This article is within the scope of WikiProject Photography, a collaborative effort to improve the coverage of photography on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.PhotographyWikipedia:WikiProject PhotographyTemplate:WikiProject PhotographyPhotography articles
"Many photographic lenses produce a superior image in the center of the frame than around the edges. When using a lens designed to expose a 35mm film frame with a smaller-format sensor, only the central "sweet spot" of the image is used; a lens that is unacceptably soft or dark around the edges when used in 35mm format may produce acceptable results on a smaller sensor."
Resolution must also be taken into consideration. A crop body camera's sensor will usually have more pixel-per-inch, making lenses appear sharper on a full frame camera. While corner sharpness does suffer from this "sweet spot" effect, the center of a full frame image will be sharper. Run some lens comparisons. Nonameplayer (talk) 07:41, 2 October 2009 (UTC)[reply]
If you run the comparison of the Canon 50mm f/1.4 on a 450D vs 1DsIII @ f/5.6 (link) you can see the full frame camera is actually sharper in the corners. Nonameplayer (talk) 07:46, 2 October 2009 (UTC)[reply]
It is true for many lenses as stated, but does not generalize to ALL lenses. Dicklyon (talk) 15:57, 2 October 2009 (UTC)[reply]
"a lens that is unacceptably soft or dark around the edges when used in 35mm format may produce acceptable results on a smaller sensor."
This simply is not true (Lens Comparison) because of the pixel-per-inch issue I mentioned above. Nonameplayer (talk) 19:48, 21 October 2009 (UTC)[reply]
It says "may"; the fact that it goes the other direction for that lens (and many others) does not make the statement less true. Dicklyon (talk) 00:50, 22 October 2009 (UTC)[reply]
Maybe not "less true" but it is misleading. The line should read: "a lens that is unacceptably soft or dark around the edges, when used near or at its maximum aperture, can give better results on a smaller sensor." Reference, Nonameplayer (talk) 20:16, 23 October 2009 (UTC)[reply]
This statement is true, lenses vignet and becomes softer near the edges, especially noticed on wide and ultra wide. Full frame is only sharper because of large pixels, but I bet some here will try to argue that its a myth even seeing results. 173.206.212.205 (talk) 00:18, 10 July 2010 (UTC)[reply]
I removed the following sentences from the article:
1 New lenses designed for specific use on DSLRs are subject to the same crop factor, depsite projecting a smaller image.
2 The standard focal length measurements are dependent on the camera sensor size, not the lenses.
3 This causes confusion because a 18-70mm digital or regular lens will never be 18-70 mm by conventional film camera standards.
4 The entire range is multiplied by the crop factor when a digital camera is used.
5 As cameras transition from film to digital so does the conventional understanding of the focal length numbers when used on the two different formats.
I understand what the original editor means, but it's phrased incorrectly and could cause confusion for users unfamiliar with the concept.
Sentence 1 is false: Lenses which cast an image circle designed for the sensor size are not subject to a crop factor, because the image isn't being cropped. I know what was meant -- that for folks used to equating certain focal lengths with certain FOVs, the smaller sensor makes such correlations inaccurate. However this has always been true, even before digital sensors came into being. A photographer who was used to large-format cameras having 120mm wide-angle lenses couldn't assume a 120mm lens on a Nikon SLR was wide-angle. Nor could a medium-format user who thought a 100mm lens was a normal lens. It all depends on your frame of reference.
Sentence 2 is also technically false. Focal length is dependent on nothing other than the actual focal length of the lens, measured in millimeters. As before, the editor is equating focal length with FOV. They are not the same. FOV varies according to sensor size, focal length of the lens and image circle size. Focal length does not vary due to sensor size.
Sentence 3 is also false. An 18mm lens is an 18mm lens no matter what camera it is attached to. A 70mm lens is a 70mm lens no matter what camera it is attached to. What varies is the FOV captured by the sensor.
Sentence 4 is correct w/r/t FOV vs. focal length.
Sentence 5 would be correct if reworded to address how FOV varies with sensor size.
Moondigger, I agree with most of your changes, but not really number 1. You also inserted the last clause of this sentence, which I have removed: "Crop factor is the ratio of linear photographic area lost in a DSLR camera compared to a standard 35 mm film frame, when used with lenses designed for that frame."
I know what you're getting at here, but if you look at how crop factor is used, it's as a characterization of a camera format, independent of any lens, irrespective of any imaging circle size, etc. It doesn't matter that the designed-for-digital lenses have their image circles cropped less; the camera still has the crop factor that it has. The term has caused great confusion by focusing on cropping. The FLM term isn't much better in the confusion that it causes, but I put that in, too, since it's also widely used. Dicklyon 20:00, 8 September 2006 (UTC)[reply]
Dicklyon, the article was written the way it was to address a term that has indeed arisen due to the existence of smaller-than-35mm frame DSLRs that take lenses designed for the full 24x36mm film frame. I realize that the term "crop factor" can be generically applied to any sensor smaller than the reference frame, but that generalization did not exist (and the term did not exist!) until the advent of DSLRs that took lenses designed for a larger format. Photographers with boatloads of lenses designed for the 35mm format were adopting new DSLR camera bodies that worked with their existing lenses -- yet the images produced were (are) radically different than the images they'd get with the same lenses on their old film bodies. The term arose specifically because of that situation. And, I should point out that photographers without a good understanding of the term are most likely to look it up here specifically because of its use in discussions/advertising/etc of lenses designed for use with the 35mm frame on their digital cameras.
So I disagree with the way you've generalized the article and removed references to the very origin of the term. I am not opposed to the incorporation of the information you have incorporated, but I am opposed to the wholesale removal of all references to the format the lens was designed for. It is not immaterial to the discussion, as you claim in your edit notes on the history page. In fact it is the very genesis of the term as applied. The fact that it can be generically applied to any sensor smaller than the reference frame can and should be in the article, but not as the primary definition.
That said, I do not wish to get into a revert war, and don't have time to re-write the article tonight to incorporate both sets of information. If you would like to edit it to include the information you excised, please do. Otherwise I'll re-incorporate it at a later time. -- Moondigger 01:56, 9 September 2006 (UTC)[reply]
I'm not sure what you want back in that I took out, so feel free to state it more explicitly here. In the mean time, I'll see if I can improve toward the needs of those DSLR shooters. I agree that's where it's mainly used (and misused), but I think the more general definitions will help people understand it better. Dicklyon 02:20, 9 September 2006 (UTC)[reply]
OK, I added various bits of info and clarification. Feel free to have the next crack at it. Dicklyon 03:25, 9 September 2006 (UTC)[reply]
I apologize if my comments came across the wrong way. I wasn't trying to exert any proprietary claim on the article, and I was in a bit of a crabby mood when I first responded. I just skimmed through the article and it looks good -- certainly better than it was prior to your edits. I'll give it a thorough once-over for copyediting purposes, but it looks really good. Thanks... -- Moondigger 03:14, 10 September 2006 (UTC)[reply]
Thanks. And if anyone knows about feeling crabby over wiki edits, it's me. Dicklyon 06:29, 10 September 2006 (UTC)[reply]
18 years later and I can see both of you, @Moondigger and @Dicklyon, are still active photographers (Moondigger hasn't contributed to WP in the last few years, but I can find them elsewhere on the web) - "mirrorless" came into use after this conversation, in 2008, to describe the then-new Micro-Four-Thirds cameras. Until then, as far as I know and regardless of image circle, designed-for-digital meant additional element coatings made necessary due to the greater reflectance of a digital sensor than of film resulting in more flares and other light artefacts in photos. But then, they said Live View had become good enough for us, and that digital sensors had advanced to the point that lens telecentricity was no longer required, so they could delete the mirror box and move the flange closer to the sensor - and that this would permit smaller lens designs and previously impossible lens parameters, and last year, Tamron released a 17-50mm ƒ4 lens which projects a proper 35mm-full-frame-covering image circle. It happened!!
