What Is ISO?

I find your response to @BillFerris unnecessarily aggressive :(.

I talked to Richard Butler on an unrelated topic, and he mentioned that all articles he has written belong to Amazon. In the case of copyright, it is always best to err on the side of prudence.

He made assumptions in his post without checking the facts of the matter.

Happy Easter :-)

The banner on the DPRevived.com landing page reads, "The community-run digital photography resource." This is what that looks like. I acted as a member to openly say that the copyrighted content (the PDFs) uploaded with the original post should be removed. There was nothing wrong with your comments or the links to the article. However, it is a fact that the article is copyrighted content. None of us on this site - not you, nor me, nor anybody who isn't the copyright holder - has the right to make a copy of the article and distribute that copy to others.

That's how copyright works. I don't say that as a "wannabe internet lawyer." I say that as a 30-year television professional who deals with copyright issues on a regular basis. I respect the rights of copyright holders. I suspect you do as well. If asked, I'd wager the vast majority of folks on this site would say they believe copyright should be respected. We're all photographers on this site. If there's a community of people who respect a person's or organization's copyright, it's us. If we don't, how can we expect others to respect the copyright each of us holds on the original works we create?

If this site is going to be "community-run," we all have a responsibility to embrace the possibility that members will occasionally respond to a post to bring order to some amount of disorder. That's all this was...a member trying to bring a bit of order to a thread. Tomorrow, it may be you who sees a need to respond; to openly act to bring order to disorder. That's OK. That's as it should be. This collection of 2,600+ registered members can become a community if we all are willing to embrace a bit of the responsibility for making this place a community. There's no better place to begin in my opinion than by respecting the rights of others, especially other creatives and their original works.

On the T&Cs you linked to it clearly states:

"This Web Site and any of its contents may not be copied, translated or distributed in any manner (electronic, web or printed) without our prior written consent. To inquire with the Web Site for such consent, please contact us."

Before making your original post why not tell everyone if you established that consent had not been given some time in the past.

In any case, I amended my op to say the information has now been posted elsewhere. If you ever become sufficiently qualified to give me legal advice then come back to me and I will consider taking you seriously.

I'll leave it at that. Happy Easter :-)

I think everybody should have learned by now that it is not okay to download anything, image or text, from the net and post it elsewhere, even when citing the ownership. That also applies to screenshots as strange as this may sound. The correct way is linking to the content. Exceptions are small citations of the content which is allowed in many countries as well as citing for educational purposes.

Every upload here should show a checkbox where the uploader agrees to being the owner of this content. If my memory does not deceive me, this is not the case so far.

I've been calm and reasoned in my comments. If this is going to be a "community run" site, then the admins and members should embrace the occasional comment to correct an issue that's come up. If members aren't welcome to do that, the whole "community run" thing needs to be reexamined.

That's not necessarily a bad thing. There are multiple ways an open discussion forum can be run. Inviting members to self-regulate so that mods/admins only step in very rarely, is an approach that can work but only if everybody buys into it. It takes time and effort to build that kind of community culture. Not all groups thrive in that environment. If folks aren't going to buy into members self-regulating, I would encourage you and the other admins to discuss other options for governance. Whatever that governance model ends up being, it will need to be sold to the community by the admins and as many members as can be recruited to model the desired behavior.

I think a lot of confusion is because ISO "implementation" includes design decisions both rooted and not rooted in ISO.
- some design decisions are there to get cleaner data, which constitutes a normal electronics design effort and shall be in place regardless of ISO (such design decisions may include selecting charge conversion gain, switching capacitors, pre-ADC amplifier),
- some design decisions are to provide other desired features, like latitude, taking advantage of low noise (like programming the amplifier to provide lower amplification to prevent highlight clipping and use a boosting curve in digital domain), also has nothing to do with ISO per se, but may look as if one set ISO to 800, while the camera used 200,
- some specific decisions aiding effective (cost/complexity) conversion of raw data to full-colour bitmap according to ISO recommendations (digital multiplication, for example).

With digital cameras, ISO is somewhat of an afterthought. People designing raw data chains are thinking of getting acceptable noise and setting the operational point ("midtone calibration in raw") according to what noise they deem acceptable, sometimes down to 6% instead of "expected" 18%, in other words, not being guided by ISO, and leave it to firmware / raw converters to deal with ISO ratings.

