What is the point of setting an upper limit in Auto-ISO?

"What is the point of setting an upper limit in Auto-ISO?"
In my tests, I found that it prevented the ISO from automatically going higher than I specified.

Yes, indeed it prevents ISO from rising above the set limit. The question is why would you do that.
In A mode it forces shutter speed below the minimum shutter speed that you set.
In M Mode it darkens the EVF.
Others have posted why they still prefer to set an upper limit.

You are again making things up.

I never said Auto iso does not cause highlight clipping. I said on my 90D I have never seen highlights clipped due to Auto ISO when the camera set ISO above base.

I normally use evaluative/matrix metering.

In my experience the 90D is over cautious when setting ISO above base with Auto iso to protect highlights after I set the exposure*.

Are you saying that 90D meters differently with Auto-ISO on and off? I have not noticed that on my R5 or other cameras. In my experience, the only way to force metering to protect the highlights somewhat was to use highlight-weighted metering, and even that does not always work.

I just posted my experience with my 90D.

The Auto iso is over cautious when setting iso above base when using evaluative/matrix metering.

So with the camera in manual mode and automatic ISO giving you iso 100, if you stop down one stop, are you saying the ISO doesn’t go to 200?

No, of course not.

But as I posted earlier, I set the optimal exposure** first and then let Auto ISO set iso where it likes as long as highlights are not clipped.

I would have no reason to stop down because the exposure* would no longer be optimal**.

Exactly. It’s simply an option for trade offs. Sure you’ve set the min desired shutter speed. But setting an auto ISO limit expresses that you’re willing to “give” on shutter (over your ideal minimum) if ISO gets too high for you. That said, I never use auto ISO.

So the ISO behavior with light level is linear? Then how is " Auto iso is over cautious when setting iso above base"?

Your last sentence above is not relevant to the issue " The Auto iso is over cautious when setting iso above base when using evaluative/matrix metering."

The op asked what is the point in setting an upper limit for ISO in Auto ISO.

I simply posted why I set Max iso to the highest native ISO value.

I know. I'm asking about something else you said: "The Auto iso is over cautious when setting iso above base when using evaluative/matrix metering."

Because when setting ISO above base it very often leaves 1/2 stop, if not more, highlight headroom according to the camera's histogram.

I rarely see auto ISO set an ISO above base resulting in highlights being close to clipping.

As noted, I merely observed the result of a test - anyone is free to think up their own reasons for setting limits.

Different behavior when setting iso above base than at base would indicate that when light is halved in M mode, iso wouldn’t double from base iso. Yet when I suggested that, you said “of course not.”

You were talking about stopping down the aperture by 1 stop and I made it clear I was talking about the situation where the aperture and shutter speed are constant after I set the optimal exposure**.

I am just describing my experience with Auto ISO on my 90D after I have set the optimal exposure**. If you do not believe what I observe that is fine 🙂

* exposure - amount of light that struck the sensor per unit area while the shutter was open
** optimal exposure - the maximum exposure* within dof and motion blur requirements without clipping important highlights.
*** under exposed - more exposure* could have been added with the DOF and blur constraints still being met without clipping important highlights.

Is the resulting OOC JPEG brightness/lightness any different than manual ISO in Av, Tv, or P modes, or do you feel that there is some special behavior regarding brightness/lightness for auto-ISO in M mode? I always assume that it is same, because "EC" just targets brightness/lightness regardless of which parameters are automated.

I shoot only raw and not raw + jpeg but I assume the sooc jpegs would look very similar to what I see on the camera's lcd screen.

I doubt very much that the ISO value the Auto ISO sets after I set aperture and shutter speed to suit dof and blur would be very different to what I would have set manually to zero/centre the camara's meter after having set the same aperture and shutter speed for the scene.

The only thing I would do differently if setting ISO manually is that if I had highlight headroom after having set the optimal exposure** then I might raise ISO to push the histogram data closer to the right but without clipping highlights.

* exposure - amount of light that struck the sensor per unit area while the shutter was open
** optimal exposure - the maximum exposure* within dof and motion blur requirements without clipping important highlights.
*** under exposed - more exposure* could have been added with the DOF and blur constraints still being met without clipping important highlights.

It's not a question of not believing you. It's a question of my understanding what you are saying your camera does.

You said: "The Auto iso is over cautious when setting iso above base when using evaluative/matrix metering."

I don't have your camera, so I can't test it.

I think it would be great if auto-ISO left more headroom at higher ISOs than it does at base ISO, but I've never seen a camera work that way, and I don't know what the camera would do with the JPEG preview images in that case.