I’ll bet this question came up with many photographers right after Nikon accounced the D850. Something like: “Oh wow! 46 MP? 9 frames per second? Ehhh, but what does this mean for my current D5/D4S/D500/D810 etc.?”
I got the D4S after my D3S and D4. And that’s where I stopped: the D5 was reported to have poor dynamic range at lower ISOs compared to its predecessors (and tests proved it). Next to that, I was very unpleasantly surprised by its price in Europe. It currently still goes for EUR 7000 while in the US it’s around $6500, so basically over $1000 more expensive in Europe! I have no idea what Nikon was or is thinking (like probably many others) but I’m not joining that party. To underline my sentiments, B&H have some interesting D5 reviews on their website. On a more positive note: Nikon has demonstrated they actually can disrupt the market through innovation with the release of the D850, so my hopes are on a D5S/D6. Time will tell soon… (expecting an announcement late this or early next year)
I got the D500 as a backup for my D4S and because of its crop sensor; so an easy way to get to 1.5 focal length on my primes and zooms. And it’s a great pro-level camera; great dynamic range (even outperforming the D5 below ISO 400…) and again a welcome surprise from Nikon in that they can produce excellent cameras at a decent price.
After my focus point misery with the D800E/D810 I decided to wait it out, and lo & behold: the D850 was announced! Its specs were so crazy that the obvious question came begging: “what do I do with my current camera(s)?” Is the D850 able to replace them all, for example? On my recent trip to the Great Bear Rainforest I shot the D500 with my 600mm f/4E FL prime, and the results were horrible: all images were soft. I still have no idea what went wrong, but I had much better results with the D850 instead. Then, a couple of weeks ago I was shooting deer here in The Netherlands and my images with the D850 were a little soft. So I tried the D500 and the result was much better! Very confusing… So, I thought: let’s do a little comparison on resolution, noise at higher ISO, and shadow recovery and see which one comes out best when adding all results together.
We’ll start with some resolution images, as usual with my standard test subject. All images were shot with the Nikon 600mm f/4E FL, at 100% crop, exported from ViewNX-i as TIFF and imported/generated in Photoshop CC as JPG. No sharpening applied.
The D500 image looks a tiny bit bigger than the D850 one, but it’s also a tiny bit less detailed, if that makes any sense. On the D850 ISO 64 vs ISO 100, I can see a small difference in RAW (in ViewNX-i/PS) but when generated as JPG that small difference is gone. Resolution is obviously much higher on the D500/D850, but the D4S is still as sharp as ever. If you can get close enough to your subject, the 16 MP on the D4S will do the job. However, it’s seriously outperformed by the newer generation D500/D850 on resolution. Another test subject to check my initial results:
Again, the D850 has the advantage over the D500, and I’m not taking into account the ISO 64 vs ISO 100 difference too much. The D4S still looks great, despite the unfair difference in resolution. Before I continue, I’ll include a last test shot, which reminded me again of my earlier article on how to use long lenses:
This was shot at 1/1000 s. as a quite a few of my other test shots. Same position, same VR setting, same everything. I remembered suddenly from my earlier tests that 1/1000 s. is a risky shutter speed in combination with VR: mostly it will work, but if you’re unlucky you’ll get bad ones as the VR sampling frequency seems to run at that speed as well. Therefore, unpredictable results.
Let’s continue with some ‘noise in the dark’ shots. For this I chose a dark spot in a bush at around the same distance as the previous test shots.
Again, a very slight cleaner image of the D850 compared to the D500 while the D4S is the best; mostly because of the much smaller resolution! (fewer pixels make bigger pixels receiving more light)
Same results around ISO 1600 as with 800: slightly cleaner images with the D850 but all three cameras perform well at this ISO. Let’s continue doubling the ISO:
The trend continues, but with the D500 I’m getting a bit more concerned than with the D850 and D4S. As these are 100% crops it will all be fine at for example 50% crop, but nonetheless there’s an early warning signal. The D4S is still clean. Let’s continue again doubling the ISO:
Similar results in the ISO 6400 range. I would want to remain below 6400 with both the D500 and D850, if I can. And if I can’t, I’d want to crop as little as possible. The D4S is still the cleanest. Moving on to around ISO 12800:
Definitely a ‘no’ on the D500, and wouldn’t want to go that far either with the D850. Here we’re getting into the realm of comparing horrible noise to horrible noise. I don’t want to start my impressions on the D5 again, but I do believe that when dynamic range goes below a certain point there’s just not much use in considering those images to be ‘beautiful shots’. Sure, for journalism purposes there may be a use to have grainy images with very little/no dynamic range at ISO 25000-100000+ for example, but it’s not exactly material you’d want to print on an A1-size poster and enjoy looking at (realising everyone’s different so I’m sure exceptions exist :))
Now: how about trying to bring back some of those shadows? I still find myself using exposure compensation in the range of 1/3 to over 1 stop in lots of situations where I’m shooting dark animals, and/or animals in dark environments. In those situations you’d want to bring some of the shadows back with as little as possible loss of image quality. Nikon (pro) cameras are well known for their capability to bring back shadows/darks during image editing, while keeping noise at an acceptable level, throughout the ISO range (except for the D5, just had to mention that again…) so I was curious how the D500, D850 and my trusted D4S would perform.
Clearly, again, the cleanest result for the D4S which only makes sense with those 16 MP. The difference between the D500 and the D850 isn’t significant but the D850 is slightly cleaner. Let’s double the ISO:
Same results as with ISO 800. Double again:
Cleanest image again for the D4S. At ISO 3200 I’d want to avoid having to add 2 stops of exposure in images both from the D500 and D850. More likely, another stop of exposure and only 50% crop would still give usable results. However, I see ISO 3200 as the start of a ‘warning number’ for shadow recovery for the D500/D850. Double ISO again:
I’d say that ISO 6400 is pretty much the limit for usable shadow recovery on all cameras, perhaps with an exception for the D4S. Results will off course be better with only 1 stop correction and not 100% cropped, but I think I’d stick to this maximum anyway. Finally, what does it all look like when we double the ISO again?
The D500 is out of the game first. And although the D850 looks a tiny bit better, I wouldn’t go that far either. Same for the D4S despite it’s bigger pixels. These images may do for journalism or UFOs, for example, but I wouldn’t use them for nature photography. Let’s see if I can summarise these (way too elaborate) test results:
Resolution and dynamic range/image quality:
Noise and shadow recovery in dark areas:
Nikon D850 vs. D500 vs. D4S: which one?
To close: the image below of a little bear cub was shot late afternoon in the Great Bear Rainforest with the D850; I must have cropped it to around 50% as I was still way too far for my D850/600mm f/4 combo. I added +1/3 stop exposure compensation on the D850 and added another stop in ViewNX-i/Photoshop CC at selective areas across the image. All this at ISO 640, and a walk in the park for the D850.
Nikon has managed to exceed many (pro) photographer’s expectations with the D850; it will be a challenge to continue that trend, but the hopes are high!
Please note that these images are protected by copyright and are not allowed to be used in any commercial way. If you’re interested in personal use only (like using as personal desktop/tablet/mobile background) then that’s fine; however any other use is prohibited by law.
And as usual, for any questions, comments or feedback, simple add comments below (preferred) or otherwise drop me a note at: firstname.lastname@example.org