I hear people say that a lense that has, say, a max zoom of 50mm is equivalent to a film lens of like 100mm (this is only an example!) But how, and why is this?
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Something I dont understand about lenses...
- Thread starter SolracSelbor
- Start date
- Sort by reaction score
Because most digital sensors are physically smaller than a film slide/negative etc.
This means that even though the lens is a certain focal length, the camera is cropped so that it only shows a certain part of that area, meaning if you had a film camera you'd need a lens of 100m(in your example) to achieve the same "crop" or shot.
Read here.
This means that even though the lens is a certain focal length, the camera is cropped so that it only shows a certain part of that area, meaning if you had a film camera you'd need a lens of 100m(in your example) to achieve the same "crop" or shot.
Read here.
Digital cameras use different size sensors. Some of the top pro cameras (D3, 1Ds, 5D) use the same sensor size as a film camera: 24x36mm
Other cameras use smaller sensors, so there is a crop factor, which piles more pixels into a smaller area size, making the output (screen, print) look larger. Many Nikons use sensors that create a 1.5x crop factor. Canons use slightly smaller sensors which create a 1.6x crop factor.
Point and shoot cameras use even smaller sensors creating even larger crop factors. With smaller sensors comes greater sensitivity and greater noise (grain.) That is why P&S cameras suck in dim/low light. They have to jack up the ISO to get the picture.
Anyway, it varies!
Smaller sensors don't pack more pixels into a smaller area than a larger sensor unless they have more pixels (mb) in which to pack. But, certainly pixels in smaller sensors are packed more densely as long as the sensors are of equal mega-pixel designation. This divides the smaller sensor into smaller pixel sizes, resulting in a decrease in sensitivity, and less capability of handling low light easily without noise.
Just because a D50 (for example) uses an APS-C sized sensor (equivalent to a 1.5x magnification of the 35mm equivalent) doesn't mean more pixels are piled into a smaller area - it all depends on how many megapixels the sensor has in the first place, and what it's being compared to. But, I get the idea - it's the size of the individual pixels that count, and naturally bigger sensors allow for bigger pixels at the same megapixel rating. But, a D50, "crop factor" and all, probably has pixels as big as a full-frame Canon with 16.7 or 21 megapixels, and bigger pixels than a D200 (D50, same size sensor, fewer pixels, so they must be bigger.)
Slightly different subject, but related to the full-frame vs. crop sensors size issue: The one thing about the crop factor which I'm not sure of is this: If I'm using a 100mm prime lens on a 35mm SLR camera it will have a certain compression, or flattening of the image, foreground to background. Now, if I put that same lens on my D50, it will be as if it were a 150mm prime lens, so I'm told. The sensor is smaller, and "crops" the image circle cast upon the sensor in the middle, sort of like zooming in to the image. Yet the physical image cast by the lens doesn't change, it's the same image through a 100mm lens, just less of it hits the sensor. In reality, it's not like seeing through a 150mm lens, because the front-to-back compression is still that of a 100mm lens.
The reason I noticed this is that I originally believed that with my 55-200 lens on my D50 that I'd get the "reach" of a 300mm lens (35mm equivalent) on the long end, but what I get is the reach of a 200mm lens, because that's what it is. Cropping into any image doesn't really "zoom" it to a different focal length, whether cropping from a print or a negative or a digital image. Would my DX lens at 200mm become a 300mm lens on my 35mm full frame camera (vignetting notwithstanding?) No. It would still be a 200mm lens (made smaller and lighter so it couldn't cover the full 35mm frame completely with it's image circle.) But, 200mm is still 200mm. At least in my book.
Anyone else care to shed some light on this...?
Fair enough. Thus, we revise the statement to say:
"The optical path behind the lens has been adjusted so that all of the lenses' light falls into a smaller area than what it did with 24mm x 35mm film. It is this ratio of the new area versus (24x35) that creates the effective magnification ratio."