(The problem of equivalent focal length, equivalent aperture, sensor size, etc. being communicated in terms which are unintuitive and confusing to most people has not yet been solved, and the frustration endures. Is is helpful to say a phone with a sensor smaller than a baby's fingernail has a 24 mm, ƒ/1.78 aperture lens, when most of that lens is glass which does not exist? Is it??! Lenses should be quantified by their angle of view, the actual diameter of their maximum aperture, and the diameter of the image circle - and on a device with a camera in it where the intention is to use the camera to take photos for the photo's sake, the actual diagonal sensor size should be present, in metric, IN DECIMAL DECIMALS - a 1/3.09" image sensor? A fraction of an inch with a decimal denominator?! It has to be intentional. It's so people think they've a phone with a one-point-three-and-a-bit inch sensor in it. And you couldn't have manufacturers giving a figure for the depth-of-field, because they'll just say it's unbelievably shallow and put "value calculated using a proprietary circle-of-confusion dimension" at the bottom of a web page with a URL which is only published on the second-to-last page of the user manual, which is only available through their support site, as a PDF, because they're saving paper by only putting a quick-start guide in the box. And then there's all these cheap "ƒ0.95" lenses where if you focus more than 2 metres in front of you, the light ends up being occluded by some bit of the lens other than the aperture blades, and you only get the light and depth-of-field you'd get from an ƒ1.4. And then there's "APO"......And then......And then......And then.....)
Aaanyway, I thought the advent of the theoretical problems with those statements being made flesh was worth acknowledgement.
I reverted the last 2 of many edits by Acerimusdux, due to an objective error. My edit summary got truncated; tried to say "shooting from the same position, with same f-number, with the same actual focal length will yield the same depth of field" is objectively incorrect as it ignores COC dependence on format size. See the depth of field and circle of confusion articles. Some of the earlier edits I'm not so keen on, either, as they seem to be suppressing the idea of "full-frame" in favor of the more ambiguous 135 film size, which has more than one format. Dicklyon 07:16, 10 November 2007 (UTC)[reply]
Actually, "full Frame" is the ambiguous term as it originally meant --and to some, it still means-- a 10x8 inch (HxV) studio camera with a crop factor of 1x. 135 has only one size; 36x24 mm. There is no other size for which 135 format means. It refers to the 35mm film in the daylight loadable cassette invented by Kodak in 1939. It is very specific down to the millimeter unlike APS-C (or 120/220 film which has several form factors, which is where the term "crop-factor" originated). Applied Logics 03:18, 31 March 2015 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
Are you confusing "full frame" with "full plate"? Dicklyon (talk) 05:29, 1 April 2015 (UTC)[reply]
I'm not, by any stretch of the imagination, a camera expert. So, that being said, I'm a little confused by the article where it states the crop factor should be applied to the 1/focal-length rule of thumb for camera shake. Wouldn't the image circle produced by the lens be the same on both crop and full frame cameras? And if so, the crop sensor just uses less area of the image circle, cropping the field of view, not really increasing focal length or camera shake.
Unfortunately I don't have a full frame DSLR I can test against my Canon 20D cropped camera for camera shake.
Eelyaj (talk) 14:03, 31 May 2008 (UTC)[reply]
Correct, the crop factor does not affect the size of the blur on the film, and it does not change the focal length. Nevertheless, the rule of thumb so modified makes sense, as you'll realize if you recall that the smaller image on your sensor needs to be enlarged more to make a print, so needs to have proportionately less blur to look equally sharp. Dicklyon (talk) 22:51, 31 May 2008 (UTC)[reply]
Smaller image and smaller subject are two different things. If the two formats capture the same FoV from the same distance then one would have had a longer lens and the old rule-of-thumb applies. If they are both capturing the same scene from the same distance with the same lens, then the subject of the photo would not need to be enlarged by a greater factor so the old rule-of-thumb still applies. If they are capturing the same FoV by one (the larger format) moving in closer, then they would need to hold their camera more still for the same sharp image (but the rule-of-thumb does not take into consideration distance since close-ups are usually done on a tripod anyway). One may argue that the fact that I am producing two 8x10 images means it does not matter what the subject is but one chooses ones lens based on the subject just like I still will not hand-hold a macro shot at f22 and shutter speed 1/180 with a 135 mm macro lens nor a 35 mm macro lens even if I were using a ring flash.
Indeed, I hand-held some shots at f6, shutter 1/180 and ISO 400 with a 200mm lens on a APS-C (CF 1.53) and got perfectly sharp images (but I also had internal image stabilization so all bets are off). Applied Logics 03:38, 31 March 2015 (UTC)
What about half-frame cameras that use size 135 film, like my Olympus Pen F? Crop factor near 1.4, depending on how you account for 3:2 vs 4:3 shape. Anyway, it's not clear what you're arguing above, or whether you're agreeing with me or disagreeing that the rule of thumb applies to the 35mm-equivalent focal length rather than the absolute focal length. Dicklyon (talk) 05:31, 31 March 2015 (UTC)[reply]
Then call it the "regular" 135 (as that article did) or the redundant "full frame 135" as 'full Frame' without a reference is ambiguous. The Half-frame article makes it clear that the term was in reference to the "intended format" so there would be Studio (8x10in) and half-frame Studio (4x5in), 135 (24x36cm) and half-frame 135 (18x24). 120 (6x6cm) and half-frame 120 (6x3cm).
Above, I am saying that the rule-of-thumb --guideline really-- does not change for the smaller format. It is still the inverse of the focal length. Would one say that the crop factor of the 6x7 is about 0.5x to the standard 135 so one can hand-hold a 200mm lens at 1/100th of a second with a 6x7 camera? It would still be 1/200th of a second (or possibly faster since the camera is heavier). The inverse focal length is just a rule-of-thumb for subjects a fair distance from the camera but with the right technique, one can hand-hold at even slower shutter speeds even without IS.
And it turns out that I lied. I actually took some very sharp hand-held macro shots with a 35mm lens at 1/80th of a second and a 50mm lens at 1/100th of second in February 2015. But again, good technique (lens in hand, back of hand on table) and IS. Normally not something I would have attempted but limited space meant no tripod. Applied Logics 08:40, 31 March 2015 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
I've taken sharp shots at 1/8 s handheld, too, back when I was steady and careful. But that's not the point. The point is that a typical handheld shot will be sharp at about 1/50 s with a "normal" lens, whether shooting your 116 roll film or half-frame (aka single-frame since the original format was from the movie film) 35 mm. You need to scale the rule of thumb to the format. Dicklyon (talk) 05:27, 1 April 2015 (UTC)[reply]
"However, the crop factor can be an advantage to photographers when a narrow FOV is desired. It allows photographers with long-focal-length lenses to fill the frame more easily when the subject is far away. A 300 mm lens on a camera with a 1.6 crop factor delivers images with the same FOV that a 35 mm film format camera would require a 480 mm lens to capture."
