Jump to content

Recommended Posts

I've just had the best time trying to understand what the ISO setting does in my new X-T30 II. ISO numbers are modeled after the old ASA numbers for film, and films were truly different to create their speeds (I was a big fan of Kodak Pan, Plus and Tri X at ASA 32, 125 and 400 IIRC).

But I find references online, in the manual, and in books, that seem to confuse all this.

So I bought RawDigger, which is software that reads RAW files including the ones my camera writes, and lets you do statistics on the pixels. Below is the table of data I generated. Filename is just the photo name, Sample_Name describes the central 600X400 pixel rectangle I did my measurements in, Shutter is the Shutter_Value reported in the EXIF (there's a rounded version of that that is more accessible but apparently less accurate), and for Red, Green and Blue we have an average and a standard deviation. This made it clear that the camera is writing 14 bit pixel values.

I took this series of photos with the Fujinon 80 mm macro lens at f/2.8, in aperture priority mode, letting the camera adjust the shutter speed to make my exposures "right" at every ISO. By "right" they seem to mean about a 14% or 16% gray, at least in the green channel which is the one I paid the most attention to. I had the camera on a tripod with the lens a few inches from a sheet of paper, but the lens was manually focused at infinity to blur any texture in the paper. The paper was taped to the inside of our downstairs sliding glass door where it was backlit by our gray day (which wouldn't have any flicker like our indoor lighting would), and I worked quickly so the light wouldn't change much. I kept shifting the ISO and taking a picture.

 Filename   Id   Sample_Name       ISO Shutter   Ravg    Rdev   Gavg   Gdev    Bavg    Bdev
DSCF0190.raf 1 2823:1885-600x400    80 0.023810 2216.86 37.914 4618.88 57.640 3319.76 47.421
DSCF0191.raf 1 2823:1885-600x400   100 0.019231 2235.03 42.159 4653.14 63.400 3344.33 52.997
DSCF0192.raf 1 2823:1885-600x400   125 0.014286 2207.49 46.383 4595.81 69.240 3302.66 57.713
DSCF0193.raf 1 2823:1885-600x400   160 0.011765 1107.19 25.931 2301.03 37.817 1654.98 31.877
DSCF0194.raf 1 2823:1885-600x400   200 0.009091 1107.62 29.042 2302.50 41.715 1656.32 35.520
DSCF0195.raf 1 2823:1885-600x400   250 0.007143 1104.86 32.353 2297.14 46.321 1652.51 39.415
DSCF0196.raf 1 2823:1885-600x400   320 0.005882 1108.27 36.223 2305.09 51.731 1659.00 44.053
DSCF0197.raf 1 2823:1885-600x400   400 0.004545 1105.60 40.168 2300.77 57.615 1655.81 49.440
DSCF0198.raf 1 2823:1885-600x400   500 0.003571 1100.87 45.328 2303.92 64.488 1655.62 55.007
DSCF0199.raf 1 2823:1885-600x400   640 0.002941 1106.97 50.532 2319.02 72.443 1668.62 61.623
DSCF0200.raf 1 2823:1885-600x400   800 0.002381 1102.26 56.709 2310.84 80.642 1663.07 69.028
DSCF0201.raf 1 2823:1885-600x400  1000 0.002000 1115.02 63.675 2338.81 90.478 1684.51 78.111
DSCF0202.raf 1 2823:1885-600x400  1250 0.001471 1103.31 70.903 2315.34 100.94 1666.91 86.671
DSCF0203.raf 1 2823:1885-600x400  1600 0.001176 1103.12 79.595 2315.34 113.06 1668.75 97.235
DSCF0204.raf 1 2823:1885-600x400  2000 0.001000 1137.55 91.134 2390.45 128.58 1722.92 110.99
DSCF0205.raf 1 2823:1885-600x400  2500 0.000714 1132.56 101.96 2378.89 144.38 1713.78 124.72
DSCF0206.raf 1 2823:1885-600x400  3200 0.000588 1152.40 115.72 2419.68 162.58 1740.69 140.23
DSCF0207.raf 1 2823:1885-600x400  4000 0.000455 1154.66 129.77 2421.64 183.12 1741.71 157.47
DSCF0208.raf 1 2823:1885-600x400  5000 0.000370 1145.31 146.93 2399.44 205.13 1723.30 177.25
DSCF0209.raf 1 2823:1885-600x400  6400 0.000312 1152.15 163.63 2413.90 229.57 1729.91 199.96
DSCF0210.raf 1 2823:1885-600x400  8000 0.000227 1198.36 190.44 2532.11 272.33 1806.28 233.41
DSCF0212.raf 1 2823:1885-600x400 10000 0.000185 1167.54 214.59 2465.34 303.63 1755.25 259.34
DSCF0213.raf 1 2823:1885-600x400 12800 0.000143 1203.58 244.62 2535.27 343.70 1800.83 294.57
DSCF0214.raf 1 2823:1885-600x400 25600 0.000071  774.31 236.88 1594.28 295.63 1159.30 279.98
DSCF0215.raf 1 2823:1885-600x400 51200 0.000036  411.89 177.73  838.00 220.37  614.93 210.30

