Fuji X10 HDR & Fast photo

Started Nov 18, 2013 | Discussions thread
(unknown member) Contributing Member • Posts: 500
Re: Fuji X10 HDR & Fast photo

Trevor G wrote:

cantanima wrote:

Absolutely nothing in your email addressed the question of exposure bracketing +/-1 with three shots. Perhaps you should reread what I'm doing, and pause to think about it before mashing the "reply" button.

Email? Mashing? You have lost your cool, I think.

Au contraire, but I'm correct. You honestly don't seem to be following what I'm saying, and keep attacking a straw man.

In any case exposure bracketing is not needed if you shoot and process RAW under the terms I have already laid out.

Yes, it is.

ON a very high contrast scene like the Albury Railway Station this still retains all highlights, even though the metering would tend to lift the exposure higher than it should be.

If you expose one image only to retain highlights, how would you retain the shadows?

EXR takes 2 simultaneous images, one at the rated speed (which will overexpose) and one at 1/2 the time for DR200 (= -1EV) and one at 1/4 the nominal time for DR400 (= -2EV). Those second frames hold all the highlights, while the first frame holds all the lowlights.

In extremely high-contrast situations, they hold neither all the highlights nor all the lowlights. There are situations where dynamic range is higher even than your train station. When the well capacity of a photosite is full, it is full -- period. The highlight will be blown even in the raw file, and in some cases you cannot get around that, even with 2EV difference.

You yourself argued here not two weeks ago that you couldn't recover shadows with the X10. (You'll recall that I didn't take a position, but I didn't buy your argument.)

Misquote, incomplete quote. I'm not "arguing", I'm stating a case.

Perhaps you should read the primary definition of the word "argue".

For the sake of argument, let's suppose you were correct. The explanation is simple: the dark exposure retains highlights, but the bright exposure was not bright enough to capture shadows. One shot in DR 400 gives only two exposures, neither of which (by your claims) contains enough information in the shadows to recover them correctly.

That wasn't my claim at all - there is plenty of information in the shadows, but Silkypix/RFC (and similar programmes) cannot lift it out accurately and retain colour fidelity while also retaining the highlight details.

You claimed the X10's sensor failed to retain shadows as well as the X20's, because of a higher noise floor within the shadows. Here is what you wrote:

  • As explained above, I am not trying to show people how to PP an image, but to accurately find out how they both respond to shadow lifting, to see where their individual noise floors lie.

I suggested then that different processing than you were using could get around that:

  • So, depending on how they were combined... that's software processing, but what I really want to know is how the software combines the two images.

Now, as I said above: for the sake of argument, let's suppose your claim above is correct. Getting around this "noise floor" is what I'm talking about. With a bracketed +1EV, the noise floor would be lowered, because you'd have an image (the longer exposure at +1EV) with more signal information in the shadows.

Now, before you press "reply", think about what I just wrote.

By bracketing, however, the +1 exposure captures more shadows than the +0 exposure, while the -1 exposure captures more highlights. The trick is to blend them accurately. The OP wanted to post process, so this isn't a problem.

In the above example, if using DR100 you cannot expose at -1 to retain highlights, you have to go lower, to around -1.7EV.

For crying out loud, I've been talking about using DR200 or DR400, not just DR100. Would you actually read the suggested method before dismissing it? (side note: I have verified that AE bracketed X10 images contain 6 exposures, confirming my suggestions.)

With an HS10EXR's RAW files, the difference between the shorter of the two exposures at "-1 EV" the and the longer of the two exposures at "+1EV" would actually be higher than 2EV.

In X and F series EXR cameras the standard exposure is at "0EV", while the second frame is 2EV lower at DR400. How is an HS10's EXR processing diferent? How can the difference between the 2 frames be more than 2EV?

Again, would you actually read the text and the suggested method?

Of course not: a JPG retains 8 bits of color information; an X10's RAW file retains 12 bits,

It's not the bits, but the exposure, I believe.

No, actually; 8 bits doesn't carry as much information as 12 or 14 bits. Again, this is elementary mathematics.

Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow