Shooting high ISO vs underexposing and lifting in post question

Started 3 months ago | Discussions thread
Slaginfected Contributing Member • Posts: 745
Re: Shooting high ISO vs underexposing and lifting in post question

bobn2 wrote:

Slaginfected wrote:

bobn2 wrote:

Poor advice, based on a prioritising a faulty theory over reality.

Well, if you are so knowledgable, show me something I can use in actual reality which makes away with a good number of artifacts.

Are you issuing this as a challenge, or do you actually want some advice? If the latter, starting your post like that is not the best way to persuade someone to give you help for free.

Life lesson: Don't ever tell anyone how their reality is supposed to look like.

I had to endure this for years, not only in these forums, and despite evidence being available left and right that my perceived reality is actually shared by many others. If you start asking question why camera makers for example do X and not Y ("Why do they give different cameras different ISO ratings?" as one example) are strong indications that the reality I'm being told is right and my actual reality are somewhat in conflict.

By now my patience with people telling me how my reality is supposed to look like is zero. Means they will get that shoved back into their face, plus they have to put proof to the pudding.

For example: Why are files at higher ISOs from the A7s line of cameras much easier to process than other cameras?

Are they? Do you have any hard evidence to support this? If you can provide some good evidence that they are 'much easier to process', and identify ways in which they are 'easier to process' then it might be possible to provide an answer.

I did like 10 years of low light photography and tested a good few cameras, including high res ones. So far this is, unfortunately, the case. I would prefer to have more resolution, but ... compromises, as you mention further below.

The funny thing is, people dump me links to DxOmark DR or PDR diagrams and then tell me "there, proof!" Not too long ago I asked whether f(x) which produces these diagrams relates directly to g(x) which actually produces the pictures we view on screens, paper, whatever. g(x) contains strongly non-linear components plus a few other peculiarities, which make me question that one can simply derive g(x) from f(x). If you can do that, go ahead, because it wouldn't just resolve a lot of useless discussions, but has also the potential to be a major step in image processing. Makes me wonder why camera makers haven't done so already, though. It is one of those pesky questions ...

I would like to have higher resolution without the downsides.

That is not what life is like. Everything is a compromise. You shift the balance of the compromise, you get different downsides and upsides. The downsides of high resolution are larger image sides, and the need to downsample if you want image resolutions smaller than the native. That's not much of a downside, given that almost every real-world situation where you use an image from a camera you'll be resampling.

Preferably for Linux and a usable UI. Show me where this reality is.

What if it isn't a 'reality'?

Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow