript, so I’m not sure omitting it is bad practice.

It’s obviously desirable to have consistency among articles, especially those in the same category (e.g., photography). Matters are complicated somewhat by the seeming inconsistency of categorizations; because this article is also included under Photometry, omitting the subscripts here leads to inconsistency in that category. But I would think this article should be consistent with others such as Light meter; it’s probably feasible (with suitable explanations) to include the subscripts in that article, but I cannot imagine doing so in APEX system without driving the reader nuts.

In a sense, we are damned if we do and damned if we don’t. But all things considered, I think the best choice is to follow most photographic texts and omit the subscripts; this avoids two different meanings for the same symbol, and avoids conflict with the sources cited.

If others insist that the subscripts are the better approach, I’m not going to press the issue too far. But I think something more expansive than “(not EV, the exposure value)” would be needed so that the reader is not left scratching her head. And related articles should be revised to take the same approach, though APEX system might need to remain an outlier.

JeffConrad (talk) 10:35, 27 November 2016 (UTC)

Deckel Light Value System
I offer the following initial thoughts on how to (somewhat) cover the Deckel Light Value System discussed under Talk:APEX system:


 * The EV concept was developed by the German shutter manufacturer in the 1950s (Gebele 1958; Ray 2000, 318). Its intent was to simplify choosing among equivalent camera exposure settings by replacing combinations of shutter speed and f-number (e.g., 1/125 s at f/16) with a single number (e.g., 15).


 * On some lenses with leaf shutters, the process was further simplified by allowing the shutter and aperture controls to be linked such that when one was changed, the other was automatically adjusted to maintain the same exposure. This was especially helpful to beginners with limited understanding of the effects shutter speed and aperture and the relationship between them.  But it was also useful for experienced photographers who might choose a shutter speed to stop motion or an f-number for depth of field, because it allowed for faster adjustment—without the need for mental calculations—and reduced the chance of error when making the adjustment.


 * The concept became known as the Light Value System (LVS) in Europe; it was generally known as the Exposure Value System (EVS) when the features became available on cameras in the United States (Arizona Republic 1957, 46).


 * Because of mechanical considerations, the coupling of shutter and aperture was limited to lenses with leaf shutters; however, various automatic exposure modes now work to somewhat the same effect in cameras with focal-plane shutters.


 * The proper EV was determined by the scene luminance and film speed; it was intended that the system also include adjustment for filters, exposure compensation, and other variables. With all of these elements included, the camera would be set by transferring the single number thus determined.


 * Exposure value has been indicated in various ways. The ASA and ANSI standards used the quantity symbol Ev, with the subscript v indicating the logarithmic value; this symbol continues to be used in ISO standards, but the acronym EV is more common elsewhere. The Exif standard uses Ev.


 * Camera and Imaging Products Association. 2016. Exchangeable image file format for digital still cameras: Exif Version 2.31 (PDF).
 * CIPA. See Camera and Imaging Products Association.

I would put this in the lead for now because I can’t think of a better place for it (I can’t really justify a section Light Value System on the basis of one sentence). The current first and last paragraphs there would remain as they are.

I rely mainly on Deckel′s 1958 patent for the new information. Ray (2000, 318) would probably support the comment about automatic exposure modes if a citation is thought necessary.

I haven′t included the full reference for the Arizona Republic article because I’m trying to figure out how to get a view of that page without starting a free trial or coughing up $20.

JeffConrad (talk) 04:20, 4 December 2016 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 1 one external link on Exposure value. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20131112083840/http://www.dpreview.com/glossary/exposure/exposure to http://www.dpreview.com/glossary/exposure/exposure

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 08:11, 28 December 2016 (UTC)

Tabulated exposure values: reverse conversion and dpreview calculator
I’ve slightly cleaned up the last part of this section, but it still seems awkward. And I wonder if this material—especially the archived version of the calculator at dpreview.com—even belongs. The explanation on that page states


 * The exposure generated by an aperture, shutterspeed, and sensitivity combination can be represented by its exposure value "EV".

This is simply wrong. EV can be applied to either side of the meter calibration equation, but not both. EV can represent either a combination of f-number and exposure time or a combination of luminance and sensitivity (and meter calibration), but not both. In a footnote “for the purists”, the author (Vincent Bockaert) further states


 * (1) Strictly speaking, the term "exposure value" is used to represent shutterspeed and aperture combinations. An exposure value which takes into account the ISO sensitivity is called "Light Value" or LV and represents the luminance of the scene[2]. For the sake of simplicity, as is the case in this article, Light Value is often referred to as "exposure value", grouping aperture, shutterspeed and sensitivity in one familiar variable.

The first sentence is correct; the rest isn’t. There is no such thing as a “Light Value” that has a universally recognized meaning, as the article Light value makes clear. I agree with Bockaert that, for digital cameras, sensitivity is just another camera setting. And there is no reason that f-number, exposure time, and imaging-system sensitivity could not be encoded in a single variable—but such a variable would not be EV unless EV were redefined. As confusing as EV seems to be for many people, a new, unofficial, meaning seems like the last thing we need.

If no one objects, I’m for getting rid of the last two paragraphs altogether.

JeffConrad (talk) 02:50, 2 January 2017 (UTC)

Relative aperture vs f-number
I am concernded about the relative aperture versus f-number confusion. In the article's Formal definition section, the symbol N is said to be "the relative aperture (f-number)". That's confusing because it cannot be both.

When a f = 50 mm lens, for instance, has its aperture set to f/4 then
 * the absolute aperture, or entry pupil diameter D, is 50 mm / 4 = 12.5 mm, and
 * the relative aperture is D/f = 1:4 (i. e. a quarter), and
 * the f-number is f/D = 4, i. e. the reciprocal of the relative aperture.

So if you're wondering what's to enter for N in the formula EV = log2(N2/t) when the aperture is f/4 — 4 or 0.25? ... after reading that Formal definition you won't be any wiser. In fact, N should be 4 (= the f-number) and not 1:4 (the relative aperture). I'd appreciate a clarification of this little but crucial point.

-- Omzu Iko (talk) 19:20, 20 February 2020 (UTC)

I just realized this topic already had been addressed by User:93.86.250.243 half a year ago, on Oct 9th, 2019. Unfortunately he had his terminology the wrong way around so his edit rightly got reverted. However the article's current version — which suggests relative aperture and f-number were the same thing — is wrong, too. -- Omzu Iko (talk) 07:15, 21 February 2020 (UTC)

Replace images at top
The current images at the top of the page (correctly exposed photos of moving water at slower and higher shutter speeds) isn't what is needed to explain the concept of EV. How about a series of normally exposed, underexposed and overexposed images? Groogle (talk) 04:00, 12 April 2024 (UTC)