sanity checks for editing Nikon NEF files

RT v 5.10 New user.

I am trying to edit Nikon Z8 NEF files under win 7. (These are the “lossless compression” NEF files, not the other camera options.)

Because it’s not clear whether any codecs I’ve installed are actually working with RT, when I open a NEF file, it is unclear whether I am looking at the embedded jpeg or the nef.

Even though I have set the option to show the raw file even if not yet edited, if RT is unable to read the NEF raw component, how do I know it is not still just showing me the embedded jpeg?

For example, is there a way to see how many bits per pixel RT thinks it is working with, for example? Or that it thinks are in the original file?

Thanks

Embedded JPEGs may be used in File Browser. When an image is opened in Editor it is definitely raw data.

Thanks, but how do I verify that is the case? It seems to me that if there is no NEF codec, then there is no way it got opened as raw data.

If RT can open your images in the Editor tab you are good to go. Did you try that?

Thanks, but you are not actually answering my questions. Maybe someone else would like to give it a shot?

To me, the lack of something not happening is not any kind of guarantee that it did happen. That is why I am referencing this as a sanity check, and asking for specific ways to do the check.

RT can’t use and doesn’t need external codecs or anything like that on any platform. It has its own raw decoder built in.

Aha, thanks. I did not realize that, so that’s very helpful.

I would still like to know, however, how I can tell in the UI what bit depth RT is working with?

Sometimes RT decoders do not support particular cameras or their specific RAW format. The best way to check is to try to load a few images from that camera.
For example, until recent addition of LibRaw I could not open files from Sony A7IV which were lossless compressed.

I believe internally RT used 32bit float.

1 Like

See The Floating Point Engine - RawPedia. But you can rest assured, if RT can decode your raw, it will use the full ‘bit depth’ of the raw file. If it cannot, it will not sneakily show the embedded JPG, but rather will stop. Further, if you can open the raw tab (with low-level raw-processing options like demosaicing), you can be 100% sure you are processing the raw data.

If that still fails to convince you, the source code is available. :slight_smile:

1 Like

Turn off the demosaic…zoom in… you will see your raw sensor data that is being used…

2 Likes

Ah, yes. I was hoping to avoid reading the source code. However, past experience indicates that that would be a nice way to blow an afternoon, and yet have no answers.

It’s just going to use the data as directed by the specified raw black and white points. These come from the exif data or from RT camera support…it’s a text file I believe and you can search for your camera and see what values are used…

The exif data does not indicate pixel bit depth, only resolution, as far as I can tell.

Thanks to all who replied. I think I’ve got it figured out now.

I note, though, that I still don’t see any way to verify that the raw file actually claims to have 14 bit color depth, although knowing that is not critical as long as I know I am actually editing the raw file, not an embedded jpeg.

Thanks. Once I eventually figured out how to do that, it is quite definitive.

you should check you camera’s firmware settings if you want to know this for sure.

That assumes that I should trust what they claim, or that i did not make some obscure mistake in the settings that modifies the bit depth, or that the firmware has no bugs. Or maybe I should ask Nikon for the firmware source code.

BTW, it would also be interesting to see what the real bit depth is, i.e. how many low order bits are indistinguishable from noise.

That is generally accounted for by the raw black setting…what camera do you have??