This way, we get rid of any mention of pixels, so the statement will be true even if we go back to film
From an absolute standpoint, a 200mm lens is still a 200mm lens. As such, elements of depth-of-field and similar compression/perception elements don't change. You're essentially correct in saying that all that the 'crop body' dSLR is automatically cropping the image for you, as it is optically configured to subtend a smaller geometric angle. Simplistically, what's being overlooked is that for a standard 200mm film lens on the digital, what's captured by the sensor is the "looks like 300mm" center, but the rest of the light is passing through the lens...its simply not hitting the recording media, so its not being recorded.
This would become much more apparent if someone were create a, say, 200mm EF-S lens and then mount it onto a full-frame camera body (with EF mount): the resulting photo would be fine in the middle, but the outside third would effectively be black (due to no exposure)...it would be something like this image.
-hh
So, basically you're confirming what I'm saying. Thanks for further explaining it, it does help the discussion.
Let's assume the following discussion refers to prime lenses that would work on both full-frame and DX Nikon (or any brand) dSLRs:
To further simplify this illustration, (and I'm thinking aloud here, perhaps trying more to cement my own understanding) let's just hook a 200mm prime lens up to a full-frame digital camera, mounted on a tripod in a stationary position, take a picture, examine the results. Now, let's consider that this same camera can also "crop" the sensor down to a smaller megapixel, DX mode, and now with the same setup other than the smaller sensor recording the image, fire away once more. Examine the results. I'm betting the image doesn't look like it was shot with a 300mm lens (1.5 magnification) but looks exactly like the middle part of the 200mm image from the larger sensor, which is what it is... same apparent telescopic effect, same detail, same resolution (presuming the sensors are equal in that area, which they probably are since it's actually the same sensor, just less of it.)
Now, let's photograph the same scene by attaching a 300mm prime lens to the camera and shooting in full-frame mode, examining the results. Next, attach the 200mm lens and switch the camera to DX mode, examine the results. Compare the images (I really wish I could do this, but I can't so I'm using my imagination here...
) I'm betting the 300mm full-frame image, while covering the same angle of view as the 200mm lens on a "cropped" (DX) sensor, will have a more compressed telescopic effect, with background objects appearing larger than on the 200mm DX combo.Which in the end makes me want my full-frame way of seeing things back. I can always crop my own images down to the DX way of seeing things. Come on, Nikon. Hurry up with that sub-$2000 FX body with optional battery/booster grip and minimum 5 fps.
So, basically you're confirming what I'm saying. Thanks for further explaining it, it does help the discussion.
Let's assume the following discussion refers to prime lenses that would work on both full-frame and DX Nikon (or any brand) dSLRs:
To further simplify this illustration, (and I'm thinking aloud here, perhaps trying more to cement my own understanding) let's just hook a 200mm prime lens up to a full-frame digital camera, mounted on a tripod in a stationary position, take a picture, examine the results. Now, let's consider that this same camera can also "crop" the sensor down to a smaller megapixel, DX mode, and now with the same setup other than the smaller sensor recording the image, fire away once more. Examine the results. I'm betting the image doesn't look like it was shot with a 300mm lens (1.5 magnification) but looks exactly like the middle part of the 200mm image from the larger sensor, which is what it is... same apparent telescopic effect, same detail, same resolution (presuming the sensors are equal in that area, which they probably are since it's actually the same sensor, just less of it.)
Now, let's photograph the same scene by attaching a 300mm prime lens to the camera and shooting in full-frame mode, examining the results. Next, attach the 200mm lens and switch the camera to DX mode, examine the results. Compare the images (I really wish I could do this, but I can't so I'm using my imagination here...
It's not the lens that changes the perspective - it's your position - the distance from your subject. If you stay in the same position and take a photo first with a wide angle, then a telephoto - the perspective will be the same for both. You should see this if you crop down the wide angle image to the same angle of view as the telephoto image. It will look different - but this will be down to a different depth of field and probably some wide angle lens distortions.
