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

Have you seen actual colors or you just have a gut feeling?

I totally agree with you.

When photographing action, especially in changing light, I set max ISO to the max value in the native ISO range. I don't see any point in setting it lower than that.

I set aperture priority mode and then the largest aperture (smallest f-number) that will give my DOF. I then set the minimum shutter speed setting to be the minimum shutter speed that will meet my motion blur constraints and then Auto ISO. In this configuration, if at base ISO the camera wants to set a slower shutter speed than my minimum, the camera will set the minimum shutter speed and raise ISO to suit. If at base ISO the camera determines there will be highlight clipping at the minimum shutter speed, then the camera maintains base ISO and sets a faster shutter speed to avoid highlight clipping.

I am thinking of the scenario where someone for whatever reason chose to first set ISO at 1600 and then realised the image was too dark. They then set ISO to 6400 and the image lightness is now ok but more of the noise will now be visible.

I'm also a fan of keeping the ISO as low as possible. Smaller sensors don't like high ISO's, nor are they keen on big exposure/shadow pushes. I shoot m4/3, so consider fast primes to be essential for any low light work that has any type of movement in them. The 1.4 primes are the sweet spot for me, equivalent to 2.8 in 135 format, can still get reasonable DOF, and combined with a trip through DXO pure raw can get quite good results, without the weight associated with some of the larger format gear. Some of my favourite tiny primes are the Panasonic Leica 9mm f1.7, Laowa 10mm f2, Panasonic Leica 15mm f1.7 & 25mm f1.4, Siggy 30 & 56 f1.4's, the Oly 45 f1.8 & Lumix 42.5 f1.7. The Lumix is great for bodies without IBIS.
I figure that if Pro's can shoot with 2.8 glass, then I should be able to manage with f1.4.

Edit -I have no hesitation in cranking the ISO a bit to prevent bad motion blur by the way.

If shooting raw, leaving the exposure the same and setting the ISO to a higher value won't affect image brightness when both images are developed with the same user intent, unless you are talking about irrecoverable blown highlights at the higher ISO setting.

Of course, at the same exposure, boosting ISO won't prevent motion blur. I think what you mean is you have no hesitation in shortening the exposure to prevent bad motion blur.

Yeah it does, as I mostly shoot Aperture mode, so that's the only variable that changes/can change (shutter speed) Which minimises motion blur.

You can push in the camera, or you can push in the raw developer. If the black point calibration in the raw developer is accurate, and you don't push more than four or five stops, there is little difference except you'll have better control of the highlights pushing in post.

The size of the sensor is not a variable here, since you can crop a large sensor to get the same effective sensor size as a smaller one.

So you're using ISO to change exposure. In the situations where I know the aperture I want and the minimum shutter speed I want, I use M mode.

Which is why I wrote I prefer to use larger aperture primes on a smaller sensor. More light collected, at the expense of less DOF. Without getting into the whole equivalence roundabout/argument.

If it's a relatively static scene, then yeah. Go for your life in M mode. But when there's a bit of action going on, with not enough time to be fiddling around setting 2 or 3 different variables, then A mode for me.

I also used to fiddle ISO in aperture priority until it gave me the shutter speed I wanted.

But then one day I realised I was wasting time fiddling ISO settings until I got the desired shutter speed.

I thought since I lnow the shutter speed and aperture I want, why not lock them both in in either manual mode or as a minimum shutter speed in aperture priority and then set ISO to Auto.

The Auto ISO ended up setting ISO to the same value I would have after fiddling with it anyway.

This saved me time by not having to fiddle with ISO to get the desired shutter speed.

just continue what your doing , there is no magic sauce, no one is posting any images to prove one method is better than the other.

This might be me only for my SOOC JPG shooting.

If I shoot RAW, I might do my setting with the extra headroom of RAW over JPG in mind. Not to discussed here since every camera might vary.

In case of my cameras, if I could get maximum exposure (by shutter speed and aperture), they can produce clean enough SOOC JPG (to my acceptable level) at ISO1600 and ISO3200 respectively. Exceed the thresholds I must be very careful on setting otherwise the output could be very grainy.

For such reason I set upper limit ISO to 1600 & 3200 on respective cameras and am happy to give the freedom for my cameras to play around within the limit but still can produce usable output.

I shoot in A principally. M only for difficult lighting conditions.

When reaching the upper ISO limit, my cameras will turn to use slower shutter speed. It is a fact that I must have to accept if I would expect to have a cleaner output. At that time depend on the shutter speed I might opt for certain changes: e.g. use a wider aperture (if possible) to relax on DoF control, or if not, look for extra stability. If still not, hold camera tighter or take multiple shot hoping I could get around handshaking problem. To the last, use higher ISO manually, more careful on exposure setting (relax the original planned margins on shadow, highlight etc).

IMHO those limits are merely safty measures that help us to work within the optimal limit of our tool. When we hit the wall, we know we have to pay more attention or made changes.

Upper ISO limit is very useful to me.

Deleted. I misread the post I was replying to.

No, I don't just doggedly stick to one particular method, I'm open minded enough to experiment & try different techniques. There's so many variables that can make a difference -such as model/age of camera within a particular format, sensors can be different even at the same MP count, the raw converter used, and even the age or updates applied to that raw converter software. I've seen some absolutely horrible results from LR when used on the early 20MP 4/3 sensor cameras, whereas now I think they're not too bad. No first hand experience on the latest LR -I gave Adobe the boot when they moved to the subscription model. I'm still tinkering with different techniques, nearly 8 years after jumping into 20MP 4/3 sensors. The goalposts seem to be continually moving.

That's what I wrote. I was referring to M mode in that comment.

Yes, I realised that after I posted my reply and have since deleted it.