Why do bright highlights lose detail on a Nikon D700, even when shooting RAW?

Asked 7/27/2010

4 views

2 answers

0

I’ve noticed that photos of very bright subjects from a Nikon D700 can seem to lose separation in the highlights, even when the exposure doesn’t appear blown and the image was shot in RAW. Midtones look fine, but bright areas can look compressed or merged together.

I also saw comparisons between a Nikon D700 and a Hasselblad H4D where the medium-format file appeared to keep more distinction in bright areas. Is this due to sensor dynamic range, tone curves, highlight handling, lens/microcontrast, or RAW processing? If the highlights are not actually clipped, what can be done to preserve or recover detail in those bright areas?

Originally by Photography Stack Exchange contributor. Source · Licensed CC BY-SA 4.0

Photography Stack Exchange contributor

16y ago

2 Answers

4

Even though I did not overexpose them by at least 2/3rds of a stop, the pictures lacked a lot of details in high key

Can you clairfy that? Particularly the first part.

Yes, a MF camera will capture more total dynamic range than a FF, but light is light. If you underexpose and pull the shadows, you should get more detail than you're describing. The MF will have better color sensitivity but it doesn't appear to be what you are describing. It sounds like microcontrast, which can be affected greatly by lens and PP. Perhaps sensor could have something to do with it, but I cannot see why, especially at low pixel densities (as opposed to P&S which are quite diffraction limited). Do you have a link to the comparison and your own pictures?

Your camera likely has some sort of highlight compensation by default, especially in JPEG. Look at tone curves in DPR reviews, many cameras compress the highlights to capture more overall detail and simulate film. This is why most camera's ISO ratings are optimistic (ie it says 1600 but is more like 1100). I'm not sure if you can completely disable it on D700 and turn it off for JPEG, I know I can on K-x (often criticized for the linear highlights without highlight compensation).

If you are using RAW, you should have complete control. The Hasselblad user could have applied a better tone curve, but it should be within reach as long you capture the extremes of the dynamic range in the high key area you are talking about. After all, you are mapping 4096+ luminosity levels to 256 when you are viewing it on a computer. You may need to post process in some local-contrast.

Originally by user389. Source · Licensed CC BY-SA 4.0

user389

16y ago

0

AI Answer

Generated from our catalog & community — verify before relying on it.

If the highlights are truly clipped, a DSLR will lose detail quickly there. With digital capture, it often helps to expose a little conservatively to protect bright tones.

If you shot RAW and the whites are not clipped, the detail may still be present but compressed by processing rather than missing from the sensor data. In that case:

  • use a RAW converter with highlight recovery
  • check that your monitor isn’t set with excessive contrast
  • adjust tone mapping/curves carefully in post
  • use local corrections such as burning if needed

The difference you noticed versus a Hasselblad may not be only the sensor. Medium format can offer more total dynamic range and better color sensitivity, but perceived highlight separation can also come from lens rendering, microcontrast, and post-processing choices.

So this is not necessarily a “DSLR software problem.” First confirm the highlights are not clipped, then process the RAW file for highlight recovery and gentle tonal separation. If clipped, the practical fix is to protect highlights at capture.

UniqueBot

AI

16y ago

Your Answer