snubbr.com

Focal length multiplier
Hello,.

I have just a couple of questions that I want to understand about FLM..

1. For example, say that I have a Canon Digital Rebel XTi with an FLM of 1.6x, if I attach a 28-200mm lens with that lens actually become 44-320mm on the camera? Or are there digital specific lenses that are not affected by the smaller sensor size than 35mm because they cover the exact area of the digital sensor?.

Fuji Film S9100My Photos: http://flickr.com/photos/kris91/s9100/s9600 Flickr Group:http://www.flickr.com/groups/37994085@N00/..

Comments (17)

Usapatriot wrote:.

Hello,.

I have just a couple of questions that I want to understand about FLM..

1. For example, say that I have a Canon Digital Rebel XTi with an FLMof 1.6x, if I attach a 28-200mm lens with that lens actually become44-320mm on the camera? Or are there digital specific lenses that arenot affected by the smaller sensor size than 35mm because they coverthe exact area of the digital sensor?.

The focal length of the lense does not change, it is an optical property of the lense itself. The field of view changes, which is what you are referring to when using lenses on sensors smaller than 35mm full frame..

Digital specific lenses, eg Olympus Zuiko Digital lenses, fit the imaging sensor perfectly and do not crop the image in any way. The FOV on a 4/3 sensor with a 4/3 lense is still similar to the FOV on a 35mmFF sensor using a full frame lense of twice the focal length..

In a direct response to your question, the FOV is increased by 1.6X when using the Canon Digital Rebel XTi. So a 28-200mm lense has the same FOV as a 35mmFF camera/lense of 44-320mm. The focal length is unchanged, but understanding this doesn't affect the outcome of the image..

Fuji Film S9100My Photos: http://flickr.com/photos/kris91/s9100/s9600 Flickr Group:http://www.flickr.com/groups//photos/timskis6/..

Comment #1

Usapatriot wrote:.

Hello,.

I have just a couple of questions that I want to understand about FLM..

1. For example, say that I have a Canon Digital Rebel XTi with an FLMof 1.6x, if I attach a 28-200mm lens with that lens actually become44-320mm on the camera? Or are there digital specific lenses that arenot affected by the smaller sensor size than 35mm because they coverthe exact area of the digital sensor?.

As Tim said, nothing about he lens changes when you put it on a different camera..

Imagine mounting a lens on a clamp stand of some sort so that it projects an image of a landscape onto a convenient wall. Now stick a piece of paper to the wall in the centre of the image circle. That is your sensor / film. How much of the landscape you get on the piece of paper depends on (i) the focal length of the lens (a higher focal length magnifies the image more, so you get a smaller part of the landscape blown up onto the paper) and (ii) the size of the piece of paper. If you replace the piece of paper with a smaller one you are selecting a smaller part of the image. This has a comparable result in practice to using a longer focal length with a larger piece of paper, but in this case the focal length of the lens is not changing, you are just selecting a smaller part of the centre of the image (which will need more enlargement to make a print later)..

Digital specific lenses obey the same rules in terms of focal length etc., it's just that they are physically smaller and cast a smaller image circle - the degree of magnification of the image in the circle is the same. If your piece of paper on the wall is only one foot across, there is no need to have a lens that casts an image circle three feet across; using smaller bits of glass you can get an image of the same magnification (focal length) but whose image circle only just covers the paper, so the outer edges of the image - which you didn;t want anyway - are lost. Very efficient. the problem is now that if you want to use such a lens with a larger piece of paper you can't, as you will get vignetting at the edges (which is why you can't use digital-specific lenses, wi ha small image circle designed for small sensors, on full-frame cameras)..

I hope this helps..

Best wishesMike..

Comment #2

Hi,.

FLM's are just a way of comparing FoVs between different formats (film or CCD etc) and are used because they are handy. (Also it saves the makers wasting money on R&D when they can unload all their old lenses on you.).

It would make more sense to describe the FoV in degrees, which some do and explain that the FoV is measured across the diagonal. Even then you can get more confused as some cameras can switch between 16:9 and 4:3 and 3:2 and so on (but I've not seen a 4:5 for some time - so it's a bit easier these days in digital)..