Point-of-view #1: Companies have this thing called "marketing" who like to talk about what fits into the .jpeg frame. You get this and you get that, but they rarely talk about tech issues.
Point-of-view #2: Laws of physics. A millimeter is a millimeter is a millimeter. If an objective has 100mm focal length, then it has 100mm focal length. That fact does not change whether the objective is used with a full frame camera body or an APS-C one. It has been made to be a 100mm lens and that's what it is.
Now, what about real world? You tend to get what you pay for. But what is it that you pay for? That's the question. Imagine this:
xxxxxxxxxxxxxxxxxxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxxxxxxxxxxxxxxxxxx
The 'X' image is what the 100mm lens produces for full frame camera. The 'O' image is what the crop-body camera sees. Marketing department wants to enlarge the 'O' image dimensions (millimeters) to as big as the 'X' image and say that crop-body camera sees "150mm equivalent" or something. In reality, the crop-body camera LOSES the rest of the image and IS ONLY ABLE TO CAPTURE the 'O' part of the "full frame". The reason why marketing departments have not been sued to death is the simple fact that they're only talking about "full frame of cropped body", which seems to be fine.
If the same lens is used with "full frame" and "aps-c" camera bodies, the center part of the image (the 'O' part) will produce identical picture. Only difference being that the full frame camera body can capture the rest of the image whereas cropped body loses the rest for good. And that's why there are aps-c sized lenses for sale; if the customer has APS-C sized image sensor, marketing thinks it is perfectly okay to sell a 30mm APS-C lens instead of 50mm full frame lens. Sure, the resulting image is equivalent in size, but the two images are different because one is shot with 30mm lens and another with 50mm lens.
Okay, enough ranting. I do have a cropped body (40D) but I like to shoot with real millimeters
Point-of-view #2: Laws of physics. A millimeter is a millimeter is a millimeter. If an objective has 100mm focal length, then it has 100mm focal length. That fact does not change whether the objective is used with a full frame camera body or an APS-C one. It has been made to be a 100mm lens and that's what it is.
Now, what about real world? You tend to get what you pay for. But what is it that you pay for? That's the question. Imagine this:
xxxxxxxxxxxxxxxxxxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxooooooooooooooxxx
xxxxxxxxxxxxxxxxxxxx
The 'X' image is what the 100mm lens produces for full frame camera. The 'O' image is what the crop-body camera sees. Marketing department wants to enlarge the 'O' image dimensions (millimeters) to as big as the 'X' image and say that crop-body camera sees "150mm equivalent" or something. In reality, the crop-body camera LOSES the rest of the image and IS ONLY ABLE TO CAPTURE the 'O' part of the "full frame". The reason why marketing departments have not been sued to death is the simple fact that they're only talking about "full frame of cropped body", which seems to be fine.
If the same lens is used with "full frame" and "aps-c" camera bodies, the center part of the image (the 'O' part) will produce identical picture. Only difference being that the full frame camera body can capture the rest of the image whereas cropped body loses the rest for good. And that's why there are aps-c sized lenses for sale; if the customer has APS-C sized image sensor, marketing thinks it is perfectly okay to sell a 30mm APS-C lens instead of 50mm full frame lens. Sure, the resulting image is equivalent in size, but the two images are different because one is shot with 30mm lens and another with 50mm lens.
Okay, enough ranting. I do have a cropped body (40D) but I like to shoot with real millimeters
JFreak - great illustration with the Xs and Os! I agree with you completely. That's why I have found my 55-200mm DX lens doesn't give me what I used to get with my old 100-300mm for film. It takes nice, sharp, clear pictures, but is slower at each focal length, and doesn't give me nearly the telephoto reach. At first I was ignorant and just assumed if I multiplied by 1.5 I'd have my effective 82.5-300mm range, but it just never felt right, and was lacking something... until I thought about it.
JFreak, that is a fantastic description.
Just a pity the dummy who sold me my 400d didn't understand the crop factor, they told it that it makes the lens bigger (I was stupid and just accepted it, not really thinking about it) when logic says that it's utter bollocks that is makes the lens bigger.
