Algebra

Algebra

I always include the Inverse Square Law in the lighting workshops that I run – I think physics is important, although some politicians know that Latin and Greek matter more (they’re wrong: gravity goes downwards even if you tell it not to in an ancient language. But to lose sight of the past is to take risks with history). And today, courtesy of Altimages, a little bit of algebra.

I blogged about a downside of film – he posted a link to an article about ISO-invariant sensors. Now, this is interesting, in terms of one or two specific situations when you might want to make use of what’s in the article, and also for those who really want to understand how their cameras work (I mean REALLY understand, in depth and detail). For most of us, it raises only one or two things to be careful about.

The gist of the article is that some sensors give very much the same results regardless of the ISO setting and exposure, and the author explains why and how. The statement is true within parameters, and there are considerations that affect how you might apply it in your pictures. And it’s also made clear that it may have no practical effect on what you are doing… Indeed, right at the start, the article suggests that many readers may want to skip to the last section, which gives practical advice for – frankly – almost all of us.

The article hinges on the fact that all digital cameras do things that amplify the electrical signal that light striking the sensor produces, and they do it at different stages between photons hitting the sensor and recording digital code on your memory card. Problems can arise at each stage, and some are inherent in the system: manufacturers work hard to design them out, and their success leads to ISO-invariance.

And the algebra? I’m sure there’s a lot involved in the design work, but the crucial issue here is that it matters when (at which stage in the proceedings) you do the multiplication. In most equations, it’s fine – the order of doing things doesn’t matter, nor does the order of writing them down. X times Y is the same as Y times X, and it’s purely a matter of convenience which way round you do things – spotting that’s one of the tricks of being good at manipulating equations.

But, in some circumstances, the order of operations is crucial (memory says). Sometimes, putting the cart before the horse is fatal to a solution, rather than merely inconvenient (have you ever seen anyone trying to get a horse to push a cart? Photographs, please!) And it’s the same with ISO. The camera amplifies the signal digitally, later in the process, and there’s an analogue amplification process earlier on. If changing the ISO setting depends on the analogue stage, the world is unchanged: if it’s done at the later, digital stage, you might as well do it in editing.

There are practical consequences that mean you probably need to know which ISO settings are ‘native’ and which are extended – it affects dynamic range. And none of it matters if you only shoot JPG images. For the rest, go and read the article. It’s repetitive in places, but that’s good with a complex subject, because you keep forgetting things. Repetition gets it into your neurons…