This site is supported by the advertisements on it, please disable your AdBlocker so we can continue to provide you with the quality content you expect.

X-T2 at high ISO

Discussion in 'Wildlife-Nature' started by Fujiphotog, Nov 5, 2017.

  1. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    It wasn't too many years ago that the best DSLRs had an acceptable ISO of up to 800, and then it moved up to 1600. Nowadays, it seems that the upward limit has climbed into the stratosphere. The Fuji system has also kept pace, performing well at higher ISOs. Some of my recent photos at the local zoo were at 10,000 ISO and still acceptable.

    This tiger was photographed in a dark corner of his cage at the Toronto Zoo. The photo is a .jpeg, with cropping and minor editing only. The other photos were from RAF files.


    BLAK5559.png

    It seems a bit strange to see red Canadian autumn leaves in a photo of a rhinoceros.

    BLAK5325.png



    BLAK5550.png
     
    DaveX, LionSpeed, CWRailman and 3 others like this.
  2. CopperpotPhoto

    CopperpotPhoto Member

    Joined:
    Jun 4, 2015
    Messages:
    66
    Likes Received:
    17
    Location:
    Usa

    -Return to Top-

    Was this done simply to make a point? I ask because given your exposure settings and the fact that you had OIS on that lens you were shooting with, there really isn't any reason to be at that high of an ISO. You could have slowed your shutter speed down easily a couple of stops as well as the ISO and wound up with better image quality. None of these animals would be moving at a fast rate of speed in that setting. I do agree that the Fujis handle noise rather well for being APS-C though.
     
  3. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    Others on this forum had advised turning off the OIS at speeds faster than say 1/500 or 1/000, as OIS can cause some blur and focus delay at those higher shutter speeds, so that is what I did. These animals were quite far away in some cases, and the photos you see are significantly cropped. The 100-400 lens fully extended, plus 1.4 TC, is equivalent to over 800 mm, so I am concerned about hand holding at shutter speeds slower than 1/1000. Currently, I usually set ISO to auto, set the shutter speed at 1/1000 or faster, and the f stop either wide open or at one stop down from wide open.

    However, if you think I could improve my image quality with a lower shutter speed using the OIS, I can experiment to see if my hand holding is steady enough at 1/250 to 1/500 to be able to use 1-2 stops lower ISO. Thanks for the suggestion.
     
    Sanemancured and Brian Kimball like this.
  4. Madhav Bodas

    Madhav Bodas Well-Known Member

    Joined:
    Apr 15, 2016
    Messages:
    93
    Likes Received:
    47

    -Return to Top-

    Wow. Great images at such ISOs. U r right about modern digital cameras performance. They have brought many lighting situations under the realms of photography !
     
  5. Irene McC

    Irene McC Premium Member

    Joined:
    Feb 4, 2017
    Messages:
    939
    Likes Received:
    637
    Location:
    Cape Town, South Africa

    -Return to Top-

    I like the giraffe especially! I agree - my auto ISO is set to 12,800 at the high end, and
    I have often used it (inside cathedrals for instance) with totally acceptable results (RAW files)
     
    Fujiphotog and Mike Gorman like this.
  6. FujiMongol

    FujiMongol Member

    Joined:
    Oct 13, 2017
    Messages:
    16
    Likes Received:
    5

    -Return to Top-

    It is something I've wondered about, XT-2 in full auto, why it mostly chooses a shorter shutter speed before lowering the auto-ISO value.
    This is a matter of priorities that Fuji makes for some (to me) unknown reasons.
    I would happily trade a 1/2000 for a 1/500 for 2 stops of lower ISO, but it seems Fujifilm thinks otherwise in full auto.
    It keeps bumping the auto ISO in the mid-range while shortening the shutter speed.
    For good reasons? Does Fujifilm thinks everyone wallows and therefore prioritizes short shutter speed before dumping the auto ISO lower?
    All idea very welcome.
     
