Hey I'm not here to try and change your mind or school you on options beyond what I've already done. If you can't find a solution, you don't need to come back and convince me. I believe you.
I'll just say that for myself personally, if something proves this aggravating and complicated in my own work, I wipe the slate clean and start over. I'm usually doing something wrong by insisting on the complicated way. My dad used to say: simplicity lies on the far side of complexity.
The inverted tripod video I shared demonstrates that the lens could be touching the ground if you need. I don't know how possible it is to go any lower.
GroovyGeek wrote:
The "try another camera" responses are silly. I don't have thousands of dollars to burn on a system switch for one feature. Nor should anyone be contemplating a system switch for one feature unless they are making money with the camera and the feature in question is essential to their work. A famous landscape photographer switched from Nikon to Canon for that reason alone. I am not a famous landscape photographer. And yes, Canon's implementation of focus shift wipes the floor with Nikon's and Sony's. Particularly Sony's which soots at a pathetically slow 2-3 fps during focus shift.
I have been with Nikon for over 20 years. From an ergonomic point of view their cameras are, IMO, better than the competition. Their lenses are top notch. The aggravating thing about focus shift is that it is one of the worst implementations of all major vendors, and it is a totally self inflicted wound. There is absolutely no reason for it to be as brain dead as it is. ...Show more →
You made it sound in your original rant like it's something very important to you, so most of us assumed you're making money from it so switching system suggestion made sense.
While the implementation of focus shift is very important to you, for most people in most uses its fine as is. Nikon may in future change it/fix it but i doubt it makes even top 20 of their todo list as like it was said - it's a very niche thing.
Immortal wrote:
While the implementation of focus shift is very important to you, for most people in most uses its fine as is. Nikon may in future change it/fix it but i doubt it makes even top 20 of their todo list as like it was said - it's a very niche thing.
Also, curious if you've written Nikon support with a detailed alternative workflow?
I've written a few companies, Nikon included, to suggest product revisions and I always come with a detailed statement on: 1) why the current implementation is not good, laying out its shortcomings and demonstrating a real world example where it makes work unnecessarily frustrating, and 2) a better implementation, complete with UI sketches if needed.
Sometimes, working through an issue like this puts me in touch with how I've been thinking incorrectly about a device or workflow. Sometimes it doesn't, and I send the info along. I've had one company write a reply and thank me for the ideas and they'e taken it to "higher ups". And Nikon implemented some things I wrote them about in firmware 2.0 - I am not at all suggesting they did it *because of me*, more that I'm suggesting they're listening and if enough people make a convincing case for something they might implement it in a future update.
Jman13 wrote:
I don't think it has anything to do with when the camera connects - but rather Nikon creates the certificate, puts it on the camera, but then I think it has to register that certificate with the group doing the credentialing. And I think that's what matters. If the shot is taken before the certificate is 'active' it's viewed as not genuine and therefore 'missing'. For instance, as I said, mine seemed to go active at 1:26 PM (well, it was actually 2:26, but I forgot to set my camera's time ahead for daylight savings in the spring). This was about 5 hours after I enabled it through Nikon Imaging Cloud.
However, I was actually out in the field shooting at this time. Was not connected to my phone and nowhere near either of the two Wi-Fi networks registered to the camera. There was no way for my camera to connect to Imaging Cloud at that time. And yet, one minute there are no credentials on the images - and literally partway through the shoot it switches, and all the rest are credentialed. ...Show more →
I root-caused this issue today. The camera is interpreting the certificate's GMT start time as local time, causing it to delay embedding credentials until the local time matches GMT. I posted details in my running C2PA deep dive thread on DPR:
snapsy wrote:
I root-caused this issue today. The camera is interpreting the certificate's GMT start time as local time, causing it to delay embedding credentials until the local time matches GMT. I posted details in my running C2PA deep dive thread on DPR:
FYI, today I found a vulnerability whereby the camera can apply C2PA credentials for photos it didn't take. I discuss my findings in my running C2PA deep-dive thread on DPR:
snapsy wrote:
FYI, today I found a vulnerability whereby the camera can apply C2PA credentials for photos it didn't take. I discuss my findings in my running C2PA deep-dive thread on DPR:
argonphoto wrote:
I was just reading the post about your discovery on nikon rumors last night. I wonder if Nikon knows about this now that your findings are out there.
Both Dpreview and PetaPixel reached out to Nikon for comment while writing their articles. PetaPixel said Nikon told them they are currently investigating:
I wrote NEF raw data encoder software that lets me take any image file and encode it using Nikon's proprietary NEF lossless compression bayer pattern. In other words, it goes backwards from an image back into the bayer pattern. This lets me replace the contents of the data inside an NEF with data from any image, including images created entirely by AI. You can read the technical details of this on my DPR post here.
Here is an AI image I created in Google's Gemini, which I then encoded inside an NEF and used the multi-exposure vulnerability to get the camera to sign:
It’s sort of terrifying how quickly you were able to expose the flaws in this verification. I’m hopeful Nikon and whoever is running C2PA can fix the vulnerabilities, since right now it is effectively useless. Thanks for rhe work in bringing this to light.
Good work @snapsy. I'm not someone who is personally using this/in need of it right now, but glad this is getting exposed quickly so that folks don't come to rely upon it erroneously. I do wish Nikon was figuring things like this out in house before releasing it to the wild.
Jman13 wrote:
It’s sort of terrifying how quickly you were able to expose the flaws in this verification. I’m hopeful Nikon and whoever is running C2PA can fix the vulnerabilities, since right now it is effectively useless. Thanks for rhe work in bringing this to light.
Thanks. Fortunately the vulnerability is limited to Nikon's oversight of letting users sign multi-exposure overlay images, with the core C2PA functionality not compromised. I have concerns about how well they're protecting the private key though - I doubt the cameras have the Hardware Security Modules (HSMs) recommended in the C2PA specification.