This is uncited and I don't believe it's a true advantage - I feel it's a misunderstanding of what's happening. You can achieve the same effect by simply cropping the FX shot. The advantage lies in the smaller form factor of the lens and camera, and the fact that the viewfinder will project the closer zoom.203.196.81.85 (talk) 15:48, 6 January 2009 (UTC)[reply]
You could cite this source for that advantage (even those example numbers; probably came from there). You can't get the same advantage from a full-frame camera that has the same number of pixels, since cropping it would leave you at a relative disadvantage in number of pixels. Dicklyon (talk) 19:12, 6 January 2009 (UTC)[reply]
Hm, but this is somewhat comparing apples with oranges. You're right, that you won't get the same effect if the full frame camera has the same number of pixels as the cropped one, however, you do get the same effect, if the pixel density does not change, which, I think, is more in line with reality. After all, we are talking about cropping, not about "compressing". Cropping = cutting away something around the edges without changing the underlaying properties of the photographically sensitive medium at the same time. A film emulsions' resolution will not change, if you crop the negative. Compare, for example, the pixel density of a Sony DSLR-A700 (12.2 MP @ APS-C) with a DSLR-A900 (24.6 MP @ FX) - they are not identical, but close enough to assume them to be the same in first approximation. The DSLR-A900 can also be forced to work in APS-C mode. That's what cropping is about. In my book, the "long lens advantage" of APS-C vs. FX is non-existant, unless you "silently" change the sensor in the middle of the comparison. However, if you do, the advantage of APS-C would not be its smaller sensor, but its - in your argumentation - higher pixel density. I might agree that a higher density could be an advantage (if there would not be problems with noise), but this might be an argument broad forward in an article about pixel density (for example, when comparing a Nikon D700 (12.1 MP @ FX) vs. Nikon D3X (24.6 MP @ FX)), not in an article about crop factor (where the pixel density is assumed to be constant). Or, if both parameters, format and pixel density, are changed at the same time in a comparison, it should be expressively noted that such an advantage, as some photographers seem to see in it, is actually the result of this /combination/ of changes, not the change of the format alone. Otherwise, it's misleading and contributing advantages to the crop factor, which are unrelated.
BTW. I just read the page of the book you were referring to. "A crop factor of 1.6x effectively turns a 300mm f/4 lens into a 480mm (300mm x 1.6) lens that still has f/4 lens speed." OMG, plain wrong statements like this are the very cause, why there is so much confusion about crop factors and focal length multipliers. However, even if repeated uncoutable times, this does not make it any true. I'm afraid, this source can be ruled out as unscientific.
Possibly an interesting read to help abandon a few myths:
One thing that hasn't been mentioned in this article is the influence of aspect ratio. If you try to compare sensors with different aspect ratio, you have a total of four different ways to perform the comparison:
You can compare the (square root of the) area of the sensor.
You can compare the diagonals.
You can compare the height.
You can compare the width.
I can see a use for each of these comparisons, but it's often unclear which is meant. In the typical case of comparing a 4:3 aspect ratio to the 35mm 3:2 aspect ratio, using the Four thirds format as an example, you get:
Square root of area: sqrt (864 / 225) = 1.96
Diagonal: 21.64 / 43.26 = 2.0
Height: 24 / 13 = 1.85
Width: 36 / 17.3 = 2.08
I'm not sure how important these differences are, but I think it should be mentioned. Comments? Groogle (talk) 01:46, 20 August 2009 (UTC)[reply]
When such differences arise in photography, the typical convention is to use the diagonal measure. I wouldn't even comment on in, though, without a source to that effect. Or if you have a source that comments on the other alternatives, you could mention them, too. Dicklyon (talk) 05:41, 20 August 2009 (UTC)[reply]
I have only seen approximate crop factors in cameras brochures.
In practice, one use crop factor to compare the new digital cameras with prior experience with 35mm cameras, when one wants to buy a lens or use an inherited one. I use the formula: because one can crop the larger dimension to get the same aspect ratio of 35mm film.
I do not have a book reference for that formula, but mathematics is a language and that formula expresses a crop factor which can be re-cropped to preserve a 35mm aspect ratio.
I did not changed the article, because I do not want to provoke the reaction of other wikipedians who want a more strict citation of everything. I do not know if the above formula is used by some famous photography author. I have not searched it. As I stated above, is just a product of the modest experience of an amateur photographer. — Preceding unsigned comment added by 189.233.106.197 (talk) 10:21, 30 October 2016 (UTC)[reply]
Sorry, but olypedia.de is not spam, promotion or any other (we say in german) "Quatsch". It is THE german free wiki about olympus - nothing to do with olympus co! You find here many informations about olympus, that you can not find in en.wikipedia.org or other wikis. The most of the autors came from germany, for many of them (also my person) is english not a favorite language.
My edit was reverted, but I would still like to state that a crop factor (or more universal a format factor) is technically not the same as a focal length multiplier. The term focal length multiplier (Brennweitenverlängerungsfaktor in German) is the technically correct term used in conjunction with teleconverters, not in conjunction with cropping. The article mixes up both terms and incorrectly uses them as synonyms.
Actually, *many* people incorrectly use the term focal length multiplier when they actually mean the crop factor, probably because they only pay attention to the field of view, not to the other properties, which won't change with cropping, but will by using a teleconverter. This wrong usage of the term focal length multiplier, however, will make it very difficult for them to actually understand why a seemingly same (focal length) multiplier will have different optical effects on the resulting pictures when using a teleconverter or when using a camera with a crop sensor. You can find all sorts of strange after-thought "explanations" in the web (and in some media) resulting from this, and some even go further and start to interpret advantages into cropping vs. using a teleconverter, which are in contradication with the laws of physics. Some photographers will even believe in some kind of unexplainable "magic" occuring when using a (cropped) digital camera vs. using a 35mm full-frame camera.
Therefore I think the article should fade out the misleading usage of the term focal length multiplier except for in some extra paragraph explaining why this is not the same as a crop factor despite its common use in this meaning by newbies.
BTW. While focal length multiplier is a term used since the advent of teleconverters, crop factor is not a new term either (as someone stated). Crop factor / format factor have been around even long before the advent of digital photography when describing differences between various (medium) formats, however, since most people are coming from the 35mm world, their first exhibition to these terms has been with the advent of APS-C DSLRs in the last decade, and since this is what most media talk about, it has become common misbelief, that these terms are new and hardwired to describe some factor based on the 24x36mm format.
-- 84.63.52.9 (talk) 02:47, 6 October 2009 (UTC)[reply]
I agree with you that *many* people...use the term focal length multiplier when they actually mean the crop factor, which is pretty good evidence that they refer to the same concept. If that usage is "incorrect", we'd need a source that says so. As for the use with respect to teleconverters, I find that term quite rare; see this book search (only four books with both "teleconverter" and "focal length multiplier", and looks like only one of those using it in the way you say). Uses with respect to the sensor size in a DSLR are much more common. I agree with you that there's a lot of misinformation published about this stuff, like that the cropping does change the depth of field, but I think what the article says now is correct: This ratio is also commonly referred to as a focal length multiplier ("FLM") since multiplying a lens focal length by the crop factor or FLM gives the focal length of a lens that would yield the same field of view if used on the reference format.
If these terms are old, show us some sources. But don't assume that the term used in German with respect to teleconverters is necessarily equivalent to the English usage. Dicklyon (talk) 04:01, 6 October 2009 (UTC)[reply]
Dicklyon, Not everything is on the Internet. The term, “Focal Length Multiplier” (FLM) is in the users manual and sometimes even printed on the sides of tele/macro converters. It has always been used that way since the dawn of lenses even in English. It is NOT a Geerman only term, although it probably originated there since most of the early work in photography optics occurred in Germany. FLM means exactly what it says; the focal length is in fact multiplied. Crop Factor/Format Factor/Enlargement Factor are also terms invented eons ago which also mean exactly what they say; ① your film is cropped from the standard (usually 10×8 inches) by a factor of,… ② your format differs from the standard by a factor of,… ③ to fit the standard, you have to enlarge by a factor of…. The last one being quite relevant since 10×8 plates were rarely enlarged., hence, they were the standard by which all other formats were judged.
Just because you are too young to have known those terms, does not make them new terms. Just because 98% of the world uses the wrong term, does not make it the right term. There is no FLM when a lens moves from one format to another. A 120mm lens on a 10×8 is still a 120mm lens on a MFT and will still bring an object at infinity in focus when placed 200mm from the film plane. You want a source, go to a library which has old books and read one (such as the Time/Life Series on Photography) or,… ask an old photographer. Instead of defending an incorrect term, you should take correction on the things that are fact such as, ① these are not recent terms if older photographers have been using them their whole life and are referenced in at least one pre-APS book —and APS predates digital cameras—, ② they do not refer to the same thing if tele/macro converters never talk about crop/format/enlargement factors but talk about FLM by stamping it in the side of the converter or spelling it out in their manual, ③ whereas multiplying the actual focal length by the crop factor will yield a focal length which gives the same FoV IF USED on the reference format, that amounts to an “if…then” of a situation which does not exists (since most people do not possess multiple formats) and it does not actually alter any focal lengths at all.