Just a pity the dummy who sold me my 400d didn't understand the crop factor, they told it that it makes the lens bigger (I was stupid and just accepted it, not really thinking about it) when logic says that it's utter bollocks that is makes the lens bigger.
Great diagram JFreak. I tried making one in my first post with "-" and "|" but that turned out crap. Good work!
I don't quite understand what the problem is and what differences you are seeing. It's certainly true that a 200mm lens remains a 200mm lens - what ever body you put it on. Saying it is a 300mm equivalent is only really a convenient shorthand for describing the reduced angle of view on a smaller sensor body. But the sizes of the objects in the image and their relative sizes will be the same for a 300mm lens on a full frame camera and a 200mm lens on a camera with a smaller sensor (1.5x).
The major difference is depth of field. Depth of field depends on focal length and subject distance. If you take a photo with a 200mm lens on a small sensor camera, then take a photo at the same distance (and same aperture) with a full frame camera using a 300mm lens to get the same field of view, the depth of field will be greater on the 200mm/small sensor image. This could be either an advantage or disadvantage, depending on what you're trying to achieve.
The fact that the small sensor only sees the central part of the image can be thought of as a good thing. This is where the lens will be sharpest.
Exactly the problem; marketing talks about *relative* things but one should first and foremost think of *absolute* terms which is how the image is taken anyway.
You can change the *relative* size of your image later on, but the image will always be what you have *absolutely* captured.
So, think absolutely and post-process relatively.
You can change the *relative* size of your image later on, but the image will always be what you have *absolutely* captured.
So, think absolutely and post-process relatively.
This is one of my big gripes with the tiny-sensor EVF cameras like my Canon S2IS. Even with the widest aperture (f/2.7) you cant' get a decent portrait because there's too much depth of field.
By your logic, if I took an image from a 50mm lens and cropped it to the equivalent crop of a 6x lens, right in the middle, I'd get the same image, more or less, that I'd get from a 300mm full frame image. That seems like a hell of a compromise in detail, and I also doubt the front-to-rear telescopic compression would be anywhere near the same (if that makes sense...) I know that's an extreme example, but I use it just to make a point.
To me the best way to describe this is by how large objects at a distance appear at different focal lengths relative to objects in the foreground at the same angle of view. A cropped 300mm lens and a straight 200mm lens aren't the same. You're right, it is a depth of field issue, which I hadn't even thought of and that's huge by itself, but I think there's more to it than just depth of field.
There's a difference in optical "zooming" and cropping an image, or we wouldn't need different focal lengths at all, we could just do it all in post processing, as I see it.
As I understand it - that is essentially correct. Even if you used a wide angle lens you could crop it severely to get the same field of view as a telephoto and you would get the compression effect. It's down to restricting the angle of view and magnifying the image, seeming to bring everything closer together.
Obviously you wouldn't want do this very often as the resolution of the cropped image would be too small to do much with.
Just did a quick search to see if anyone agreed with me...
http://www.luminous-landscape.com/columns/composition-3.shtml
"Telephoto lenses do not compress distant scenery any more than wide-angle lenses distort scenery (except for fish-eye lenses). The perspective created by a telephoto lens is the same as that of a wide angle lens, except that only a fraction of the wide angle photograph is visible in a telephoto view. Distant objects are naturally compressed because as objects recede in the distance (get further away from us) they appear to be closer and closer together. This effect is caused by perspective, not by lenses."
http://www.luminous-landscape.com/columns/composition-3.shtml
"Telephoto lenses do not compress distant scenery any more than wide-angle lenses distort scenery (except for fish-eye lenses). The perspective created by a telephoto lens is the same as that of a wide angle lens, except that only a fraction of the wide angle photograph is visible in a telephoto view. Distant objects are naturally compressed because as objects recede in the distance (get further away from us) they appear to be closer and closer together. This effect is caused by perspective, not by lenses."