Regards, David..

Comment #3

It would make more sense to describe the FoV in degrees, which somedo and explain that the FoV is measured across the diagonal. Eventhen you can get more confused as some cameras can switch between16:9 and 4:3 and 3:2 and so on (but I've not seen a 4:5 for some time- so it's a bit easier these days in digital)..

The diagonal is a good way to express angular field of view, but it can cause confusion when trying to use it to determine crop factors (aka focal length multiplier). The ratio of the diagonal of a sensor to the diagonal of the 35 mm frame only gives the correct crop factor when the aspect ratio of the sensor is 3:2. As you note, there are other aspect ratios out there. For instance, the ratio of the standard four-thirds sensor diagonal (22.5 mm) to the 35 mm frame (43.26) is about 1.92, while the correct crop factor is 2.0. The only way to determine the crop factor reliably across different formats is to compare the longest dimension of each frame..

Davehttp://www.pbase.com/dsjtecserv..

Comment #4

The information given by the previous posters is accurate and excellent. If you want to learn in more detail how and why the crop factor (aka FLM) is the way it is, you might want to see this thread: http://forums.dpreview.com/...forums/read.asp?forum=1002&message=26735620.

Davehttp://www.pbase.com/dsjtecserv..

Comment #5

Dsjtecserv wrote:.

It would make more sense to describe the FoV in degrees, which somedo and explain that the FoV is measured across the diagonal. Eventhen you can get more confused as some cameras can switch between16:9 and 4:3 and 3:2 and so on (but I've not seen a 4:5 for some time- so it's a bit easier these days in digital)..

The diagonal is a good way to express angular field of view, but itcan cause confusion when trying to use it to determine crop factors(aka focal length multiplier). The ratio of the diagonal of a sensorto the diagonal of the 35 mm frame only gives the correct crop factorwhen the aspect ratio of the sensor is 3:2..

This is wrong. The ratio of diagonals *is* the correct way to measure crop factor - it equates to the diameter of the image circle and is independent of the aspect ratio..

As you note, there areother aspect ratios out there. For instance, the ratio of thestandard four-thirds sensor diagonal (22.5 mm) to the 35 mm frame(43.26) is about 1.92, while the correct crop factor is 2.0..

Also incorrect. The sensor diagonal of a Four Thirds sensor is 21.63 mm giving a crop factor of exactly 2.0. See http://www.four-thirds.org/en/about/standard.html..

Of course, not all "Four Thirds" cameras necessarily conform precisely to the standard..

The onlyway to determine the crop factor reliably across different formats isto compare the longest dimension of each frame..

The only way is to compare the diagonals...

Comment #6

Steve Balcombe wrote:.

Dsjtecserv wrote:.

It would make more sense to describe the FoV in degrees, which somedo and explain that the FoV is measured across the diagonal. Eventhen you can get more confused as some cameras can switch between16:9 and 4:3 and 3:2 and so on (but I've not seen a 4:5 for some time- so it's a bit easier these days in digital)..

The diagonal is a good way to express angular field of view, but itcan cause confusion when trying to use it to determine crop factors(aka focal length multiplier). The ratio of the diagonal of a sensorto the diagonal of the 35 mm frame only gives the correct crop factorwhen the aspect ratio of the sensor is 3:2..

This is wrong. The ratio of diagonals *is* the correct way to measurecrop factor - it equates to the diameter of the image circle and isindependent of the aspect ratio..

As you note, there areother aspect ratios out there. For instance, the ratio of thestandard four-thirds sensor diagonal (22.5 mm) to the 35 mm frame(43.26) is about 1.92, while the correct crop factor is 2.0..

Also incorrect. The sensor diagonal of a Four Thirds sensor is 21.63mm giving a crop factor of exactly 2.0. Seehttp://www.four-thirds.org/en/about/standard.html..

Of course, not all "Four Thirds" cameras necessarily conformprecisely to the standard..

The onlyway to determine the crop factor reliably across different formats isto compare the longest dimension of each frame..

The only way is to compare the diagonals..

Steve:.

Good points. There is, however, another side to both of them..