  7. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    I agree, even the excellent Fuji .jpegs are less than stellar at higher ISOs, but the RAF files hold up very well.
     
  8. CopperpotPhoto

    CopperpotPhoto Member

    Joined:
    Jun 4, 2015
    Messages:
    66
    Likes Received:
    17
    Location:
    Usa

    -Return to Top-

    Even with the OIS off, the Giraffe photo for instance could have easily been 1/800th of a second, drop the aperture to F/8 which would be wide open and get that ISO down to like 1600-ish range. The tiger photo didn't have much wiggle room without OIS, but that would be a perfect shot where you would want to use it. That tiger would have been barely moving and if it was moving, it would have been very intermittently and easily predictable. Therefore you could have easily engaged OIS, dropped your shutter speed to 1/250th of a second, and as a result, dropped the ISO to under 3200. OIS should not cause blur at those speeds. However, they are right in saying it can slow down autofocus. However, these are fairly static subjects. These are a prime example of when OIS would be useful.

    Furthermore, ISO is the absolute last resort for me. I would rather sacrifice a slight amount of DOF or IQ based on opening up the aperture wider than I would raising the ISO. The ISO is going to destroy IQ more than most anything else you can do (assuming your shot is in focus to begin with).
     
  9. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    I have read some articles here (e.g. by Dan Bailey) and elsewhere that have persuaded me that ISO is no longer as important as it used to be.

    For example, referring to the X-T1, :

    Please login or register to view links

    The expressions "ISO Invariance" and"ISO-less cameras" are probably exaggerations, as dynamic range and noise do vary with ISO in all cameras, but the degree of variance has become so reduced in the X-T2 and some other newer cameras as to make higher shutter speeds and higher ISOs desirable, within certain limits. Those limits have increased versus a few years ago, as the loss in IQ with RAF files after processing with the latest software is greatly reduced (at least 1-2 stops).

    If that is correct, with longer lenses like the 100-400 at 400 + 1.4 TC, it is now desirable to reduce the risk of motion blur by increasing 1-2 stops of shutter speed and ISO. In other words, with these cameras and RAF files, at wide open aperture, the trade-off between shutter speed and ISO has changed.

    That makes it desirable to turn off OIS as we move to higher shutter speeds.

    It also makes Auto ISO (again, within limits) more useful, thereby making the use of the camera less stressful because we no longer need to be continually changing as many settings.
     
    trainer and lawsofphysics like this.
  10. lawsofphysics

    lawsofphysics Premium Member

    Joined:
    Apr 17, 2013
    Messages:
    2,403
    Likes Received:
    879
    Location:
    usa

    -Return to Top-

    The XTrans III sensors actually have two ISO invariant ranges: 200 to 640 and 800 or above. Here's the Please login or register to view links.

    The lower range is ISO invariant to within ~ 1/3 stop. The upper range is invariant to within 1/3 stop or less.

    These data mean with raw files one only can only use two ISOs.
    • In bright light where dynamic range can be important use ISO 200.
    • In low light where S/N (shadow region detail) is important use ISO 800.
    In both ranges, typical levels of underexposure do not significantly increase read noise levels. It also means the primary noise source is photon (shot) noise. The camera electronics' noise levels are insignificant unless there is extreme underexposure (~ 3 stops or more underexposure at ISO 800)).

    The single problem with this approach involves inconveniences. You have to use raw files and in-camera image review becomes less useful as underexposure increases.
     
    trainer likes this.
  11. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    I looked at the data in the graph you linked, but I don't understand what you mean by "two ISOs", 200 or 800. I thought each of these is not a single number but a range.

    Why can we not "use" ISO settings of 400 or 600 in the first range and 1600 or 3200 or higher in the second range?

    I agree that the use of raw files is an inconvenience, but as I can't seem to get my .jpeg files to be ready to be used straight out of the camera in most shots, shooting raw is necessary, hence only a minor inconvenience.

