• Members 221 posts
    April 16, 2023, 2:34 p.m.

    Whose is it then?

    The phrase more ideal suggests something which is closer to an ideal. You can find several varied examples of the use of "more ideal" here.

  • Members 976 posts
    April 16, 2023, 2:59 p.m.

    I can find a lot of nonsense, like exposure triangle. Not interested.

    The viewpoint that lossy is pseudo-raw and not raw isn't just my viewpoint, it's a simple fact. Lossy isn't raw neither by Sony, nor by Nikon technical definitions.
    Nothing further. Ciao.

  • Members 1737 posts
    April 16, 2023, 4:01 p.m.

    I'm glad you're clear that they are indeed images. Now let's work on viewable. The raw image data in a raw file (as opposed to the JPEG preview image) is stored in a single plane. That makes it a monochromatic image. If can be viewed in the same way as any monochromatic image, ignoring compression if present. It may not look like what you want it to look like (it has a gamma of unity and is not demosaiced), but you can't say it's not a visible image. It is no more latent than any monochromatic TIFF file.

  • Members 711 posts
    April 16, 2023, 4:29 p.m.

    Sorta, kinda like "very unique."
    😏

    Rich

  • Members 27 posts
    April 16, 2023, 5:25 p.m.

    I neither understand your optimal exposure definition nor your maximum exposure definition. Neither was either has to do with dof and motion blur.

  • April 16, 2023, 5:36 p.m.

    So, what do you think are the constraints which stop you setting an arbitrarily large exposure?

  • Members 27 posts
    April 16, 2023, 5:39 p.m.

    Agreed, but it can't hurt to understand how to maximise image quality - doesn't "good enough" really equate to just optimising image quality.

  • April 16, 2023, 5:44 p.m.

    Interesting semantic point. Is 'good enough' optimum? I suppose, if there are trade-offs against 'better'.

  • Members 132 posts
    April 16, 2023, 7:05 p.m.

    Editor/camera histograms vs. RAW histograms

    The two SOOC jpegs examples below were shot the same except for the dynamic range setting (Fuji X-T2) - DR200 (basically off) on the left, and DR400 (2 stops of dynamic range "enhancement") on the right with their corresponding histograms in Lightroom directly below them...

    2023-04-16 13_57_09-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg
    2023-04-16 14_05_32-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg
    Flatter highlights in the DR400 example (R), but similar histograms.

    Below are the corresponding RAW histograms (via RawDigger). As can be seen here, the DR400 RAW histogram (R) is 2 stops darker (further to the left), in this case indicating 2 stops less exposure (due to the DR mode requiring a higher ISO setting and, relative to the first example at base ISO, a 2 stop reduction in sensor exposure.

    2023-04-16 14_55_52-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg

    So why isn't the image 2 stops darker in Lightroom too? Because the camera's jpeg processing has compensated for the reduced exposure used by the DR mode automatically, and Lightroom will similarly read a metadata tag in the RAW file telling it to also compensate by moving an, invisible to you, "exposure" slider 2 stops to the right before you ever see anything, producing an initial image of similar brightness to the example with "normal" exposure..A RAW histogram used in this way is a great way to understand what's actually going on "under the hood".

    But what does that mean in the real world? Well, in this case it shows that while the resulting images look similar at first glance, there is a price to pay for a sub-optimal exposure, pretty easily seen here upon closer inspection...

    2023-04-16 14_30_35-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg

  • Members 457 posts
    April 16, 2023, 7:33 p.m.

    DR400 is not two stops less exposure. DR400 keeps the set exposure (shutter speed and aperture) and lowers ISO by two stops (hidden). That is why the lowest possible ISO at DR400 is two stops above the base ISO. When the camera converts the raw to JPEG, it boosts the shadows and midtones (by two stops) while keeping the highlights intact.

    Adobe reads embedded DR mode data to lift the brightness of shadows and midtones while keeping the highlights intact.

