sevo
Fokutorendaburando
No, it is a Nikon phenomenon. But as far as I can make out, third party lenses do also have the offset going from 4 2/3 to 5 stops for lenses faster than f/2 - at least on my one f/1.8 third party AI lens.
As I already said, my guess is that they did it so that F2 era cameras did not need a screen correction factor with the most popular fast lenses (the F came with a impressive table of lens/screen correction factors, and no screen - other than plain matte - that needed no corrections throughout the entire Nikon lens range). Later cameras could undo that correction wherever necessary (i.e. if their metering system was not affected by the focus aids) as AI lenses also encode the absolute aperture value on another tab inside the AI mount (which the F2 however could not read).
Mark C
Well-known
Are you saying that AI Nikkors have the same AI cut for f1.8 and f2 lenses, which is what the OP is saying. My lenses are mostly non-AI, so I can't verify it myself.
sevo
Fokutorendaburando
Yes - and f/1.4 is the proper 2/3 off f/1.8, so that change in offset is not undone as the aperture series continues.
Fraser
Well-known
Exploring the possibility of converting non AI lenses to AI I have found that f/2 and f/1.8 have the exact same AI settings.
That means essentially that Nikon disregards a 1/3 step difference in metering...
Going back to the original question, if thats the case why does the meter change when i change the aperture from 1.8 to 2?
But on the other hand if i mount a f2 lens and tell the camera its a 1.8 the aperture ring will still say f2 and not be able to reach 1.8 (only just though) which tells me f2 and 1.8 are in a different position, but there is so much play in the aperture ring on old nikkors maybe some of them could get to 1.8.
sevo
Fokutorendaburando
Because the change is a change in the reference point for a relative value. It is not that f/2 is the same as f/1.8 on the same lens, but that f/2 (and below) has a different definition of "fully open" as f/1.8 and above. Without that correction, the F2 TTL meter would give the same readout for f/1.8 on the faster and f/2 on the slower lens with identical subjects, where there ought to be a 1/3 stop difference.
sanmich
Veteran
I just realized that I don't fully understand how AI works.
If it is meant to tell the camera what is the maximum aperture, the one the metering s done with, how come it moves with the aperture ring?
The old, non AI mechanism is clear: you do bring the lens to the maximum aperture that is recorded by the camera, then the position of the aperture ring allows to meter.
But with AI, how can the camera know the difference between an f/2 lens set at f/4 (aperture for metering is f/2), and an F/4 lens??
If it is meant to tell the camera what is the maximum aperture, the one the metering s done with, how come it moves with the aperture ring?
The old, non AI mechanism is clear: you do bring the lens to the maximum aperture that is recorded by the camera, then the position of the aperture ring allows to meter.
But with AI, how can the camera know the difference between an f/2 lens set at f/4 (aperture for metering is f/2), and an F/4 lens??
sevo
Fokutorendaburando
It does not! That is, while AI tells it that one lens is a lens stopped down by two stops while the other is fully open, it does not know the speed of either - the latter will only indirectly be inferred from the amount of light that passes through the open lens! For basic AI (what e.g. the F2 DP-11 or -12 do), the camera does not need to know the lens speed - it can read the in-finder aperture display off the scale printed on the lens, while the aperture tab only gives the difference between fully open and the current position, which is all that is relevant for aperture priority AE or manual TTL metering. A f/2 lens set to f/4 has exactly the same AI ring and tab position as a f/2.8 lens set to f/5.6 (that is, "2EV down from start"), while the f/4 lens at f/4 has the same AI ring position as a f/2 lens at f/2 - 0EV. What we are talking about is that there is a small offset of the start value (by 1/3EV) where lenses faster than f/2 are concerned.
Nikon also made provisions for cameras that need to know the absolute lens speed - the tab at the bottom of the mount does encode that (even though none of the cameras around at the start of AI could read it).
radi(c)al_cam
Well-known
It's explained quite well here:
http://www.thephotoforum.com/threads/meter-coupling-lever.371863/
And, again: Minolta had it already in 1966