Here's what I learned:

If by "sensor" we mean the sensor chip that includes analog to digital conversion, then over most of its range the ISO setting does change the sensitivity of the sensor by changing the analog amplification before the digitization. This, coupled with the changing automatic shutter speed (which was either mechanical or electronic at different speeds), kept the raw pixel values at around 14% or 15% of the full scale 2^14. At ISO 80, 100, and 125, which are all specially preceded by an "L" in the display, there's something else going on that I haven't figured out, but it would have overexposed areas that were just 2 stops brighter than my sample area, so not much headroom for typical photography. At ISO 25600 and 51200, which are both specially preceded by an "H" in the display, the exposure fell way off, and I think for ISO 12800 the analog amplifier must have already been at maximum gain, so all they could do is encode someplace that the jpg conversion should bump up the brightness for the two highest ISO settings. Note that over most of the range, ISO numbers go up by around 1.25X per step, but these last two are going up 2X.

It is interesting to see that the noise, or at least the standard deviation as a fraction of the average as a proxy, goes up steadily. In those last couple steps this standard deviation over the mean goes way up.

I might be able to clean this up with a more constant light source (which of course must not flicker), and maybe a frosted glass filter threaded into the front of the lens so there's less texture. All the same, I feel now like I know how to use the ISO -- whereas all the references I had left me nearly clueless what to do.

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Similar Content

  • Posts

    • I found the reddit topic i refere to :  https://www.reddit.com/r/davinciresolve/comments/1pc3f1e/cant_apply_new_fujifilm_gfx_55_lut/ "Update for y'all, It's just like what @ExpBalSat said, it's because of the backslashes in the names break them. I changed the file name and it works now. "   For me it was the solution. Realy annoying if it doesn’t work for you 😕  
    • Here is the solution to using the Eterna 55 file simulation LUTs in Davinci Resolve.   In general, do not use the FLog2C to film simulation LUTs as they are not supported by Davinci Resolve for two reasons: 1) Davinci Resolve does not support Fuji Gamut Color Space and 2) Davinci Resolve does not support FLog2C gamma.  Instead, use Flog2 which is supported by Davinci Resolve.  Here is an example.  Let's say that you want to use Classic Chrome simulation.  Do the following: Complete your color grade and use a CST to get to Rec 709. Add a node.  Use a CST to convert from Rec 709 to FLog2.  Output Color space is Rec 2020 and Outout Gamut is FLog2. Add a node.  Apply the FLog2 to Classic Chrome LUT Create a combination node from node in steps 2 and 3. Apply a Key to the combination node and adjust the Key Output Gain to get the amount of the combination node that you want applied. So that you do not have to do this over and over again, generate a LUT for the combination node.  Remember to turn off all other nodes before generating the LUT. Hope this helps others. Don  
    • Thanks for the insights. I think it's really hard to make a decision without having the two side by side! 
    • I don’t have the 23 f2 but I have read several times that it is considered a little soft at close distance, compared to the 23 f1.4 lenses. These will also focus at shorter distance from the subject, esp. the new one. So that might make a difference. The new 23 f1.4 LM WR  has better resolution, esp. in regard to the 40Mpix sensors, which you don’t have on the X-T2. What practical difference that makes for the value of the pictures one makes is disputable and subjective.  f1.4 will gather more light but with a smaller DOF, which may be desirable in some situations but not so in others, depends. If you like to shoot close ups, you will probably use higher f numbers to get a bigger DOF. Same for landscapes. If you are a bokeh fan, yes the f1.4 lens are better.  The older 23 f1.4 lens that you are considering is a very good and respected lens. The f1.4 vs f2 aperture per se is perhaps not so important. The 23 f2 is very small, light and practical and a great lens for travel and landscapes. So, go figure. I am afraid I just sent you further down the road to insanity !
    • First post here but long time fuji shorter. I use the XT2 with the 23mm f2 / 35mm 1.4 / 16-80mm f4 I'm considering the 23mm f1.4 r (Non-WR) About me: - I shoot black and white only. - I like macro details to wide open landscapes and everything in-between. - I shoot mostly for art, intrigue and creativity of the image. My question - is the 23mm f1.4 going to offer me any meaningful difference over the f2 for the above scenarios Thanks and sorry for bringing it up again...
×
×
  • Create New...