Narew

With macro rust can surely produce some creative stuff

At the release of Apple Silicon, there was this repos https://github.com/hollance/neural-engine That reference lot's of discovery and reverse engineer on the ANE.

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.

I use it via llamacpp and codex-cli.

Disclaimer : I worked at DxO.

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.

It's deprecated not removed yet.

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

The french gouvernement publish LaSuite recently https://lasuite.numerique.gouv.fr/#products They only host it for themself. And this time it's quite good because they don't try to reimplements stuff, they reuse some existing open source project. The repository for this : https://github.com/suitenumerique/

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 NCS is from 2018 so it's really outdated now.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.

It's more a Slack alternative than a Trello alternative ? Their website is so bad that I had to go to wikipedia to remember what mattermost is.

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.

Really nice 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.

It remind me a devlog from andrew kelley on the same topics : https://ziglang.org/devlog/2024/#2024-11-04

I'll interpret a bit. But I think the remark is more did we have a cross platform and complete gui library for swift.

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.

yes and it's only work because we prevent the output to be in the input.

I haven't read how they apply the constraint. But there is similar stuff when you force llm to generate structured output like Json format. llama.cpp allow to match a custom grammar for example.

here a example of it : (I only remember the first version) https://www.reddit.com/r/dosgaming/comments/86yxp4/question_...

for example "flashback" ask you a writen code in the manual at the beginning of each level if my memory is good. So it's was not a one time activation

For the moment it's a bit free wheel. And I agree the code is better but they could probably not reproduce it themself. I honestly don't know how to "force" them to understand the code the llm write if the code is clean. But this happen when the code produce by llm is over complicated or bad and we catch that by doing code review. I have the impression it will create even more disparity between student, students that just use llm code and the ones that try to understand it.

Honestly, the code from my interns have greatly improve since they use AI. And there is lots of really ugly and hard to read code from papers. So I don't think it will be an obvious loss of readability to have code completely generated by AI :)

It's not only fog. Depending on pupil size it can also be rain. But during my thesis 10y ago one of the big limitation on horizontal line of sight laser telecommunication was scintillation due to turbulence. And this phenomenon appear in clear weather condition with heat. So it was limited in clear and sunny weather. I don't know what up time did there expect ? But the bandwidth should fluctuate a lot.

Does anyone know a similar image viewer but that manage correctly the colorspace of the image and the colorspace of the screen ?

From my experience on C++ vs Rust on test algorithm. For a naive algorithm implementation rust is usually slightly faster than C++. But when you try to optimise stuff, it's the opposite. It's really hard to optimise Rust code, you need to put lots of unsafe and unsafe is not user-friendly. Rust also force you on some design that are not always good for performance.

I don't remember where I read that and was not able to find it again. There is a web/desktop app (like zoom) that install a font when you install the app, and the web app check if this font is install to trigger the open in app popup.

I don't know if it's good or bad, with pseudo code you put no constraints on how it should be implemented. It's known that some kind of algorithm are really hard to implement in Rust (every one use the link list data structure as an exemple). So having the article that use rust is good to see it can fit to rust constraints but at the same time does this constraint limit the algorithm itself ?