I took the dimension of the Four Thirds sensor from dpreview's list of sensor sizes: http://www.dpreview.com/...learn/?/Glossary/Camera_System/sensor_sizes_01.htm. Note that this gives the diagonal specifically as 22.5 mm and the sensor width as 18.0 mm; the only way that data creates a crop factor of 2.0 is using the width, not the diagonal. While I generally take dpreview to be authoritative, it is possible that they need to reconcile their list with the Olympus standard..

The Olympus page doesn't indicate standard sensor dimensions that create the diagonal of 21.63, but if it has an aspect ratio of 4:3 if must be 17.304 by 12.978. Do you have a reference that confirms that?.

I didn't make up the rationale for using the longest dimension as the baseline for the crop factor; I read it a while back from a source I took to be authoritative. I doubt that I can remember the source, but I'll look into it. At any rate, this is what I understand to be the reasoning behind using the longest dimension:.

As you point out, a crop ratio can be determined from the diameter of the circle in which the rectangular sensor dimensions are inscribed; this would be the diagonal of the respective formats. If our pictures were circular that would be that; we would know all we need to know about the respective fields of view. However, actual sensors are rectangular, and there are an infinite number of rectangles that can be inscribed within a circle of the same diameter. Different rectangles would give very different impressions of the amount of cropping, depending on their aspect ratio. For instance, 3:2 ratio sensor of about 18 mm x 12 mm would have the same diagonal (21.63 mm) as the standard Four Thirds sensor, as would a square sensor of 15.92 mm. The 3:2 ratio sensor would give the impression of a uniform crop around all edges, while the square sensor would appear to have a greatly narrowed horizontal FOV, but less loss of vertical FOV..

In other words, there are really three crop factors for any given rectangular sensor size, of which the diagonal is one; the other two will be different if the aspect ratio is different. What I gathered from my reading is that comparing the horizontal field view (which is normally the longest dimension) provides a truer representation of the relative fields of view since it follows the change in width of the view. That would be a psychological/aesthetic rationale, but it would seem to be no less useful (and perhaps more) than use of the diagonal as a reference point..

At any rate, it makes sense to me, and it seems that other reputable sources, including dpreview, follow the same method. My original comment was too categorical in referring to the longest dimension as the "correct" method, but it does seem to be widely used..

Thanks for the feedback; I appreciate the opportunity to go back and ponder the origins of what I think I know, and perhaps reassess it..

Dave.

Http://www.pbase.com/dsjtecserv..

Comment #7

Dsjtecserv wrote:.

I took the dimension of the Four Thirds sensor from dpreview's listof sensor sizes:.

Http://www.dpreview.com/...learn/?/Glossary/Camera_System/sensor_sizes_01.htm. Note that this gives the diagonal specifically as 22.5 mm and the sensor width as 18.0 mm; the only way that data creates a crop factor of 2.0 is using the width, not the diagonal. While I generally take dpreview to be authoritative, it is possible that they need to reconcile their list with the Olympus standard..

Well, it's wrong. You've then gone on to infer that the crop factor must be based on the horizontal dimension, simply because the numbers happen to fit - and now you are stating that inference as fact. It's not, it's just your assumption on incorrect information..

The Olympus page doesn't indicate standard sensor dimensions thatcreate the diagonal of 21.63, but if it has an aspect ratio of 4:3 ifmust be 17.304 by 12.978. Do you have a reference that confirms that?.

I gave you a link to the official Four Thirds consortium web site, isn't that authoritative enough?.

I didn't make up the rationale for using the longest dimension as thebaseline for the crop factor; I read it a while back from a source Itook to be authoritative. I doubt that I can remember the source, butI'll look into it..

Under the circumstances it would be worthwhile to find it if you can..

At any rate, this is what I understand to be thereasoning behind using the longest dimension:.

As you point out, a crop ratio can be determined from the diameter ofthe circle in which the rectangular sensor dimensions are inscribed;this would be the diagonal of the respective formats. If our pictureswere circular that would be that; we would know all we need to knowabout the respective fields of view. However, actual sensors arerectangular, and there are an infinite number of rectangles that canbe inscribed within a circle of the same diameter. Differentrectangles would give very different impressions of the amount ofcropping, depending on their aspect ratio. For instance, 3:2 ratiosensor of about 18 mm x 12 mm would have the same diagonal (21.63 mm)as the standard Four Thirds sensor, as would a square sensor of 15.92mm. The 3:2 ratio sensor would give the impression of a uniform croparound all edges, while the square sensor would appear to have agreatly narrowed horizontal FOV, but less loss of vertical FOV..

