4
\$\begingroup\$

After reading this question and its answers, I came to wonder if Nikon full-frames handle ISO more or less in the same way as Canon does.

Quoting an answer of this question:

The native ISO for almost all Canon DLSRs over the last few years has been ISO 100. 'Full stop' intervals, such as ISO 200, ISO 400, ISO 800, etc. increase the analog amplification of the signal readout of the sensor. The 1/3 stops in between those full stops use software adjustments during in-camera processing of the data coming off the sensor. Here's what happens when shooting in P, Tv, or Av mode if you select, for example, ISO 160 when you take a shot. The sensor is set to ISO 200. The camera overexposes the shot by 1/3 stop by increasing Exposure Compensation (E.C.) 1/3 more stops than the user selected value. When the data from the sensor is read into the processor, 1/3 stop of pull is applied to the data.

If so, it apparently would mean that for example ISO 4000 would be noisier than 5000. Can anyone confirm or disprove that for Nikon too? Examples would be a plus.

My camera is a D610, but out of curiosity I would like to know more in general.

\$\endgroup\$
6
  • 1
    \$\begingroup\$ Name specific cameras please. \$\endgroup\$ Commented Feb 28, 2017 at 7:20
  • \$\begingroup\$ If you are going to downvote, please at least tell me what's wrong with the question. \$\endgroup\$ Commented Feb 28, 2017 at 7:36
  • \$\begingroup\$ With Nikon it depends on the specific model. Some sensors in Nikon cameras are done in-house, some are farmed out to companies like Sony and Toshiba. \$\endgroup\$
    – Michael C
    Commented Feb 28, 2017 at 7:54
  • \$\begingroup\$ So the "full-stop ISO use" would be a "general purpose" solution, as I understand... \$\endgroup\$ Commented Feb 28, 2017 at 8:43
  • \$\begingroup\$ @Noldor130884 It depends on the specific sensor and how the signal is multiplied for specific ISO settings. There is no general purpose solution. For every Canon camera since around the time of the 1Ds Mark III through at least the end of 2015 the sensor has handled "partial" ISO settings the same. I haven't looked to see how they are doing it with the newer sensors in the 80D, 1D X Mark II, 5D Mark IV, etc. which have a different way of applying some on-sensor NR. \$\endgroup\$
    – Michael C
    Commented Feb 28, 2017 at 16:39

2 Answers 2

2
\$\begingroup\$

Just a hint for the experiments. I am not really answering this question, as I do not like to compare little tiny piece of the algorithm inside the camera, but rather end to end result, and I am a fun of constructive and creative learning.

I think except automatic programs no one cares about 1/3 ISO increase. Important is how much noise you have when you are making bigger steps. In the links below you will find great shots done using big ISO (not very big, but quite big).

Look at images done with Nikon 610 at: ISO from 1600 to 3200 and compare them with e.g. ISO from 6400 to 12800 or e.g. Canon EOS 5D mark IV ISO 1600 - 3200 and ISO 6400 - 12800

When you had a closer look some of them are of similar quality.

But anyhow I recommend to make yourself the following experiment. Use RAW images and no JPGs. Use rather Manual, Aperture, Shutter speed or Program mode and not any kind of Auto, to avoid uncontrolled ISO changes.

Take tripod. And in good lighting conditions shot photo of the light object in the ISO 3200, then increase it by 1/3 and shot again, and then again 1/3 - shot, and then take ISO 6400. Then take the same object and made the same session in the dark place with poor lighting (do not use flash). Poor for me means e.g. object lightened by candle or some ambient light reflected by the light wall. Additionally copy of the images convert to BW.

If you are patient you can repeat photo session with dark object. The result will be interesting. As shooting light object in the light environment is more like shooting edges. Light object in dark place is shooting contrasts. Dark object in light environment is like catching contrasts and dark in dark is again about edges.

You will get your photos and compare them in sense of noise. Do not post process images. Observe them in RAW in full size. JPGs and any kind of scaling will change noise.

Get your feeling not how "big" is nose, but rather how equally distributed is it. What do you think when you look at images. It is like nice image with few damaging oversaturated places or evenly distributed noise which gives a bit perception like you look at old photo or some fuzzy memories or almost forgotten dream.

At the end you will have a full table of ISOs and you will see how your camera is working and how your mind is interpreting results.

At the second and later end - the photography is not about ISO, it is about the final perception of the image.

Completely another story is how to post-process images to change influence of the noise made by increased ISO or how to bring the noise to the low ISO images to make them looking like old photos.

Have a fun with shooting and less care about technical interiors of the camera :-).

\$\endgroup\$
4
  • \$\begingroup\$ While you have some generally good tips and approach at qualitative assessment of images at different ISO values, this doesn't really answer the OP's question. It is well known that Canon's cameras have reduced noise performance for ISO values that are not full-stop multiples of ISO 100. The OP's question is, do Nikon cameras exhibit the same behavior (especially the D610)? \$\endgroup\$
    – scottbb
    Commented Mar 14, 2017 at 3:00
  • \$\begingroup\$ My approach was to make experiment and step 1/3 around given ISO. This experiment I made with my camera. And I know practically results of the stepping 1/3 :-). That was my proposal. \$\endgroup\$ Commented Mar 14, 2017 at 5:08
  • \$\begingroup\$ Upvoted because I think you wrote some nice stuff about experimenting, but I can't mark as question answered, since it didn't quite answer the question. I do have fun in shooting, but I'm kind of a noise fetishist... I bought a full-frame, because I happen to shoot a lot in dark environments, so I could use the better high-ISO performance. I do post-process my noisy images, but I'd rather keep that to a minimum, since I don't want my images to lack contrast or be "too smooth", and that's the reason behind the question. \$\endgroup\$ Commented Mar 14, 2017 at 6:55
  • \$\begingroup\$ One more comment. I just once more looked at settings of my camera regarding ISO. I have in the camera option to reduce ISO noise. This is as well influencing final result. BTW I was experimenting with this option and I was not quite satisfied, when it was totally switched off, as I got much more over saturated points. I like random noise like it was on the old analogue films, but my DLSR sometimes ends up with bigger over saturated (red or white) spots, which I have to manually repair, and that one's I do not like :-). \$\endgroup\$ Commented Mar 14, 2017 at 7:16
0
\$\begingroup\$

Noldor, all cameras' images get noisier as your crank up the ISO. Each camera has a native ISO, and only applies its own internal image processing as ISO is increased. Unfortunately, as more light is applied to the sensor, interference a.k.a. noise increases between adjacent cells. This causes the 'noise' we see in our images.

I read an article where the author made an argument that you could process the images better than the camera, that you should shoot at native ISO all the time regardless of how dark the image was and then process your own shots in something like Photoshop. I tried it, but it was a failed experiment. My final images didn't look as good. Actually I think this is logical since the manufacturer will optimize their own post processing and could therefore logically do it better than you or I could. Anyway...

\$\endgroup\$
2
  • \$\begingroup\$ See my comment to Seweryn's answer. Specifically, this doesn't really answer the OP's question. It is well known that Canon's cameras have reduced noise performance for ISO values that are not full-stop multiples of ISO 100. The OP's question is, do Nikon cameras exhibit the same behavior (especially the D610)? \$\endgroup\$
    – scottbb
    Commented Mar 14, 2017 at 3:02
  • \$\begingroup\$ I agree with scottbb. I do know that high ISO multiply the noise, and I do know that cameras have a native ISO (range) where they are optimized, but I'm inquiring about something else here :) \$\endgroup\$ Commented Mar 14, 2017 at 6:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.