[MadMathMind]

I have a general question regarding how stop-down metering works. My camera is the K-5 IIs, if that matters at all.

My understanding is that when you use a lens newer than the A series, it works like this: you dial in the aperture electronically, but the lens is not physically moved to that aperture until you actually press the shutter. If you look at the front of the lens, you see no differences when you dial up/down the aperture. It's always wide open until the very moment you shoot. That means all metering is done with the lens wide open, but because the camera knows the aperture you will use, it computes how much less light (e.g., on a 1.4 lens, if you set it for 2.8, then incoming light will be half) and adjusts the shutter speed accordingly. In short, the aperture doesn't actually figure into the metering but only the shutter speed, which is a number the camera calculates after taking the metered value (taken wide open) and plugging it into some formula that includes the aperture.

Let me know if that's not correct.

What I don't understand is why Av setting with K/M lenses will result in incorrect metering values unless the lens is wide open. When I dial in the aperture on my K lens, I can see the blades physically move. Here it is wide open:
Attachment 199961

Here it is at f/2:
Attachment 199962

and here it is at f/8:
Attachment 199963

If you change the aperture on a modern lens and look right into the lens, you see no change whatsoever. It will always look like the first image.

What I don't understand is the necessity for stop-down metering with this sort of lens. My understanding of it is that the camera pretends to take a photo by having the diaphragm set the aperture as it would on a modern lens and then taking a measurement of the actual amount of light that comes in. But isn't that what I am doing when I use the aperture ring? When I dial in a new aperture, shouldn't the amount of light coming into the camera change immediately--after all, the hole got a lot smaller--and thus change the metering appropriately?

That's what I don't get. I thought I was setting the diaphragm when I use the aperture ring. Is that not what is happening? What exactly am I seeing happen in the above pictures?

---

Last edited by MadMathMind; 10-09-2014 at 09:44 PM.

[elliott]

With all variations of K mount lenses the camera holds the aperture wide open with the aperture lever to allow for easier focusing and composing. This is why you don't see the viewfinder getting darker or brighter as you turn the aperture ring with the lens mounted. When you press the green button or use the DOF preview it releases this lever, which allows it to get a light reading with the aperture where you want it. With the lens unmounted, like in your pictures, nothing is holding the aperture lever.

If you were using an adapted M42 lens Av mode would work exactly as you describe, since the camera has no control at all over the aperture.

[jatrax]

With the lens unmounted, like in your pictures, nothing is holding the aperture lever.

This is the key, do your test again this time with the lens mounted on the camera. The blades will not move.

[MadMathMind]

With all variations of K mount lenses the camera holds the aperture wide open with the aperture lever to allow for easier focusing and composing. This is why you don't see the viewfinder getting darker or brighter as you turn the aperture ring with the lens mounted.

This is the key, do your test again this time with the lens mounted on the camera. The blades will not move.

Yep, it's clear now. Thanks guys! I never saw what happens when the lens is mounted because I'm usually on the other side of the camera when it's mounted!

I never thought about the viewfinder dimming with the aperture. It's pretty obvious now that you mention that. I just did another test with a f/2.8-4.0 zoom lens. I set it for the minimum 2.8 at the wide end and then zoomed in; the camera naturally adjusted to f/4.0. I didn't see any decrease in brightness as I did so. I realize that it's only one stop different, but that should be somewhat noticeable, no? With zoom lenses of variable f, am I always seeing 2.8 light, and then the aperture gets moved to the appropriate value for the focal length when I press the shutter? Or is there a physical adjustment automatically from 2.8 to 4.0 as I zoom in?

[stevebrot]

This is the key, do your test again this time with the lens mounted on the camera. The blades will not move.

In fact, they won't even stop down at exposure time when shooting in other than M mode on the K-5 IIs.


Steve

[stevebrot]

I didn't see any decrease in brightness as I did so.

You won't see it with the stock viewfinder until the aperture gets narrower than about f/4.


Steve

[elliott]

I never thought about the viewfinder dimming with the aperture. It's pretty obvious now that you mention that. I just did another test with a f/2.8-4.0 zoom lens. I set it for the minimum 2.8 at the wide end and then zoomed in; the camera naturally adjusted to f/4.0. I didn't see any decrease in brightness as I did so. I realize that it's only one stop different, but that should be somewhat noticeable, no? With zoom lenses of variable f, am I always seeing 2.8 light, and then the aperture gets moved to the appropriate value for the focal length when I press the shutter? Or is there a physical adjustment automatically from 2.8 to 4.0 as I zoom in?

Since the f-stop is just the focal length divided by the diameter of the opening, the variable aperture a function of that. If the focal length gets longer, but the diameter of the wide open aperture stays the same then the f-stop will go up. This doesn't always relate to a physical diameter of an opening, just the effective diameter, which is how constant aperture zooms are possible.

With old variable aperture zooms you just have to know the aperture is really f4 even if the aperture ring is set to f2.8, just offset it in your mind. More modern ones will tell the camera electrically when the aperture changes, which is just a matter of shorting a different set of pins in the mount.