In other words, there are really three crop factors for any givenrectangular sensor size,.

I'm sorry but you are in danger of descending into absurdity in your attempts to justify your mistake. Sometimes the horizontal field of view is more important, sometimes the vertical, depending on the subject and composition. But the crop factor is a single representative figure for the system, it has nothing to do with subject matter..

Of which the diagonal is one; the other twowill be different if the aspect ratio is different. What I gatheredfrom my reading is that comparing the horizontal field view (which isnormally the longest dimension) provides a truer representation ofthe relative fields of view since it follows the change in width ofthe view. That would be a psychological/aesthetic rationale, but itwould seem to be no less useful (and perhaps more) than use of thediagonal as a reference point..

At any rate, it makes sense to me, and it seems that other reputablesources, including dpreview, follow the same method..

No, dpreview doesn't use that method. You have jumped to that conclusion based on the width of the non-standard Olympus sensor which simply happens to be half the width of a 'full frame' sensor..

Read the opening paragraph on this page about crop factors (referred to, unwisely in my view, as focal length multiplier):.

Http://www.dpreview.com/.../?/Glossary/Optical/Focal_Length_Multiplier_01.htm.

My originalcomment was too categorical in referring to the longest dimension asthe "correct" method, but it does seem to be widely used..

Widely? Errors do have a nasty habit of propagating when they appear credible, but I'm not aware that this particular error has done that..

Thanks for the feedback; I appreciate the opportunity to go back andponder the origins of what I think I know, and perhaps reassess it..

Dave.

Http://www.pbase.com/dsjtecserv..

Comment #8

Steve:.

I'm very sorry if I have offended you; from your reaction this seems to be something you take personally. As I indicated, the information I have was credible; not only that it makes a degree of sense. Unlike the focal length itself, there is nothing inherent in the physics of either lenses or sensor that decrees that the diagonal is the measure of what we have come to call the "crop factor"; it is just one of three dimensions that might be used. The fact that it is constant regardless of the aspect ratio is both an advantage and a disadvantage: while it does provide a consistent relationship between two imaginary circles, it says little about how this plays out on the actual rectangular sensor; I believe that was the point of the material I read. Clearly you disagree, and I agree there is precedent for the diagonal approach, but the rationale for "longest dimension" approach is neither absurd nor illogical. Whether mistake or not, it is consistent with dpreview's report of the Four Thirds sensor dimensions.

If the diagonal is 21.63 mm then the side CANNOT be 18.0 mm. If dpreview got the dimensions wrong, so be it; perhaps you should draw their attention to that. But it (and other dimensions in the table) is consistent with the method I understood to be accepted, and my reference to that hardly warrants your derision..

I appreciate the additional references. I don't appreciate your dismissive tone. Good day..

Davehttp://www.pbase.com/dsjtecserv..

Comment #9

Dpreview is guilty of publishing an assumption as fact - i.e. that the size of Olympus sensors (presumably where the 18x13.5 figure has been taken from) conforms with the Four Thirds standard. You then took that misinformation (hardly your fault, obviously) and drew another assumption from it, and published that as fact. An understandable mistake, perhaps. But what you then did is go on to *invent* a justification for your mistake, and your latest post continues to do that. The justification is an invention, just as your supposed definition of crop factor was an invention.



There is far too much misinformed nonsense provided in the name of fact on this forum, and if I have bitten in your case, that's too bad. And note that I did no more than correct the mistake in your first post, and took the trouble to look up the authoritative reference for you. It was your second post that caused me to bite...

Comment #10

Dsjtecserv wrote:.

... there is nothing inherent in the physics ofeither lenses or sensor that decrees that the diagonal is the measureof what we have come to call the "crop factor"; it is just one ofthree dimensions that might be used..

