From what I personally tested Ornith-1.0 35B is slightly better than Qwen-3.6 35B.
My tests are tasks that consist of adding/modify feature in a big C++ codebase.
The part that I find interesting is that the model is way faster than Qwen3.6 35B. It seems Ornith produce a smaller chain of thought.
On my test it can be 3 time faster to produce the answer.
DNG is more a container in which you can store lots of different think and not only RAW images.
For example PureRaw output Linear DNG that is not a RAW anymore. In the same way ProRaw is not a RAW image dispite it's name. But yes PureRaw and some ProRaw are compressed internally with jxl.
Ubisoft release some tools to simulate colorblindness. It's easy to extract de color transformation from their shader directly.
I'm not colorblind but I use that quite often to check roughly if the color palette I choose is fine
https://github.com/ubisoft/Chroma
There was in the past.
Google had Coral TPU and Intel the Neural Compute Stick (NCS).
NCS is from 2018 so it's really outdated now.
It was mainly oriented for edge computing so the flops was not comparable to desktop computer.
cxx only support a part of c++ STL container. It doesn't support template and custom container implementation. So no generic, no variadic number of argument, no automatic type déduction etc...
Appart from cxx.rs. I think most binding between C++ and Rust are in fact C and Rust binding. So you need a C API to your C++ library (no template, no std etc...). So for me you can't use C++ library from Rust.
I'm not a front end dev and only use JS stuff time to time for small personal project.
There is so much JS framework out there that appear and disappear so fast. I don't know if we can call it innovation. I have the impression they just reinvent the wheel with so little value added.
I prefer to keep on React at least it will not disappear the next time I will do some change on my project.
That said, I do want to point something out: the “Fast (because Rust)” or “Lightning fast” tagline that’s increasingly common in Rust projects isn’t great. Just using Rust doesn’t automatically make something fast — real performance comes from design, algorithms, and benchmarks.
These days, whenever I see “Lightning fast” in a Rust project, it makes me think the author hasn’t really focused on optimization, and that the project might not actually be as performance-tuned as the claim suggests.
Some of the distortion shown in the article is call "Volume Anamorphosis". It's a distortion that strongly deform face and person. This deformation is really visible for short focal lens.
Disclaimer: I work for a photo processing software.
There is. If you use lightroom app for example you can have access to raw pixel.
But I'm not sure there is a way to get all the images the camera app from the iphone take.
Phone don't take one shot to create the final image. they take hundred of shot and combine them.
It's funny because I have completely the opposite stance. When I code in rust (mainly algorithm), I always struggle to change what I want to do to what rust allow me to do. And all this complexity has nothing to do with the problem.
I personally think that the discovery of rust API is awfully as soon as macro is involve. It's impossible to know what argument is taken by macro.
For example the "simple" `println!`, without example in the docstring it's impossible to guess what are the argument even when we have an IDE that show you the implementation.