70-200/2.8 II Interesting Focal Length Behavior

It's kind of a well known fact that the Nikon 70-200/2.8 VR II lens is only 200mm when focused at infinity, and this lens gets a whole lot shorter when you approach the MFD.

This topic was broached a day ago, and when I checked the Canon 70-200/2.8 IS II, I was surprised to note that the lens seems to have the opposite behavior. This is kind of interesting, so here is a follow up.

I shot the lens at three distances and took a picture of a tape measure. For the example, see the MFD and 178" shots below. I used a 1D Mark IV for this, so note the 1.3X sensor.

My findings were as follows:

When I shot the tape measure at 373" distance, the image recorded a 52.3 inch wide FOV. This works out to an 8.02 degree horizontal AOV which is about 202mm focal length.

When I shot the tape measure at 178" distance, the image recorded a 23.9 inch wide FOV. This works out to a 7.69 degree horizontal AOV which is about 210mm focal length.

Finally, right near the MFD I shot the tape measure from 44.6" distance and recorded a 5 inch wide image. This works out to a 6.41 degree horizontal AOV which is about a 253mm focal length.

So the Canon 70-200/2.8 IS II looks to be a true 200mm at infinity but it actually grows longer to as much as 253mm when you focus at the MFD.

That is really unusual so far as I know, but a nice feature and much better than what the Nikon does.

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

In what circumstances would you want a wider FOV at MFD rather than a narrower FOV?

This is quite interesting, and I'm sure that it has something to do with the differences in design. Here is a great article that I found somewhere (lots of info about the zoom lenses, but the 70-200mm VR II should be near the bottom):
http://www.pierretosca​ni.com/echo_telezooms_​english.html

One of the benefits of the Nikkor design is ultra fast autofocus. I don't have experience with any 70-200mm zoom though, but I'm guessing the autofocus speed is comparable to the Canon equivalent.

frugivore wrote in post #13591068
In what circumstances would you want a wider FOV at MFD rather than a narrower FOV?

This is quite interesting, and I'm sure that it has something to do with the differences in design. Here is a great article that I found somewhere (lots of info about the zoom lenses, but the 70-200mm VR II should be near the bottom):
http://www.pierretosca​ni.com/echo_telezooms_​english.html

One of the benefits of the Nikkor design is ultra fast autofocus. I don't have experience with any 70-200mm zoom though, but I'm guessing the autofocus speed is comparable to the Canon equivalent.

^^ Faster.

I use Canon primarily because of the 70-200mm f/2.8 IS II. It's the best 70-200 out there bar none.

The Nikkor is nice though. Personally I wish the Canon 70-200 wasn't white because I can't get into any NHL Hockey games with it! They know what it is right away.
I even thought about getting the old 80-200L just because it was black. Haha

JeffreyG wrote in post #13590707
My findings were as follows:

Just to be clear, were these distances measured from the sensor plane?

btdvox wrote in post #13591567
I can't get into any NHL Hockey games with it! They know what it is right away.

That's a problem with your arena, not the lens. I've had my 100-400 to a few games and I see 70-200s in the crowd often.

krb wrote in post #13591592
Just to be clear, were these distances measured from the sensor plane?

Yes, measured to the sensor plane mark on the body.

Hi Jeffrey,

From which point did you measure this? Based on the above measurements, I get rather different values, namely 240 mm, 240 mm and 194.5 mm as FLs going from longest to shortest distance, purely based on magnification at those distances.

However, I am assuming that you are using the distance from sensor to object. This probably is not a correct assumption, because the MFD of the 70-200 II is 47.6 " (1.2 m), and you mention a distance of 44.6 " for the shortest distance used here.

So second guess is from front of lens to object, but I need to be sure here . Could you please elaborate?

Kind regards, Wim

Edit: I was thinking too long here. You already explained. However, in that case there is something wrong here, possibly with my formula implementation. I'll be back.

Other details,

I used 28.3mm for the sensor width. I'm very confident in my sensor plane to subject distances, especially at the MFD which was easy to measure.

My method of focus for MFD was to enable live view, turn the focus ring as far to MFD as possible and then move the tripod towards the subject until I achieved maximum focus. Then I twiddled the focus ring a hair to sharpen it up that last little bit (probably less than 5mm final adjustment).

If the MFD is better than Canon's published best, so be it. I trust the measurements.

JeffreyG wrote in post #13591684
If the MFD is better than Canon's published best, so be it. I trust the measurements.

According to The Digital Picture measurements MFD IS 1148mm.

sosrah wrote in post #13593152
According to The Digital Picture measurements MFD IS 1148mm.

And according to me it is 1132mm. That 16mm difference is about 1.4% difference of opinion.

So I reported the focal length of the 70-200 was 253mm at the MFD. If my measurement was bad and I was actually at 1148mm, then the focal length was actually 249mm

So 249mm vs. 253mm.....either one would not change my conclusion which is that the 70-200 is very close to 200mm at infinity focus and grows substrantially longer as you approach the MDF.

JeffreyG wrote in post #13593516
And according to me it is 1132mm. That 16mm difference is about 1.4% difference of opinion.

So I reported the focal length of the 70-200 was 253mm at the MFD. If my measurement was bad and I was actually at 1148mm, then the focal length was actually 249mm

So 249mm vs. 253mm.....either one would not change my conclusion which is that the 70-200 is very close to 200mm at infinity focus and grows substrantially longer as you approach the MDF.

Yes, i wrote my post to confirm your measures not to confute them.

Jeffery - very interesting test and results. Thanks for sharing what you did.