Actually there is something inherent in the physics of lenses which makes the use of the diagonal appropriate. That is, the image circle of the lens. Ignoring the use of internal baffles, the lens produces a circular image, into which could be fitted any sensor having a diagonal equal to the image circle..

It is notable that the Four Thirds System specification states the image diagonal which is to be used..

Contrary to common belief, the four thirds system does not necessarily mean the image aspect ratio is 4:3. The implementations so far happen to have that aspect ratio but as I understand it, there is nothing to stop future designs using a 3:2 or square sensor which would still have the same crop factor.Regards,Peter..

Comment #11

Sherwoodpete wrote:.

Dsjtecserv wrote:.

... there is nothing inherent in the physics ofeither lenses or sensor that decrees that the diagonal is the measureof what we have come to call the "crop factor"; it is just one ofthree dimensions that might be used..

Actually there is something inherent in the physics of lenses whichmakes the use of the diagonal appropriate. That is, the image circleof the lens. Ignoring the use of internal baffles, the lens producesa circular image, into which could be fitted any sensor having adiagonal equal to the image circle.It is notable that the Four Thirds System specification states theimage diagonal which is to be used..

Contrary to common belief, the four thirds system does notnecessarily mean the image aspect ratio is 4:3. The implementationsso far happen to have that aspect ratio but as I understand it, thereis nothing to stop future designs using a 3:2 or square sensor whichwould still have the same crop factor.Regards,Peter.

Peter:.

The image circle of the lens is independent of the sensor size, and the focal length of the lens, for that matter, and thus doesn't factor into the crop factor. I agree that any sensor with a diagonal less than the diameter of the a circle (disregarding the lousy image quality at the edge) could use that lens. But, for instance, both a full frame camerra and a crop sensor camera could use a lens made for 35 mm, so the diameter of it's field doesn't tell us anyting about the crop ratio of the sensor involved. Lenses made exclusivley for crop sensor cameras can take advantage of this and economize by not having as wide a field as is needed for full frame, but the actual diameter of the field doesn't have a direct relationship to either the sensor size or the focal length..

You make a good point the "Four Thirds" doesn't necessitate an aspect ratio of exactly 4:3. Freed of that contraint, a sensor could have a 2.0 crop ratio at both the diagonal and the longest side. The data provided by dpreview, however, is consistent with a 4:3 ratio and a 2.0 crop on the side, but not diagonal. That may be wrong, but it is what is..

If the only thing defining Four Thirds is the diagonal, and every rectangle with that diagonal is considered to have the same crop factor, the point made by my original source is reinforced. A Four Thirds sensor with a 3:2 ratio and one a 1:1 ratio would be very different sensors, with very different applications because of their different rectangular fields of view. The crop factor based only on the diagonal would obscure those differences, while one based on the side would aid considerably in understanding the usable field of view relative to 35 mm. While I didn't originate it, I continue to think that concept is of value..

Dave.

Http://www.pbase.com/dsjtecserv..

Comment #12

As a newbie to the DSLR world, I find the discussions of crop factor and FLM quite interesting. I agree that using the diagonal dimension is the only way to determine the FLM. I am still trying to get a grip on how various sensors are sized. From the four-thirds specification I see that it is aimed at lens design. It is definitely necessary to standardize that..

However, my camera is a Pentax K100D which has a 23.5x15.7 sensor. That gives a 28.3 diagonal which is larger than the four-thirds standard. It is not unique in the fact that most of the Cannons and Nikons have a 27 or 28 mm diagonal. I guess these are what is called an APS-C sensor. Anyway a lens designed to the four-thirds specification would not give full coverage on these sensors. I assume that Pentaxs DA lenses and others designed for these must be designed to a different standard.

Shouldnt the primary standard cover the larger sensor (1,5 FLM) which would also include the four-thirds..

I definitely find these forums educational and thought provoking. Thanks for the good discussion guys..

DavidDallas, TX..

Comment #13

DSHarned wrote:. Shouldnt the primary standard.

Cover the larger sensor (1,5 FLM) which would also include thefour-thirds..

I definitely find these forums educational and thought provoking.Thanks for the good discussion guys..

Dave,.