A crop is an actual crop of the reference format, a format is an actual thing and an enlargement factor is exactly how much one needs to enlarge (the width) to fit (the width of) the reference factor. The proper terms are crop factor (most common today), format factor, and enlargement factor (least common today since people rarely enlarge anymore). To see how the term is wrong, one only has to see how it is used. “Since the APS-C camera has a smaller image sensor, the focal lengths of these lenses have to be altered using a focal length multiplier.” → (LifeWire). “Focal Length Multiplier is a not-exactly-correct-but-helpful term that many like to use to describe the Field of View Crop Factor.” → (The-digital-picture.com). “It is really necessary to understand that the visible change of focal length is due to a crop, and absolutely not to any form of increase of the focal length. This one follows the optical laws : it is the distance which separates the optical center of the focal plane. We indeed speak about focal equivalence.” → (Within Lights). “…A 50mm lens on DSLR with a focal length multiplier (FLM) effectively becomes mid-telephoto. …Each camera’s effective focal length will be different than the one listed on a lens.” → (Digicam Help).
I must also point out that I checked with the manual of my camera and nowhere does it say that my camera has a FLM (nor a crop factor for that matter). If you want to find a reference to the term, “crop/format/enlargement factor”, find a book on using the new 120 travel TLR cameras (or the new 5×4 portable cameras) and not one on using the new APS-C DSLR. The.Kings.Servant (talk) 20:54, 31 January 2017 (UTC)[reply]
Here's the format (and actual camera) that I first did extensive photography and learning on.
Just because you are too young to have known those terms – Well that's just the sweetest thing I've heard in decades. Cute. And I don't need to go to the library to get the Time-Life series on photography, as I picked up a compete set at a garage sale in 1974. Dicklyon (talk) 21:19, 31 January 2017 (UTC)[reply]
Just because 98% of the world uses the wrong term, does not make it the right term. – OK, what should we call it then? Dicklyon (talk) 21:19, 31 January 2017 (UTC)[reply]
There is no FLM when a lens moves from one format to another. – Did someone say there was? FLM (meaning crop factor) is a property of a format, not of a lens or its movement. Dicklyon (talk) 21:19, 31 January 2017 (UTC)[reply]
I do agree with you that these terms are widely misunderstood, misused, and abused, and that this FLM is not the same kind of thing that your teleconverter does, which may have also been called that sometimes (though I'm too old to remember for sure). Teleconverters do literally multiply the focal length of the attached lens, by a factor called the extension factor, or magnification factor, and yes even focal-length multiplier according to this book; thanks for jogging my memory (and thanks, Google). I've actually done pretty extensive research on ideas related to format, DoF, etc., and can send you a draft paper about all that if you email me. Dicklyon (talk) 21:19, 31 January 2017 (UTC)[reply]
If a format has a “Focal Length Multiplier”, it must have something which multiplies the focal length. Since we both agree that moving the lens from one format to another does not change the focal length, then no Format has a “FLM”. It is that simple. Tele/macro converters do have a FLM because, when attached to a lens, it changes the focal length by adding additional optics to the optic formula. To recap, formats do not have FLMs, Converters have FMLs. Additionally, a crop factor is only relevant to a known and used reference format. It is a useless term to a new photographer with only one format and no reference.
P.s., I have seen your draft paper before and I think I commented to you that I consider your problem to be that you think in terms of the Circle of Confusion with a 10×8 inch reference (with enlargement and viewing distance coming into play) and not in terms of the Cone of Confusion where all the uncontrollable/unpredictable variables disappear. With The Cone of Confusion being set as a 0° 1.7' arc for the average adult human eye with 20/20 vision in a well lit room, enlargement amount, viewing distance and format size becomes clearly irrelevant to DoF calculations. One cannot always force someone to examine an image from 25cm away nor be sure that someone will not want a 90×60cm enlargement, affecting the Circle of confusion, but Cone of Confusion is always the Cone of Confusion. Hope that clears up any confusion. ;-) Applied Logics 23:41, 31 January 2017 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
You say If a format has a “Focal Length Multiplier”, it must have something which multiplies the focal length. I do not agree; that's not what it means. Think of it this way: I have a lens, which has a focal length, on my camera that has a focal length multiplier. If I multiply my lens's focal length by my camera's focal length multiplier, I get a different focal length, which tells me what lens would have the same field of view on a 35 mm (or whatever reference format) camera. I did the multiplication; my camera does NOT have something which multiplies. Dicklyon (talk) 00:03, 1 February 2017 (UTC)[reply]
Re cone or angle of confusion, I think we totally agree; or partly. This is Moritz von Rohr's method that I call the "outside the box" method. It removes all the confusion induced by format, CoC, etc. Dicklyon (talk) 00:06, 1 February 2017 (UTC)[reply]
So then we agree that ① a format cannot have a FLM nor a crop factor, nor a format factor since it has to be in relation to another specific format, that is to say, two formats can have a crop factor, etc., and ② Whatever it does has nothing to do with the focal length but everything to do with the FoV or Angle of Veiw or crop amount and definitely amount of enlargement needed to mach size, and ③ Your justification for calling it a focal length multiplier also justifies calling it an f-stop multiplier, an ISO multiplier, an EV multiplier, an ND multiplier, or anything else which may need adjusting to take the Same Image.
…Or… we can end the confusion by calling it what it is, a format factor between any two stated formats, not unique to the DSLR world or 135 vs APS-C, and only relevant when two formats are in use together, affecting several properties which need adjusting to take the Same Image with both formats, especially enlargement, and certainly not just focal length. …As opposed to a true focal length multiplier on a converter which actually multiplies the focal length of whatever lens it is attached to by a factor imprinted on the side of the converter as the FLM is a property of that converter and nothing more. Applied Logics 18:12, 1 February 2017 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
I think we agree on the basic facts. But it's not up to us to decide what it's called; we need to report what it's called and what it means. It is frequently called "focal length multiplier", or was last time I looked. Dicklyon (talk) 18:22, 1 February 2017 (UTC)[reply]
We do not need to decide what it is called (an opinion), we need to decide what it is (a fact). Wikipedia is about correcting facts, not going with the popular opinion. A cottonmouth or a water moccasin? Both is what it is called, (also a copper head water rattler) but none of those is what it is. It is Agkistrodon piscivorus. No one really calls it that. They call it (in Florida) a cottonmouth or (in Georgia) a water moccasin. Likewise, the difference between two formats is the format factor, and an FLM is a property of a converter. A converter has an FLM. No camera has an FLM and no camera has a format factor. Two cameras may have a format factor, but two cameras do not have an FLM. An FLM of a device is the factor by which a focal length changes when added to a device. People may incorrectly call it an FLM but no matter how many of them do so, they are wrong.
When a 100mm lens moves from a 135 to an APS-C, there is no change in focal length. When one moves from shooting with a 135 to an APS-C, one has to adjust the focal length, the f-stop (so as to keep the same aperture), and the amount of light or sensitivity of the film, (to keep the same amount of motion blur, same shutter speed) and all of that is done by multiplying/dividing by the format factor, which is the ratio of the width of the two formats. (The width is used to get the same FoV or Angle of View, as measured across the image. Use of the diagonal was is a long story but was a compromise way of addressing different aspect ratios.) Applied Logics 19:23, 1 February 2017 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
Since I have both a full-frame 5D Mark II and a APS-C sensor 7D, I've photographed a still life from the same position with both camera. This comparison could replace the one showing the simulated crop lines. Included in the image is a tape measured focused on the 10cm mark; crops from the tape measure could be used to illustrate the effect of sensor size on depth of field. The white balance isn't well matched between the two images, but otherwise they are the same 85mm @ f/1.8 with a focus distance of 85cm.