JeffreyG wrote in post #13591684
Other details,

I used 28.3mm for the sensor width. I'm very confident in my sensor plane to subject distances, especially at the MFD which was easy to measure.

My method of focus for MFD was to enable live view, turn the focus ring as far to MFD as possible and then move the tripod towards the subject until I achieved maximum focus. Then I twiddled the focus ring a hair to sharpen it up that last little bit (probably less than 5mm final adjustment).

If the MFD is better than Canon's published best, so be it. I trust the measurements.

Ah, in my enthusiasm, I completely overlooked the fact you used the 1D IV - makes quite a difference, and I'll have another go at it .

Kind regards, Wim

Ok, worked out the new values.

What I have done is used the magnification factors at these distances, and worked those back to actual FLs, with th eaid of th elens formula, keeping the focusing distance the same. Gong from wide away to nearby, I get the following results.

Focusing distance:
9339.6 mm - 4485.0 mm - 1123.9 mm
Expected magnification:
0.022 X - 0.049 X - 0.301 X
Measured magnification (sensor width / width of measuring tape in image):
0.021 X - 0.047 X - 0.225 X
Nominal FL:
200 mm - 200 mm - 200 mm
Calculated FL based on true magnification:
~190 mm - ~193 mm - ~169 mm

IOW, this confirms that the FL gets slightly longer first when focusing closer, but towards MFD it gets quite a bit shorter again, and ends at appproximately the same FL as the older 70-200 F/2.8L IS, Mk I (~170 mm).

Note that my calculations take the expected magnification into account, and as indicated in the other thread, the image circle gets bigger when focusing closer, and the image circle therefore is cropped more and more when focusing closer, which may seem to have the effect of an increasing FL, while in effect it may not. This because part of the image is no longer visible by the sensor.

If this really was a 253 mm lens, I am pretty sure Canon would have said it was a 70-250 lens or thereabouts. Based on the expected magnification, it is clear that this lens does show focus breathing however, but within limits.

Kind regards, Wim

Hi wimg. I find your calculations interesting, but I'm not sure why you and I get different answers.

I measured the field of view at three focal lengths. From this I calculated the angle of view via simple trigonometry (simply using the tangent of the angle being equal to the opposite leg of the triangle divided by the adjacent leg of the triangle). From there I was able to calculate the focal length.

I guess this is the biggest thing I do not understand in your calculation: I am 100% sure that I was measuring a narrower angle of view from the lens as the focus distance grew shorter. In my understanding, narrower AOV = longer focal length. The way I calculated the focal length certainly assumes this to be true.

So why does your magnification based focal length calculation suggest that an AOV of 8.02 degrees is 190mm focal length and then and AOV of 6.41 degrees is only 170mm? That makes no sense to me.

I am 100% confident that my calculations of AOV are correct, and to me focal length and AOV are directly correlated. So there is some kind of disconnect between how you and I measure focal length.

JeffreyG wrote in post #13595682
Hi wimg. I find your calculations interesting, but I'm not sure why you and I get different answers.

I measured the field of view at three focal lengths. From this I calculated the angle of view via simple trigonometry (simply using the tangent of the angle being equal to the opposite leg of the triangle divided by the adjacent leg of the triangle). From there I was able to calculate the focal length.

I guess this is the biggest thing I do not understand in your calculation: I am 100% sure that I was measuring a narrower angle of view from the lens as the focus distance grew shorter. In my understanding, narrower AOV = longer focal length. The way I calculated the focal length certainly assumes this to be true.

So why does your magnification based focal length calculation suggest that an AOV of 8.02 degrees is 190mm focal length and then and AOV of 6.41 degrees is only 170mm? That makes no sense to me.

I am 100% confident that my calculations of AOV are correct, and to me focal length and AOV are directly correlated. So there is some kind of disconnect between how you and I measure focal length.

Hi Jeffrey,

Although your calculations are correct, the assumption that AoV means a different FL is only true for infinity. This is why I asked for the tape measure photographs, because that way I would be able to calculate true magnifcation at these focusing distances.

With the focusing distance given, and true magnification known, it is then possible with the aid of the lens formula and an algebraical transformation to calculate the actual FL. There always is only 1 FL possibe for a certain magnification at a certain focusing distance (depending on accuracy of measurement available; in practice, when measurements don't go to 100th of inches, it is a very small range at longer FLs, of which I took the one closest to the magnification obtained), and this is what my (quick) calculation is based on.

Do note that you measured the AoV purely based on the image you received, which does not reflect the actual projected image circle increase, but only the diminishing size of the FoV in the image (and in the VF). As mentioned before, the closer you focus, the more you crop the projected image circle, in principle anyway. This is also why in principle the AoV should become narrower when focusing closer (just like the AoV with the same lens on APS-C is smaller than on FF; the image is cropped, or cropped more if you like).

What is very interesting however, is the relatively large decrease of AoV first, which indicates that something non-expected is going on, and indeed, the FL does increase a little first, prior to decreasing again at even closer focusing distances. This is interesting from the POV that I personally certainly never have seen this happening before in any other lens.

BTW, I deliberately provided expected magnifications at 200 mm, in order to be able to compare the values obtained in real life, and because I really did a quick calculation with th eaid fo th emagnification calculator I created a while ago. This is not 100% accirate, however.

As to decreasing AoV when focusing close by, f.e., at 1:1 the image circle is 2X the diameter of the image circle at infinity, which means that at 1:1 the image is cropped by a factor 4 (2X2), or a 25% crop, hence a drastic decrease of the AoV at the focusing distance, which at 1:1 happens to be 4X the FL. However, FL hasn't changed, in principle anyway.

HTH, warm regards, Wim