Not sure I understand the question, but in simple terms, 4/3 is more than a sensor size change. The mount diameter and distance to sensor (registration distance) were both changed, so there's no compatibility with anything APS..

Whereas, APS cameras in most (all?) cases are designed to use legacy FF camera lenses, so there's a level of interchangeability between APS and FF lenses (that varies with manufacturer)..

Arguably, it's an advantage for 4/3 that they have "optimized" for the current sensor size. APS lenses only work with APS sensors, but suffer a bit from the restrictions placed on them by the FF mount..

FF lenses obviously work on FF sensors and also therefore on the smaller APS sensor which of course is just a crop of the same image..

So deductively, 4/3 can't go any larger, but could theoretically go SMALLER...a cropped 4/3? Not sure they're would ever be any advantage to it, unless they could put in some superfast shutter and AS mechanism that would give 1/500 sync speed and 5+ stops of AS due to smaller masses involved..

That would make for an interesting daylight telephoto cam, but it probably wouldn't reduce cost, which was the main reason for the cropped sensor in APS in the first place..

Greg..

Comment #14

Its really more of an observation than a question. I know that the Pentax DA lenses dont cover a full frame, but do cover the APS-C sensor. Since Tamron and Sigma and others make lenses specifically for digital cameras, I assume they cover the APS-C sensor. If you make a lens for a four-thirds camera like an Oly, then youre eliminating that larger Cannon, Nikon, Pentax APS-C market. I just never thought of it that way before. If I were a third party lenses designer, I dont guess Id mess with a four-thirds design..

DavidDallas, TX..

Comment #15

Well, from a pure design standpoint, there is plenty of overlap. Many of the expensive telephotos I believe come in both 35mm mount and 4/3 mount...they just replace some of the gear on the back of the lens..

Whether they are "compromised" and a "pure" 4/3 design would be "better", I don't know the answer to that question. One would think that could be the case, just due the nature of compromises..

Greg..

Comment #16

Hi,.

Well, I dunno, but does it matter that much? I decide on things by looking through the view-finder or whatever. Never ever by saying that I need a horizontal FoV of 23, thanks to zooms..

Having said that I can add that I know the 4:3 isn't quite like 35mm film and so can quote that a "28mm" on 4:3 is the height of a 24mm but the width of a 28mm. Not that that is of any use in the real world but in forums....

As for APS etc: has anyone else ever wondered why they don't use lenses from their APS cameras? At one time (up to about 5 years ago) all the makers did APS (film) cameras in P&S, compact zoom and SLR's but what happened to their lens' designs? Were they re-hashed? Just curious..

Regards, David..

Comment #17

Click Here to View All...

Sponsored Amazon Deals:

1. Get big savings on Amazon warehouse deals.
2. Save up to 70% on Amazon Products.


This question was taken from a support group/message board and re-posted here so others can learn from it.

 

Categories: Home | Diet & Weight Management | Vitamins & Supplements | Herbs & Cleansing |

Sexual Health | Medifast Support | Nutrisystem Support | Medifast Questions |

Web Hosting | Web Hosts | Website Hosting | Hosting |

Web Hosting | GoDaddy | Digital Cameras | Best WebHosts |

Web Hosting FAQ | Web Hosts FAQ | Hosting FAQ | Hosting Group |

Hosting Questions | Camera Tips | Best Cameras To Buy | Best Cameras This Year |

Camera Q-A | Digital Cameras Q-A | Camera Forum | Nov 2010 - Cameras |

Oct 2010 - Cameras | Oct 2010 - DSLRs | Oct 2010 - Camera Tips | Sep 2010 - Cameras |

Sep 2010 - DSLRS | Sep 2010 - Camera Tips | Aug 2010 - Cameras | Aug 2010 - DSLR Tips |

Aug 2010 - Camera Tips | July 2010 - Cameras | July 2010 - Nikon Cameras | July 2010 - Canon Cameras |

July 2010 - Pentax Cameras | Medifast Recipes | Medifast Recipes Tips | Medifast Recipes Strategies |

Medifast Recipes Experiences | Medifast Recipes Group | Medifast Recipes Forum | Medifast Support Strategies |

Medifast Support Experiences |

 

(C) Copyright 2010 All rights reserved.