Your 5D Mark II picture of 1280 pixels long has a focal-plane pixel size of 36mm/1280 = 0.028125 mm, which is smaller than the customary circle of confusion criterion of 0.030 mm for that format, so it's not enough resolution to see the DoF. The full-size original image could be used for this. Same problem with the other one. See DOF Master, which shows DOF of 84.5 to 85.5 cm for the 5D, and 84.7 to 853 for the 7D (assuming it's a 1.5 crop factor, CoC 0.020 mm). Or you could adopt a different criterion and compare this pictures as they are, but it's a subtle thing to judge by eye. Dicklyon (talk) 03:35, 7 October 2009 (UTC)[reply]
The 5D Mark II has an effective 5616 pixels across a 36 mm image giving 0.0064 mm per pixels. 4.68 pixels are needed for a 0.030 CoC. The 7D has 5184 pixels across a 22.3 mm sensor giving 0.0043 mm pixel width. 6.97 pixels are needed to cover a 0.030 mm CoC or 4.65 pixels for a 0.02 mm CoC. Any APS-C camera with better than 1.9 megapixels already has a pixel width of less than half the 0.02 mm you specify. Since pixel sizes are considerably smaller than the CoC you reference, CoC becomes unimportant and irrelevant in this discussion about sensor sizes (and not film/sensor resolution, enlargement sizes and distance of viewer). When an image is re-sized for the Internet, all bets are off due to the screen size/resolution of the viewer. My 4k monitor of 3840 pixels across an 858 mm screen (rated 39" monitor) has a pixel width of 0.22mm, 7.4 times larger than the CoC so there is no point. My FullHD monitor (rated 23") of 512 mm wide has a pixel width of 0.26mm which is even worse. This all assumes that the pixels have negligible space between them. Applied Logics 19:01, 26 March 2015 (UTC)
Your test shots are inaccurate, the center point on the two are different, bring it into Photoshop and shrink the 7D one to 62.5%, you'll see that they don't line up. Please use a sturdier tripod (Manfrotto or Gitzo) and a head that locks the axis movements (three-way head, not ball head) next time.
Like Dicklyon says, you cannot tell the difference by eye, try with a subject MUCH further away for this kind of test. By my eyes, the crop has ever so slightly shallower DOF.
In this image, I've moved the 7D image using layer mode difference so that they line up perfectly. As you see they are still not the same, due to your tripod not being sturdy, the image is slightly rotated when comparing to the other.
Parallax would affect the lineup more than the ball/three-way head because one can pull off the QR mount, place it on the other camera (if it does not have its own) and place it on the locked ball/three-way head in the same position but the lens may be higher/lower on each of the two cameras. Alternately, one can take a 135 format image and crop it via software to approx 1.5 and realize that the DoF does not change by cropping an image. Image quality may change but that is more to do with sensor/film choices than format. DoF does not change, CoC is irrelevant. Applied Logics 19:01, 26 March 2015 (UTC)
In this image, 7D on the left, 5D2 on the right. I've cropped out the 5D2 image to the same size as the 7D, since I downsized the 7D, there was no upsize or program interpolation in the 5D2 image. This also means that the crop will appear to be sharper due to down sampling, please ignore that and focus on the FOV and DOF. The crop is marginally less DOF to my eyes. Again, if you are to do this kind of test, something like f/1.2 - 1.4 would be better, with a subject much further away so the DOF is obvious. Channel 49 (talk) 16:50, 9 July 2010 (UTC)[reply]
Actually, the DoF would be less obvious when the subject is farther away since the DoF will be greater. You would want the DoF to fall off quickly. What would help is putting the tape measure on an incline so that it could be read more clearly (although it would not be reading actual distance). If the angle of incline is known, (such as 53.14 degrees in a 3/4/5 triangle), the actual distance may be calculated easily (such as multiplying by 3/4). Applied Logics 19:01, 26 March 2015 (UTC)
Since the image is effectively cropped, less light is available. Should be 1/form factor², probably a bit dampened by cos^4 law (probably not much, since lenses are built to avoid vignetting). This should (these are my ideas, not read somewhere) act like an additional aperture (cameras do not care since they measure the light anyway). But the additional "aperture" is a crop only, i.e. it does not make the image sharper since the original aperture is responsible. It does however cut a lot of vignetting, fringing etc. that appeear in the outer regions. This means using "analog" objectives on APS-C should result in unchanged central image, less fringing etc but less light and thus potentially more ISO/noise (i.e. F5.6 is really about as dark as F8). May also be a good idea to use "A" mode for fixed aperture because camera may go for ISO 100 and F5. The sharpness is of course reduced by form factor compared to 35mm, since the pixels are simply smaller. --88.74.172.201 (talk) 07:14, 2 June 2010 (UTC)[reply]
The same amount of light is hitting any given square mm area of the image. Cropping off 99% of an image does not reduce the amount of light hitting the remaining 1%. Yes, 99% less light (assuming even lighting) is now being used overall but the amount of light that hit that 1% of the image does not change. f5.6 on a 135 format lens on a 135 format sensor is still f5.6 on an APS-C sensor with an APS-C designed lens or the 135 format lens. That is why hand-held light meters, whether incident or reflected, do not care what camera one uses, just aperture, film sensitivity and shutter speed. Applied Logics 19:12, 26 March 2015 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
On a 1.6 crop, a 85mm would have the field of view of 136mm (FF equivalent), while maintaining its 85mm perspective.
At the same perspective using 7D vs 5D2 with 85mm f/1.2, subject distance 10 feet, according to DOF calculator [1], a 7D would have 0.19ft DOF, while a 5D2 would have 0.29ft DOF. When comparing these cameras at the same perspective, it is clear that the crop sensor has less DOF. However, photogs will "foot zoom" just to make frames fill, without considering perspective, and since a 5D2 has a bigger FOV, they move closer. As they move closer to the subject, the DOF will shrink, thus making 5D appear to have a shallower DOF. Remember the perspective has changed, so the two shots are not the "same", even though they both fill up the frame.
Crop cameras has less DOF, but their FOV makes the photog move further away, thus increasing the DOF in practice. A good comparison would be a FF with 80mm and crop with 50mm both at f/16, subject 10ft away, since both will have the same FOV, the subject will be framed the same way. The FF will give DOF of 4.68ft while the crop will give DOF of 8.41ft, so when used for a real world application of shooting a shot with the same framing, the FF will give less DOF.
---
Shoot two shots a full figure of the same framing with a 100mm and 200mm. The 200mm will appear to have a blurrier background, as the photog has moved 2x further away to accomodate for FOV, and the change in perspective will result in 1/4 of the previous background coverage to be captured on film. This background will fill the same space and appear to give a better bokeh/blurrier background, despite the increase in DOF.
Despite what many say, 50-100mm is not best for shooting full length portraiture, as it does not isolate the subject. The effect I'm talking about is like the way that 85 @ f/1.2 isolates the face and blurs the background to oblivion. This is better achieved on a super telephoto, just like the sports shots you would see in the media, done with 300-600mm.
A good visual tutorial on perspective and focal length [2]
---
" On the other hand, compared to the 35 mm-equivalent focal length shooting a similarly-framed shot, the smaller camera's depth of field is greater. "
-This is partially correct, but it does not explain the cause, which is not the sensor size, it is the FOV, "foot zooming", and change in perspective, not a direct action through the sensor size. There needs to be an explanation stating that using the same lens on the two camera shooting the same subject from the same spot will yield shallower DOF on the crop, not the FF... The statement that is there now will be very misleading for amateur, it will make them believe that the crop DOF is greater inherently.
"When comparing two sensors of the same resolution, the crop sensor has much smaller pixels and it will exhibit less contrast and saturation than its full frame counterpart."
-This is also not a myth, check out Ken Rockwell's tests [3], just like bigger aperture, bigger sensors collect more light.
