Nikon Z Series High Efficiency RAW Support

[mikae1]

@martinus

This leaves us with the conundrum: should we use the high-efficiency RAW files in the field and convert back to DNG on our system, or should we just go for the full-size lossless RAW?

In camera lossless proprietary raw and wait for DNG v1.7 files with lossless JPEG XL compression to be supported in darktable, would be my advice.

[paperdigits (Mica)]

@martinus

But it seems (after reading numerous articles) that it is visually identical and the processing capabilities don’t seem to change.

Of course that is what they tell you. For me, personally, the point of a raw file is that I get unadulterated data from the camera sensor, as much as possible. So using a lossy compression format on a raw file just does not compute for me. I also don’t need the extra frames-per-second that this lossy format was developed to enable, so its pretty much useless to me.

@martinus

It is a pity that camera vendors don’t release libraries to decode the files.

Well Nikon licensed this compression technology from a 3rd party vendor, so there must be something special in it, I guess. It is patented up the yingyang, so it seems unlikely that it’ll be supported by FOSS raw processors any time soon.

@martinus

This leaves us with the conundrum: should we use the high-efficiency RAW files in the field and convert back to DNG on our system, or should we just go for the full-size lossless RAW?

For me that’s easy: I use the format supported by the FOSS community, this gives me the most assurance now and into the future that I’ll be able to read and process my raw files.

[martinus]

@paperdigits

Of course that is what they tell you. For me, personally, the point of a raw file is that I get unadulterated data from the camera sensor, as much as possible. So using a lossy compression format on a raw file just does not compute for me. I also don’t need the extra frames-per-second that this lossy format was developed to enable, so its pretty much useless to me.

Well, it is not only Nikon telling, I have read many articles that seem to verify the claim. You maybe don’t need the speed. But there are types of photography where the speed matters. So in some situations I am willing to trade of maybe fraction of image quality versus speed because that gives me the shot.

[artichoke (Nikolai)]

Even if darktable/libraw were to create an implementation of Nikon’s high efficiency raw format, the chances of Nikon going out of their way to enforce patent protection are very low. Patent holders go after the money, not hobby source repositories.

[paperdigits (Mica)]

@artichoke

the chances of Nikon going out of their way to enforce patent protection are very low.

the patent isn’t Nikons, they licensed the compression from another company.

Also, welcome.

[Terry (Terry Pinfold)]

@artichoke

Nikon going out of their way to enforce patent protection are very low.

In the past I have known of Nikon threatening an online poster of one of their repair manuals even though Nikon was unwilling to supply the repair manual or the $20 part to repair my Nikon because it was over 10 years old. This nasty experience put me off Nikon and I now look at a my Nikons as a ten year product and then probably no parts available.

I find it disappointing that any camera manufacturer will not freely supply the code required to process their captures in a user’s software of choice. This would be a selling point to me when I buy a camera. If DT can’t process the images I don’t want the camera.

[Donatzsky (Nis Donatzsky)]

@artichoke

the chances of Nikon going out of their way to enforce patent protection are very low.

Nikon probably won’t, true, but IntoPIX very well might. They make money not just from licencing the encoder, but also the decoding SDK. And it seems that they insist their SDK must be used. If you want the situation to change, you should get in contact with IntoPIX.

https://www.nikoncafe.com/threads/nikon-he-he-nraw-support-as-of-31-march-2025.331229/

Nikon HE* decoder

I spent the last month having a fun little science project. Even though not owning a Nikon camera at all, I tried if it’s possible to get a working decoder for the HE* bitstream with the help of Claude. When using samples from: https://photographylife.com/nikon-z9-sample-images-raws (all copyright belongs to the author) the decoder produces this: [decoded] So you can see the colors are still off even when using WB coeffs from the EXIF and Color Matrix from Adobe. So the inverse MCT is still …

intoPIX TicoRAW | RAW Bayer CFA Sensor Compression (IP cores & SDK)

[paperdigits (Mica)]

To be completely blunt, releasing an open source TICO raw decoder is exactly the kind of legal challenge that both patents and LLMs need.