This is how Adobe are describing the challenge in their DNG specifications:

BaselineExposure
Tag 50730 (C62A.H)
Type SRATIONAL
Count 1
Value See below
Default 0.0
Usage IFD 0
Description
Camera models vary in the tradeoff they make between highlight headroom and shadow noise. Some leave a significant amount of highlight headroom during a normal exposure. This allows significant negative exposure compensation to be applied during raw conversion, but also means normal exposures will contain more shadow noise. Other models leave less headroom during normal exposures. This allows for less negative exposure compensation but results in lower shadow noise for normal exposures.
Because of these differences, a raw converter needs to vary the zero point of its exposure compensation control from model to model. BaselineExposure specifies by how much (in EV units) to move the zero point. Positive values result in brighter default results, while negative values result in darker default results.

Zero point above is close to 12.7%, half a stop below the "expected" 18%.

Maybe worth mentioning that:
- ISO standard for digital (12232) is about the signal from a digital camera, that is it deals with data processed into a fully rendered full-colour image (bitmap/TIFF/JPEG/HEIC/etc); not about the signal from a sensor
- ISO 12232 isn't applicable to raw
- ISO setting appeared on cameras before ISO 12232 was introduced
- ISO 12232 "was designed to harmonize with earlier standards developed for film-based photography", the goal that these days isn't strictly necessary and may result in rather negative consequences if misapplied / misunderstood
- ISO 12232 advises that "there are differences between electronic and film-based imaging systems that preclude exact equivalency" when it comes to the subject of the standard
- ISO 12232 explicitly refers to both "analog or [/and] digital gain" as the ways to "implement" ISO and thus nether "ISO is just a pre-ADC gain" nor "ISO is just a digital gain" are correct
- amendment to ISO 12232 (Amd.1:2020) tells us: "[ISO speed ratings] assume processing of the sensor signal will be applied, either in‐camera or after the fact, to achieve the desired tone reproduction", thus, saying that ISO rating can be established / adjusted in a raw converter or in an image editor.

Getting back on topic are we :)
To me what you wrote means that the text should clearly make the distinction between the standard [only the one for digital IMO] and the implementation [and the freedom manufacturers have in that]. Most on the latter and forget what the letters ISO stand for.

Is there a simple term that we can use to correctly describe modifications caused by varying ISO settings (disregarding dual conversion gain)?

The standard hints at "electronic or [/ and] digital gain" "to reach proper / desired tone reproduction".

The industry had to come up with something that made sense to allow for a smooth conceptual migration from film. This was about two decades ago and can be safely discarded at this point.
There is already a way to specify a tone curve in RAW files, which along with the white balance information provides for an unambigous way for the camera platform to specify the perceptual intent alongside the RAW data, without any recourse to ISO, and it is this mechanism that should be standardized instead.

As I understand it, ISO refers to any and all manner of brightening (be it pre or post ADC) being applied to whatever base sensor exposure in order to arrive at the final jpeg output brightness. And though an ISO setting may or may not ultimately affect the raw file, ISO values aren’t directly applicable to RAW files at all. Someone please correct me if I’m wrong on this..

Hi Erik,
That's how I understand it too.

"Brightening" an exposure (meaning, the result of the exposure) is IMHO still a problematic term (for example, only things that have brightness to start with can be brightened or darkened), but all manner of processing to reach proper tone reproduction, at a first glance, is what they mean.

JPEG, PNG, TIFF, HEIC, whatever the camera outputs as a fully rendered full-colour image is what the standard is about.

I think of the "ISO" setting on a camera as a control that tells the metering system to give less exposure, compared to that given at the "base" setting (typically 100).

In practice it is commonly used to allow faster shutter times, to help with hand-holding.

One can then lighten the image in post. Obviously this works best on raw files.

Don

Thanks, I’m not happy with ISO “brightening” either, but it tends to get the point across and does describe the end result - a brighter image. Proper tone reproduction, to my way of thinking anyway, could also include color and contrast processing independent of ISO. It’s a very difficult subject to converse about, lots of problematic language to contend with.

To clarify, in most cases the ISO values do affect the raw values (clipping, noise if not ISO-invariant).