    BTW, the DR modes are mostly useful for JPEGs.

  • Members 536 posts
    April 16, 2023, 7:35 p.m.

    You can get otherwise absent values if the camera does the integer math that causes the gaps first, and then interpolates replacement pixels for mapped-out, defective ones.

  • Members 1737 posts
    April 16, 2023, 7:37 p.m.

    OSPDAF pixel interpolation can do that, too.

  • Members 536 posts
    April 16, 2023, 7:38 p.m.

    There are usually more positive outliers than negative ones, from thermal noise.

  • Members 27 posts
    April 16, 2023, 7:59 p.m.

    Based on the motive, and I am a so called slow shooter, I set ss and fstop given the motive. That may limit the exposure I want, i.e. I'd like my in camera histogram to move to the right (...). If the motive is static, and the camera is on a tripod, I can adjust ss to get maximum (large?) exposure. If not, i.e. hand hold, I, as far as I know from manyyearso phogography, I am onl left with adjusting the ISO. What a the constraints fro your vanatge point?

  • April 16, 2023, 8:18 p.m.

    With respect to shutter speed, much the same as yours. With respect to f-number, how much DOF I need. Thus, my exposure ends up being determined by motion blur and DOF unless:
    1. It's limited by the maximum exposure the camera can accept.
    2. If I don't need much DOF, I might stop down a little to get the lens at its sharpest - rare because todays lenses tend to be very good wise open.
    So in the end motion blur and DOF are major factors in determining exposure.

  • Members 132 posts
    April 16, 2023, 8:23 p.m.

    DR400 does reduce exposure by 2 stops relative to shooting at base ISO at DR200, the post above clearly demonstrates that. Your are correct that if it makes sense to be at higher ISO anyway it just plays with the ISO implementation and does not result in a loss of sensor exposure, but that's not what happened with the examples in the post above. I did nothing except change from DR200 to DR400, and, with Auto-ISO active, selecting a DR mode is directly responsible for the automatic change in ISO implementation and the 2 stop loss of exposure relative to shooting at 200 instead.

    The camera's jpeg processing will apply a softer curve in DR mode (primarily in the highlights), but Adobe simply adds two stops of compensating linear gain with no tonal processing to the DR mode RAWs.

    2023-04-16 15_55_57-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg
    2023-04-16 15_58_47-New Improved Catalog Oct 15-2-2 - Adobe Photoshop Lightroom Classic - Library.jpg
    DR200 vs.DR400 RAFs in Lightroom with their nearly identical histograms.

    And yes, DR modes are for jpeg shooters (and can be useful for them, typically with some positive exposure compensation). I know how they work, but I normally shoot RAW and don't use them myself.

  • Members 27 posts
    April 16, 2023, 8:29 p.m.

    I wonder what optimum vs maximum exposure entails, as these two terms seem to be used in a not consistent way. If optimum is good enough, some sensor exposure potential is possibly left on the table - not only non-critical highlight clipings are ok, but there is potentially underexposure beyond that, IMO. So maximum to me is when no channel is clipped, but are exposed as close to clipping as is possible, from a RD analysis point of view. This may mean, though, that non even so called non-critical highlights are not clipped. But is that the exposure that adequately considers the motive? How often does a motive have non-critical highlights? So in the spirit of the discussion in this thread, we should IMO not use the term optimum exposure, or maximum exposure, but motive adequate exposure - maximising exposure where I need /want it in the motive, and ignore any highlight clipping(s) where I don't care loosing details.

  • Members 132 posts
    April 16, 2023, 8:51 p.m.

    From my point of view, an "optimal" exposure is really just getting the exposure you were trying to achieve. From a technical standpoint, that generally means maximizing the signal to noise ratio by recording to brightest highlight detail that you want to retain as far to the right as possible - which will often mean that letting direct light sources and specular reflections "go" is very often going to be necessary for achieving an "optimal" exposure. In my book, anyway.