[Terry (Terry Pinfold)]

@martinus

But it seems (after reading numerous articles) that it is visually identical and the processing capabilities don’t seem to change.

My simple answer to this is if DT or my program of choice can not process the image then the processing capabilities have changed and are more limited. I personally avoid compressed raw because not all software is capable of supporting it. I also wonder what is being loss to compress it, but that is just an academic rather than practical issue. Anyway memory cards are so cheap now compared to the early days of digital cameras.

[martinus]

@Terry

My simple answer to this is if DT or my program of choice can not process the image then the processing capabilities have changed and are more limited. I personally avoid compressed raw because not all software is capable of supporting it. I also wonder what is being loss to compress it, but that is just an academic rather than practical issue. Anyway memory cards are so cheap now compared to the early days of digital cameras.

I agree, that if your editor of choice doesn’t support it, then you can’t use it. And thus, I shoot the normal raw. However, if the support is there, and the processing capabilities don’t change, I will use it.

I don’t find it cheap, especially with the high memory prices… And it is not only the memory cards. I don’t know about you but have multiple backups… so it is 4 times (memory card, local disk, usb backup disk and cloud) and that adds up.

[Donatzsky (Nis Donatzsky)]

@paperdigits

To be completely blunt, releasing an open source TICO raw decoder is exactly the kind of legal challenge that both patents and LLMs need.

Yes, calling their bluff by releasing a decoder is also an option.

[Toast]

Processing apps should do their best to handle everything because people often have a library of shots already. You can’t expect people to reshoot old images when they change tools. Obviously not everything can be supported.

[jjbraun (Joachim Braun)]

I was wondering if it would be a solution for you, to shoot in high efficiency and later convert to dng?
I know not the answer you are looking for. I am doing the same with pixelshift nefx files from my z8 as those are not supported in darktable.

[martinus]

I have been trying that out, but I went back to just the regular NEF files. It is an additional step which takes time, espacially with 1000s of photos. And it added only the additional dng checksum as an advantage

[Terry (Terry Pinfold)]

We should all whine to Nikon. We buy a camera and we want to process the images in the software of our choosing. The code should be freely available.

[g-man]

https://www.nikonusa.com/content/contact-us

[martinus]

@Terry

We should all whine to Nikon. We buy a camera and we want to process the images in the software of our choosing. The code should be freely available.

I have reached out to Nikon. But the conversation went a bit a like:

1. Well it is third-party, but hey you can use NX Studio to open the file and convert to tiff.
2. O you don’t like TIFF, well you can you de DNG convertor from Adobe!
3. O you don’t want to you that well you can always shoot the uncompressed raw!
4. O but the want to use the high efficiency Raw? mm well go to 1)

But yes, the more people the reach out… the better…

[g-man]

Nikon had a market cap of 3.5Billion. What incentive they have to release their code?

[martinus]

It is not even their code, it is third party. So maybe we should start whining with IntoPix who developed TicoRaw

[ggbutcher (Glenn Butcher)]

The only reason I’d have for a ‘whine’ is if they did away with the uncompressed format. To my way of thinking, buying more storage is a far more cost/effort-effective solution than getting all pissed off about one format option.

@artichoke

Even if darktable/libraw were to create an implementation of Nikon’s high efficiency raw format, the chances of Nikon going out of their way to enforce patent protection are very low. Patent holders go after the money, not hobby source repositories.

Nope, first-hand experience. Decades ago, I wrote a little TCP server program for a model railroading application, got a take-down notice from a vendor of a similar program that was covered by a patent for a ‘TCP server applied to a model railroading application’. That was the basis of the patent. He was later trounced by other hobby developers for a copyright snafu.

Oh, and reading this thread, compelled to point out the fundamental difference of a patent from copyright. The former applies to a mechanism, in any of its forms, the latter to a specific encoding of an idea. So, just because you came up with your own clean-room implementation of someone else’s patented mechanism doesn’t mean you’re in the right. And, ‘clean-room’ isn’t disassembling someone’s executable, it’s implementation of the functionality without any contact with the original implementation.