    The in-camera image review issue is a major inconvenience. Unless I bracket by 1/3 to 2/3 stop or even more, what I see on the camera's screen when reviewing my photos is not quite what I get in the raw file when loaded into the computer. When I am reviewing my recent shots in the camera, I believe that the tiny histogram in the display back setting, as well as the image on the screen, reads the .jpeg the camera makes from the raw file, not the raw file itself. If that is correct, then the .jpeg based histogram I see in the camera when reviewing my raw file will be more to the right than the histogram of that raw file when i look at it in Lightroom or ACR.

    Do you know whether the spread or gap in exposure between the camera made .jpeg and the raw file is the same EV at all ISOs, or if not, how the relationship changes? Does the gap increase with higher ISOs or does it increase only with underexposure, regardless of ISO?
     
  12. lawsofphysics

    lawsofphysics Premium Member

    Joined:
    Apr 17, 2013
    Messages:
    2,403
    Likes Received:
    879
    Location:
    usa

    -Return to Top-

    For raw files, you only must use one of two ISO values (200 or 800) to achieve maximum technical image quality (S/N). There is no fundamental need for other values because,
    • the photo-diode conversion gain increases at ISO 800 and above
    • camera read noise is constant to within ~1/3 stop for both ISO ranges
    • the appropriate global image brightness can be achieved during post-production rendering.

    Increasing rendering brightness will not degrade IQ.

    For in-camera JPEGS you need to rely on in-camera signal amplification to achieve the appropriate image brightness.

    Bracketing raw file exposure (shutter time and or aperture) is very useful because you can select the exposure that retains only important highlight detail. This means you will have the best possible shadow region IQ (S/N).

    Underexposure has no direct dependence on ISO. It only depends on shutter time and aperture. So there is no gap. ISO signal amplification occurs after the shutter closes.
     
    Last edited: Nov 11, 2017
    Fujiphotog likes this.
  13. Fujiphotog

    Fujiphotog Amateur photographer.

    Joined:
    Jan 10, 2015
    Messages:
    1,265
    Likes Received:
    904
    Location:
    Toronto, Canada

    -Return to Top-

    I have been mulling over your two posts,combining what you said in both of them, and trying to figure out how to change my normal photographic process to take advantage of this new information about ISO and IQ, but not sure what to do.

    If I went back to the zoo, I would again want to shoot RAF at 1/1000 shutter speed to avoid motion blur with the lens at 400 + 1.4 TC (840 equivalent in FF) , and shoot with the aperture wide open because that lens and TC combination is already fairly slow (f8). If I now set the ISO at 800 rather than using auto (which would set ISO at 3200 or more) the meter in the camera would show 2-3 stops underexposure and the captured image I would see when reviewing it in the viewfinder would be very dark. In principle, as you have suggested, I could add 2-3 stops in processing without reducing IQ, but if I had underexposed by more than that, there would be a loss in IQ. How would I know whether I had underexposed too much? What, if anything, could I bracket that would make a useful difference?

    Conversely, if I was shooting a white sailboat in bright sunlight with the same lens and TC, using fill flash, f 8 and 1/1000 shutter speed, ISO 200, that could still blow out some important highlight detail. The histogram would be reading a .jpeg file, not the RAF file, so would show blown highlights even if they were still usable in the RAF file. How would I know whether I had overexposed too much? What, if anything, could I bracket that would make a useful difference?
     
    FujiMongol likes this.
  14. lawsofphysics

    lawsofphysics Premium Member

    Joined:
    Apr 17, 2013
    Messages:
    2,403
    Likes Received:
    879
    Location:
    usa

    -Return to Top-

    At the zoo...

    The exposure is limited by the required aperture and shutter time. No matter what you do, the sensor will be underexposed. The question then becomes what is the best approach to produce the desired global brightness when you render the image after the shutter closes.