"An example would be taking an image of a rock using two cameras with the same lens. The first camera a 18mp full frame and the second a 18mp APS-C, both shooting the same composition in a stationary position. The first image will be more "wide" while the second image will be more "magnified". After bringing the results into an image editing program and enlarging the first image so that the rock is the same size in both images, one will see that the enlarged image is approximately 160% (1.6x) of the original."
-This is just an example to help explain the crop factor for those who don't understand after reading, simply means that the image is pre-cropping before going to the sensor, as explained in the article. See the images above with the lemons resized.
The above is Channel 49's attempt to justify the material of his that I reverted. See fractured conversations on his talk page and mine.
The stuff about DOF and perspective is mostly confused, or at least imprecise. Some of the other bits are strange, like citing Rockwell's page in support of contrast and saturation issues, when that page doesn't even mention that, and there's clearly no significant connection of those to pixel size.
I wouldn't say confused, it may not be precisely worded. Nothing is easier than practical examples to demonstrate.
"Having bigger pixels on a larger format means you can use cheaper lenses and usually get better results than the best lenses on a smaller format. The only reason I use the weasel word of "usually" is because as of 2007 we're only talking about a 1.6x linear format difference among consumer (under $9,000) DSLRs. In film, with a 2x linear difference from one format to the next, the crappiest lens on the next bigger format always smokes the very best lens on the next smaller format." -Look at his test images, there might be a difference in technology but not by much. He then goes on to demonstrate other things related to full frame such as better contrast, sharpness, etc. We're talking about bigger pixel as in the pixels on the sensor, the actual physical pixel that collects light. A bigger piece of film will collect more light than a smaller piece of film correct? This is no different. Compare a 35mm to 4x5 large format, the large format has major advantage on contrast and saturation on top of its larger resolution. Of course you also have to take into account that lenses for the larger format may not perform as well optically due to its size and being harder to construct.
"Bigger pixels collect more light (photons). More photons means more clean image compared to the relatively constant amount of electrons making noise. (Yes, I'm simplifying, but it's why bigger sensors are quieter and cleaner.)" -He uses the words "bigger pixels" here again.
"By about 10 MP in small formats, most of the extra pixels today are splitting up the limitations of your lenses (and our own photo abilities) into smaller pieces instead of showing us more detail. Having more pixels on a small format doesn't make a sharper picture if all these extra pixels are doing is splitting up a lens' limited resolution (or subject motion or limited depth of field or small-aperture diffraction) into more pixels." -Again, splitting pixels = smaller pixels.
Is there anyone here who can comment and help him get it right? He's inserted a lot of the same strange statements about perspective into a bunch of Canon lens articles. Dicklyon (talk) 07:32, 10 July 2010 (UTC)[reply]
One more thing: 173.206.212.205 appears to be Channel 49, not logged in. He's new around here. Dicklyon (talk) 07:35, 10 July 2010 (UTC)[reply]
Hey, what about a table with common photography formats, movie formats, and sensor sizes in comparison to 35mm in photography and Academy format movie 35mm? I'm thinking of a table which has 6 columns for each entry: Width, height, diagonal, image area, and area size ratio compared to 35mm and Academy format each as an integer. It'd help people about calculating equivalent focal lengths to achieve identical angle of view. We could also have [4] in the weblinks, people could look up the image/sensor dimensions here, then enter the values into that calculator.
I'm thinking about one table as described above (6 columns on each entry) for the common sensor sizes such as those at Image sensor format#Table of sensor sizes (keep in mind there's three different aspect ratios when it comes to sensors, 4:3, 3:2, and 16:9!), a second for the most common movie formats (70mm, 65mm, Super16, 16mm, 9.5mm Pathe, Super8/Single8, and Regular8), and a third for the most common photography formats.
Also, angle of view greatly affects not only the amount of how much of your scenery you can capture, but even moreso your perspective. It's why these issues matter so much beyond simply how much moving space the photographer has available at a given location.
You may move back and forth to capture more or less of your scenery, but you'll never be able to do the same as when you're changing your angle of view! Long lenses compress a perspective, wide-angle lenses stretch it, see perspective distortion (photography). That's why 35mm equivalent and crop factor are of such significance in photography and cinematography, people wanna know how to achieve a particular amount or lack of perspective distortion, but focal lengths are by default given for 35mm only. A single line about "BTW, perspective will change, too" just doesn't cut this. --79.193.41.249 (talk) 00:17, 20 August 2010 (UTC)[reply]
Do we have to multiply by the crop factor, or not?
I'll give an example to make this clearer: On an APS-C DSLR (eg a Canon 550), a lens might be marked 50mm, f/1.4. But I still mentally think in 35-mm terms.
So, in 35-mm equivalent terms this is about an 80 mm lens, i.e. moderate telephoto. Should I consider it as being an "80 mm f/1.4 lens", or should I consider it as an "80mm f/2.2 lens" ? — Preceding unsigned comment added by 87.194.171.29 (talk) 21:40, 15 December 2011 (UTC)[reply]
That depends on what you want to do with the aperture. For exposure calculation, relative to an ISO setting, it's still f/1.8. For DOF equivalence, it's the absolute aperture diameter that matters, do the DOF is like an 80/2.2 on a full-frame camera. Dicklyon (talk) 07:39, 16 December 2011 (UTC)[reply]
Never change the f-stop amount in either case. The f-stop is entirely dependent on lens properties and does not change regardless of sensor, film, enlagrement or anything else. It is a ratio of the diameter of the lens to the effective aperture of the lens. Regarding DoF, Dof is affected by these variables; the actual focal length of the lens, the aperture used and the distance from the film plane to the subject in focus. It is so fixed that it is often printed on older lens barrels so that one can focus in a studio without looking through the camera (or stopping down on a reflex camera). No matter the format, from an 8x10 field camera to a 4/5th format, any given focal length at the same aperture focused at the same distance would give the same DoF. Applied Logics 20:15, 26 March 2015 (UTC)
Applied Logic is correct that "The f-stop is entirely dependent on lens properties and does not change regardless of sensor, film, enlagrement or anything else." But he neglects the role of the circle of confusion criterion for acceptable blur, which is generally taken to be proportional to format size, in computing DOF, so the rest of what he says is incorrect. Dicklyon (talk) 05:40, 31 March 2015 (UTC)[reply]
The DoF is not altered by sensor, film nor enlargement for the “same image”. ①Sensor/film: as long as the sensor has an adequate resolution such that a 0° 1.7' arc (from focal center to film-plane) is significantly larger than the size of the pixels/grain, sensor/film is irrelevant. ② Enlargement: when an image is viewed at the perspective correct distance, enlargement is irrelevant. On Canon's best 135 sensor, when viewed from 1mm away, all DoF is gone but no one views an image from that sort of distance (except perhaps a billboard technician placing/removing a banner intended to be viewed from several meters away, etc.). Circle of Confusion is a farce. Cone of Confusion is far more meaningful.
The same aperture on a 100mm lens and a 35mm lens gives the same DoF at the same distance to subject. Reason; the same aperture (not f-stop) gives the same cone of confusion. DoF (Cone of Confusion) only affected by effective aperture for any given subject distance. Sensor size, enlargement, focal length all not relevant This Cone of Confusion remains the same when an image is viewed at the perspective correct distance. Stand closer and your DoF decreases, stand further back and the DoF increases. In both cases, the Angle of View changes with change in viewing distance.
DFN → “Same Image”: Both images, taken with different cameras, has precisely the same perspective, framing, focus, DoF, motion blur, exposure. Perspective is preserved by using the same position in three dimensional space in reference to subject and focal center of lens. Framing is preserved by using the appropriate focal length lens. Focus is preserved by placing the lens/focal plane appropriately. DoF is preserved by using the same aperture (not focal ratio or f-stop). Motion blur is preserved by using the same shutter speed. Exposure is preserved by adjusting the lighting (such as in a studio), sensitivity or using ND filters. Regardless of how large the first format is or how small the second format is, when enlarged to the same size, they will have the same DoF (when viewed from the same distance).
