More on D500 Dynamic Range

Updated

My friend Iliah Borg—one of the folk behind RawDigger and FastRawViewer—sent me an email tweaking me on my comments about the D500 dynamic range versus the D7200. He’s got a point, so let me repeat it here.

The D5 and D500 have a different approach at the Bayer filtration than most previous Nikons. In particular, Iliah pointed out the following WB coefficients for the two cameras (here under Daylight WB, and I’m going to round Iliah’s actual measurements so that’s it’s simpler to see the difference):

  • D7200: 2.40, 1, 1.60
  • D500: 2.27, 1, 1.53

Note that the Red (first number) and Blue (last number) values are lower for the D500 than the D7200. This is also true of the D5, which has lower coefficients, as well.

What’s that mean? 

Simplest answer: we’re seeing less color noise, even though the opposite should be true if the filtration was reduced. Moreover, Nikon's not jerking the raw data values around as much to create RGB, so the dreaded “tint” issues that come up at high ISO values are lessened. I suspect white balance preconditioning.

Iliah also pointed out something else I had noticed but hadn’t yet formed a full opinion on: the low level noise pattern on the D500 is more random and grain like. The D7200’s low level noise pattern tends to have slightly wormy patterns to it.

Thus, Iliah’s conclusion is that he believes the D500 is more like a half-stop to two-thirds stop of useful gain over the D7200 at the higher ISO values. 

I’m still cogitating on that. I’m also still in the middle of doing my own more extensive dynamic range testing. Like Iliah, I don’t just use the standard tests, but also use exposure sequences with tough-to-resolve detail, which I then visually examine closely. This gives me a better sense of the boundaries of accurate data acquisition and, with JPEGs, a better idea about how Nikon is applying curves to the linear data.

I will say this: I do see differences in the JPEGs that clearly favor the D500 over the D7200 in certain tonal areas, especially at higher ISO values. I see the lower presence of color noise. I see the lowest level noise being remarkably random. I’m still trying to understand where those things come from and how much better that makes a D500 capture than a D7200 one. Of course, all that testing would be easier if I didn’t have to normalize 20mp against 24mp ;~). 

Suffice it to say that I’m happy with the D500’s dynamic range. It’s certainly a step upward from the D7200, which was already essentially as good as it gets in DX/APS. Is the step a fifth of a stop, third of a stop, a half stop, or two thirds of a stop, I can’t say with any certainty. I’m currently on the low side, Iliah’s on the high side. I value his tests and opinions, so maybe he’s onto something I’ve not yet seen.

Still, there’s no question in my mind that the D500 is performing quite well in low light conditions. Better than my D7200 and better than the Canon 7D Mark II I’ve used a couple of times now. That’s really all most people need to know. 

text and images © 2020 Thom Hogan
portions Copyright 1999-2019 Thom Hogan-- All Rights Reserved
Follow us on Twitter@bythom, hashtags #bythom, #dslrbodies