    The answer depends on the ISO amplification electronics' design. For camera A, increasing the signal levels using electronic amplification also decreases the amplification noise level contributions. So you want to use the ISO (or range of ISOs) than minimizes electronic noise. For camera B, the noise levels are constant (to with !/3 stop) as signal amplification increases. Now there is no S/N advantage to changing ISO.

    If you use camera A, shadow region S/N will depend on ISO. This is not so for camera B.

    "How would I know whether I had underexposed too much?"

    This question is moot. The exposure of the sensor (shutter time and aperture) is limited by the circumstances (minimize blur and the aperture is maximized). If a 1/800 sec. shutter time could have resulted in acceptable blur levels, then you underexposed. Otherwise, the exposure was optimal. The image will be sharp.

    "What, if anything, could I bracket that would make a useful difference?"

    Since the aperture can't be increased, bracket the shutter time in 1/3 stop steps. Use the raw file with the best compromise between motion blur and shadow-region S/N.

    For the boat...

    With the X-T2 you couldn't sync the flash at 1/1000 sec. Also, lighting the entire boat with a flash in bright sunlight would require a very high-power strobe (or multiple flashes close to the boat). With proper lighting, dynamic range is no longer an issue. The shadow region S/N is high because the strobe added more signal (light).

    In this case, the JPEG histogram is a useful estimate of the highlight regions' overexposure. Based on that estimate I would bracket the aperture by 1/3 or 1/2 stop increments. The shutter time must be the shortest possible to sync with the strobe (1/250?) I would select the raw file that retained important highlight details. I would overexpose all other highlights. In this case, the important highlight details would be the white sail. An example of unimportant details could be sunlight reflections from the boat's chrome fittings.

    The light meter and, or histogram just provide useful estimates for the initial shutter time and aperture.

    The goal is to always maximize raw-file exposure as prescribed by Emil Martinec. (Please login or register to view links)

    "... maximize subject to three constraints:

    (1) maintaining needed DoF, which limits how much you can open up the aperture;

    (2) freezing motion, which limits the exposure time;

    (3) retaining highlight detail, by not clipping wanted highlight areas in any channel.

    Note that ISO is not part of exposure. Exposure has only to do with aperture and shutter speed. Maximizing exposure guarantees that one captures as many photons as possible subject to photographic constraints, and therefore optimizes S/N.

    How does ISO enter? It enters as a subsidiary aspect of optimizing S/N. On many cameras (those with CCD sensors, and the newer Sony Exmor sensors), there is little or no advantage to raising the ISO, which aids point (3) -- leaving the ISO at a low value may leave the histogram "to the left" for your chosen exposure, it will give more highlight headroom but will not degrade S/N; such cameras can safely be operated at close to their lowest ISO (the precise optimal ISO depends on the details of a given camera design). On the other hand, for many other CMOS sensor'd cameras, such as Canon's offerings, and Nikons with Nikon-designed CMOS sensors (D3/D700/D3s, for example), noise relative to exposure is improved by increasing the ISO; after you have maximized the exposure (ie by satisfying criteria (1) and (2)), you have a tradeoff to make for (3) -- raising the ISO lowers shadow noise (up to a camera-specific point of diminishing returns, usually about ISO 1600), therefore improving S/N, but reduces highlight headroom for your chosen exposure, so one has to decide how high the ISO can go and still keep wanted highlights unclipped."

    Anyway, the prescription is to set the exposure (shutter speed and aperture only) according to (1) and (2); back off the exposure if at base ISO and you are compromising (3). If you are compromising (3) with your chosen exposure and you are not at base ISO, then you should have started with a lower ISO. Afterward, depending on the specifics of the camera's noise profile, further optimization results from raising the ISO, up to the limit specified by (3), or the camera's ISO point of diminishing returns, whichever is arrived at first.

    So, it's (almost) all about ME
    ."
     

Share This Page