DFN → “Perspective Correct Distance”: The focal length of the lens used multiplied by the amount of enlargement done. For a 10×8 inch format contact print (enlargement factor of 1) done with a normal lens, (245mm focal length), at an aperture of f/20, the perspective correct distance is approximately 10 inches or 24.5cm to be precise (1×245mm). This is why DoF is often referenced as a circle of confusion on a 10×8 inch print held 25cm (or, imprecisely, about 12 inches) away from the face. However, using an APS-C camera (with a crop factor of 10.1), one would use a 24mm lens (more precisely, 24.26mm) with an aperture of f/2 and enlarge by a factor of 10.1 to fit a 10×8 inch frame, and still view it at 24.5cm away (10.1×24.25mm). That is why Cone of Confusion is a better term.
Due to technical reasons, the 10×8 inch plate has an effective size of about 240mm, not 245mm and the ‘normal lens’ is 240mm . The APS-C frame is actually about 23.7mm, not 24.26mm but a ‘normal lens’ is still approximately 24mm (23.7mm). This brings the crop factor back to 10.1 anyway. I merely used approximations for easier maths. Also, if both cameras are 24mpx, each pixel will get exactly the same amount of light. The APS-C has smaller pixels but its sensor is receiving more light per surface area. Cell saturation is also not a problem since a less sensitive sensor setting or ND filters (or less light) will be needed for correct exposure.
That being said, a 240mm lens on a 10×8 inch camera at f/20 and the same 240mm lens on an APS-C at f/20, focused on the same subject from the same spot, still gives the same DoF when viewed from the perspective correct distance because the Cone of Confusion has not changed. Applied Logics 23:03, 31 January 2017 (UTC)
Why does the table in the article, per its footnote, give crop factors calculated from the height ratio? I thought the crop factor was calculated on the diagonal? At least I am used to think that a 1/1.7" sensor had a crop factor of 4.55, not 4.2 as stated. Even worse, the table is inconsistent. The crop factor for 1" sensors (RX100) given is the one based on the diagonal (2.7). Based on height, it'd be 2.5. I think this table should agree with the one in Image sensor format. Lupo 22:51, 11 November 2013 (UTC)[reply]
Crop factor or Format factor or enlargement factor, is actually the ratio between the width of two factors, indicating the ① amount of enlargement one needs for the smaller format to fit the width of the larger format (directly proportional), ② the the focal length of the lens needed on the smaller format to get the same Field of View (FoV) or Angle of View (AoV) as the larger format (inversely proportional), and ③ the change in f-stop required on the smaller format to give the same aperture and thus Depth of Field (DoF) as the larger format (inversely proportional) and ④ the change in Exposure Value (EV) needed on the smaller format to attain the the same shutter speed, thus motion blur, as on the larger format (inversely proportional).
Because FoV or AoV is determined by the horizontal distance (width) and not the height nor diagonal, one ought to use the width. The use of the height came into play during the “format wars” of the 120 era, when the standard 5:4 frame dimension ratio was being replaced, first by a 1:1 square, then the 3:2 golden ratio, then the 2:1 wide ratio, etc. To fit the 10×8 inch print in a reasonable way, people started to use the diagonal as a compromise comparison of apples and oranges. This actually throws FoV and AoV out the window. Why that table uses height is beyond me. Perhaps it was used for portrait photographers, more concerned with fitting an image in the vertical space. Applied Logics 16:45, 2 February 2017 (UTC)
This entire section needs to be deleted since one cannot estimate sensor performance based on sensor size just like that. It has so much more to it. HP5 vs XP2 vs Delta400 are all the same size and speed film but different performance even from the same manufacturer, Ilford. the comparison gets more dramatic if we included the Kodak Tri-X and T-Max 400 films. The same is true for digital sensors. A 24 megapixel Sony APS-C CMOS sensor with a PRIME III image engine (Pentax K-3) can yield better performance than a 16 megapixel 135 format CMOS sensor with an EXPEED 3 image engine (Nikon Df) in terms of noise, light sensitivity (ISO 100-51,200 vs ISO 100-12,800) and metering (-3 to 20 EV with an 86,000 pixel sensor vs 0 to 20 EV with a 2016 pixel sensor) even though the former has much smaller pixels (0.0039 mm versus 0.0073 mm) and much higher resolution. Applied Logics 20:15, 26 March 2015 (UTC)
The discussion compares photon-limited performance for sensors that "are made using similar technology", at equal "exposure". It's still pretty biased by equating exposure, or focal-plane illuminance and time, but what it says is basically true, and would apply to the "ideal sensor" that faithfully records the exact location of every incident photon. If instead one would fix the aperture diameter and exposure time, then ideal sensors of different sizes would all see exactly the same set of photons and would behave alike and make identical photos. Dicklyon (talk) 21:03, 26 March 2015 (UTC)[reply]
But this is Wikipedia and we need information that would be useful for the general audience. That sort of theoretical analysis which does not serve the public is not only useless to them but misleading. Someone trying to get information before buying a DSLR may say, "I do not want a high resolution, APS-C camera but a low res full frame because I read somewhere that they have better signal to noise ratio and I need that for low light." That would lead them to buy the totally wrong camera. The theoretical information you gave is only relevant to the engineers who make this stuff and they are not looking to Wikipedia for answers.
This section remains in error since no one in the real world can use sensor size nor sensor resolution to estimate sensor performance because there is no real-world case where their exists "sensors of 'similar technology'" especially when you have not stated what "similar technology" is. I compared two CMOS sensors and did not even go into CCD sensors which are in theory more prone to saturation (and 'signal overflow') yet better at low noise in low light (in theory). None of that section actually works in the real world and is extremely misleading. We can even look at the case where Pentax had switched to the Sony sensors a few iterations back. So the last three iterations all used the Sony sensor technology, each iteration having higher resolution than the last (meaning smaller pixels) and all having better performance than the predecessor despite having the same Sony technology. In addition, the Sony cameras had the exact same sensors as the Penntax cameras but the Pentax had out-performed them in some areas, not due to the sensor, but due to the PRIME II engine and/or the lack of a AA filter.
The only place that section belongs in is a, "Problems with engineering a digital sensor" but this article is about "crop factor" not Digital sensors. Crop factor became an issue back in the fifties (1950s) when 135 format film hit the market and everyone thought it was just a play thing. The format only took off when photojournalists started using it and from there, consumer interest grew and from there, it worked its way into some other professional markets. All the same discussions we are having now are the same discussions we had then but you have thrown irrelevant misinformation about digital sensors into the mix with this section.
No consumer, nor professional can actually use this section in the real world when picking out a camera. Applied Logics 15:01, 27 March 2015 (UTC)
If you see a way to improve the section, do it; if it gets reverted, then per WP:BRD, let's talk. Deleting it does not improve it.
There is no way to improve a section which is both irrelevant to the article and states full misinformation except to 1) delete it and 2) explain why it is misinformation. The second option is like an article on vegetables vs fruits and a section on mayonnaise which states, "mayo is made from animal fat and is often put on bananas," and the correction reading, "mayo is made from oil, vinegar and eggs and is used as a general condiment." Although the 'correction' is adequate, it is still irrelevant to the main article.
This article is about crop factor which is a concept from the days of 120 roll film which is used in several different size formats from 6x9cm to 6x3cm and how it and the 4x5 inch large format relate to the 'normal' size of 8x10. This section relates to digital sensor performance (which has nothing to do with crap factor) and erroneously claims it has --or only has, as it mentions no other factors-- a direct relationship to sensor size.
If it belongs anywhere, it belongs in an engineering article on digital camera sensor design and not in an article about a concept born over one hundred years ago. Applied Logics 02:11, 31 March 2015 (UTC)
This would be a good place to "explain why it is misinformation". Then we can discuss, and if there's a consensus perhaps delete or improve it.
