Image size difference at same focal length?

Why is an image on a sensor significantly larger or smaller on two different lenses at the same focal length? I know I've heard this explained at some point in the past, but I can't remember the specifics.

To clarify, for example, an image at 100mm from a 100mm Macro lens at 4 feet is significantly larger on the sensor than the same image from a 24mm-105mm at 100mm or even 105mm at the same distance.

I've never noticed it before, but recently while swapping lenses on a tripod mounted 5D MKII body viewing a static subject, the size difference was dramatic.

Could it have anything to do with the difference in maximum aperture between the two lenses? Thanks.

There are several reasons for the difference:
1) The nominal focal length is just that - at best only an approximation of the real focal length.
2) Conventionally, the focal length is measured at infinity. As you focus closer, the actual focal length of the lens changes. For example, my EF 50mm f/1.4 and EF 85mm f/1.8 get longer as you focus closer, whereas all my zoom lenses get shorter as they focus closer. A lens' change in focal length depends on its optical design.
3) Barrel and pincushion distortion can provide marginally more field of view at the edges of the frame. This is why a 15mm Fisheye sees much more in the corners than a 14mm rectilinear lens.

acornsarebitter wrote in post #11886540
There are several reasons for the difference:
1) The nominal focal length is just that - at best only an approximation of the real focal length.
2) Conventionally, the focal length is measured at infinity. As you focus closer, the actual focal length of the lens changes. For example, my EF 50mm f/1.4 and EF 85mm f/1.8 get longer as you focus closer, whereas all my zoom lenses get shorter as they focus closer. A lens' change in focal length depends on its optical design.
3) Barrel and pincushion distortion can provide marginally more field of view at the edges of the frame. This is why a 15mm Fisheye sees much more in the corners than a 14mm rectilinear lens.

very interesting. Thank you. Am I correct in assuming then that no two lenses, primes or zooms that share the same focal length in their range would every produce the same exact size image?

Thanks, I did not know about #2. I was thinking #1 but the OP said "significant difference" and it's
hard to believe that the lens manufacturer would be that far off. How big of a difference does #2 make?

Pinto wrote in post #11886596
very interesting. Thank you. Am I correct in assuming then that no two lenses, primes or zooms that share the same focal length in their range would every produce the same exact size image?

Not necessarily. They could. There are different ways to design a given focal length, and many more ways still to zoom around that. Two lenses of similar design could have the same focal length at a given distance, or even two lenses of very different design might too, by coincidence or design.

Also, consider this: the Nikon 70-200mm f/2.8 VR II lens shows markedly different "focus breathing" than the Canon 70-200mm f/2.8L IS II, but of course there is some setting on each that will produce near-matching field of view - they continuously zoom through essentially the same range, after all.

It's not just zooms versus primes either, and it's not true that all primes get longer with closer focus (macro lenses are a counterexample) or necessarily true that all zooms get shorter with closer focus.

Just one more example: I once tested my Canon 24-105mm f/4L IS and Canon 100-400mm f/4.5-5.6L IS on my 5D Mark II at near infinity, and found there is a gap between their focal length ranges - that is, the 24-105 is shorter at a nominal 105mm than the 100-400 is at 100mm.

TripleG wrote in post #11886616
Thanks, I did not know about #2. I was thinking #1 but the OP said "significant difference" and it's
hard to believe that the lens manufacturer would be that far off. How big of a difference does #2 make?

According to reports on the Nikkor 70-200mm f/2.8 VR II, it can make a massive difference. This guy claims that lens is 134mm at MFD at a nominal 200mm (and provides the theory to back it up):

http://www.bythom.com/​nikkor-70-200-VR-II-lens.htm

acornsarebitter wrote in post #11886656
Not necessarily. They could. There are different ways to design a given focal length, and many more ways still to zoom around that. Two lenses of similar design could have the same focal length at a given distance, or even two lenses of very different design might too, by coincidence or design.

Also, consider this: the Nikon 70-200mm f/2.8 VR II lens shows markedly different "focus breathing" than the Canon 70-200mm f/2.8L IS II, but of course there is some setting on each that will produce near-matching field of view - they continuously zoom through essentially the same range, after all.

It's not just zooms versus primes either, and it's not true that all primes get longer with closer focus (macro lenses are a counterexample) or necessarily true that all zooms get shorter with closer focus.

Just one more example: I once tested my Canon 24-105mm f/4L IS and Canon 100-400mm f/4.5-5.6L IS on my 5D Mark II at near infinity, and found there is a gap between their focal length ranges - that is, the 24-105 is shorter at a nominal 105mm than the 100-400 is at 100mm.

Very interesting. Thank you.