Also, I admit I'm not familiar with the early history of "crop factor" that you describe re 120 and larger formats. Do you have references to suggest? Dicklyon (talk) 05:37, 31 March 2015 (UTC)[reply]
I just started discussing why it is misinformation. What do you mean, "start discussing?" Is there anything more to add to this discussion as to why it should remain? The discussion has started quite a long time ago. (20:15, 26 March 2015 to be exact). I have explained why it was misinformation and needs to be removed. You alone claimed it was "correct" information based on "similr technology" then I explained why the "similar technology" argument does not hold up except for sensor design engineers. To which you suggested an edit to improve the misinformation to which I explained why the deletion is the proper edit. Why now, "this will be a good place to begin... then we can discuss"? It seems to me that you are dictating the content of the page. Having taken everything I said about that section, is there anyway which you can see edits being made to the current content to make it more useful and less misleading without deletion? Please help me edit it thusly as there is no correlation in the real world between the two plus sensor performance still does not belong in a discussion on crop-factor as it still applied to the Nikon 35mm SLR film camera vs its APS-C SLR film camera, neither of which had digital sensors. Indeed, they both had film and it was indeed film of "similar technology" as used in their 35mm offerings as I could put TMax 400 or Fujifilm Velvia in either camera but we do not have that choice in digital cameras, hence, no similar technologies.
Of course, that brings up the issue of Ilford HP5 in 35mm vs Kodak TMax100 in APS-C but discussions on film technologies or digital sensor technologies belongs in articles about film & sensor designs & technologies and not in articles about crop-factors.
A recent reference to crop factor regarding 120 and medium formats, etc.? Why no! But if you want me to dig up an early Photog book pre the 35mm film era, then sure, I'll look into that. In the mean time, my references are old photographers, most of whom are dead. I'll try find a book on the obvious as it seems unreal to believe that "crop-factor" is something which only came into being with the invention of digital cameras when it must have clearly been around from the day the first 5x4 inch camera was invented to compete against the 10x8 inch behemoths. I will actually do this search. It may take time. In the interim, think about the first 5x4 inch camera. Applied Logics 20:13, 14 April 2015 (UTC)
I have quoted and analyzed a lot of the old-book commentaries on different formats and how they relate to DOF in my draft paper on DOF at my website. The term "crop factor" is modern by comparison, as it came about from the idea that lenses designed for one format (35 mm full frame) would be used on another (e.g. APC-C sized digital). In previous generations, the idea of taking a line of lenses designed for 8x10 and using them on 4x5 was not really thought of in those terms (though it does happen). As far as I can tell, no nineteenth- or twentieth-century book contains this term.
I remain unclear on why you think what the article says now is "misinformation", or why the best way to fix it is to delete it. Dicklyon (talk) 20:33, 14 April 2015 (UTC)[reply]
Why is it unclear? Sensor technology is different in each and every camera for several reasons.
They all do not use the same sensor.
Even those that do use the same sensor, do not use the same processor chip.
Even those that do use the same procesor chip, don not necessarily use the same software.
Sensors come in different sizes in different formats such that a 135 format may have a 16MPx sensor while an APS-C format may have a 36Mpx sensor and one sensor is not necessarily worse than the other in low light, high ISO, SNR, etc, as a function of its size nor its resolution.
Sensors come in different technologies such as CCD or CMOS and have different behavior characteristics.
Even CMOS sensors come in different technologies such as Bayer vs X-Trans, RGBG vs RGBE, OLPF vs no AA filter, et al.
In light of all this, the section on estimating sensor performance is absolute misinformation since sensor performance cannot be correlated to sensor size in the real world. Because it does not have a correlation in the real world, it does not belong in an article on Crop Factor. Applied Logics 17:31, 2 February 2017 (UTC) — Preceding unsigned comment added by Applied Logics (talk • contribs)
On your other changes, like this one, that's just wrong. Note the DOF depends on sensor size via the Circle of confusion criterion; ignoring that gets people into trouble, or into errors, like yours. Dicklyon (talk) 03:47, 28 March 2015 (UTC)[reply]
We can discuss the DoF/CoC stuff but I made an error in making several major correct edits at one time and you reverted them all because of a disagreement on one small issue. I will redo all the other edits and discuss the DoF/CoC issue when I put all my thoughts together. Applied Logics 02:11, 31 March 2015 (UTC)
I'm afraid Dicklyon is right about DoF. Samsara 03:10, 31 March 2015 (UTC)[reply]
Not when one speaks of the "same Image". If it is the "Same Image" criterion, then their is no difference. The reason why Dicklyon seems confused is because the CoC is not defined as a width of a circle but as a cone of confusion or feild-of-view of confusion. Otherwise, he would also have to state that DoF depends on the ultimate size of enlargement and the distance to which the enlargement will be viewed. By defining the CoC correctly as an angle of approximately 0° 1.7′ ( horizontally, for an average European adult with both eyes open in a well lit room), then lens makers can (and do) put DoF rings on their lens as it has nothing to do with enlargements or distance to image while taking the picture. What becomes important is that angle of 0° 1.7′ which remains the same on a 10x8 inch, 4x5 inch, 36x24mm or 24x16mm film frame.
If one continues to define it for a 10x8 inch enlargement held at 30 cm (or 25 cm) from the face, then one has to talk about the enlargement of the subject and not the frame to compare the "same Image" and one would end up with the same enlargement factor, hence no change in DoF. That is to say, any given lens focused on any given subject will create an image of that subject on the film of exactly the same size requiring exactly the same enlargement regardless of whether that film is 24 mm or 24 ft wide. It is a lot to explain the CoC properly but it is based on optics. The size of the pupils causes diffraction, producing a circle of light surrounded by fainter rings of light even for a point source. Therefore two point sources have to be separated by a certain angle of view to produce within the eye, two distinct circles/rings of light. With two eyes beside each other, we do better at this separation for two sources of light beside each other than above each other (but neither here nor there for this discussion). This angle of separation on the film dictates the DoF and that does not change with enlargement of the subject (as subject size on film is the same) or when viewed from "perspective correct" distance (that is the distance such that the viewer sees what the photographer saw in the camera and is dependent on the size of enlargement and focal length of lens). Applied Logics 19:37, 14 April 2015 (UTC)
That 0° 1.7′ angle of confusion criterion (0.5 mrad) corresponds to a print COC near 1/2000 of the print diagonal at normal viewing distance (more or less, depending on exactly what you use for normal viewing distance). But since normal viewing distance is generally proportional to print size, that makes the result dependent on any crop that happens. Manufacturers mark it as you describe for a normal lens for a format. For longer lenses the angular cone of confusion requirement is tighter, and for a wide angle lens it's looser. In general, it's adjusted by something like the ratio of format diagonal to focal length. Format matters. See my draft paper linked in the section above for the historical roots and viewpoints on this. For a "perspective correct" viewing distance, it's a completely different result; but nobody does that. Dicklyon (talk) 20:51, 14 April 2015 (UTC)[reply]
The 0° 1.7′ Angle of confusion does have a correspondence to a certain size Circle of Confusion on a certain size enlargement at a certain distance from the face. But that is the point. If one enlarges a 10×8 to 30×24 and views it from the same 25cm, the DoF changes but if one views it from 75cm away —no one is expected to view a print that large from 25cm away— then the DoF becomes normal again because one still has the 0° 1.7′ Cone of Confusion. DoF becomes an irrelevant myth when viewing distance and enlargement is taken out of the picture. Cone of Confusion removes that confusion. Manufacturers mark it on many lenses including zoom lenses because the Cone of Confusion is fixed by the actual aperture. The FoV and AoV is adjusted to the width of the format because that is how FoV and AoV are measured; along the horizontal plane. But since the Cone of Confusion has no bearing on the field of view, it is not adjusted at all. Applied Logics 17:09, 2 February 2017 (UTC)
I have just modified one external link on Crop factor. 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:
When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.
This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 18 January 2022).
If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
If you found an error with any archives or the URLs themselves, you can fix them with this tool.
