891–900 of 1001 posts
So far, they are based on ViTPose+ Large pose estimation in the context of different sports.
My demos include:
- Comparing dancers' sync performing the same choreography and computing a sync score
- Visualizing a runner's ankle path while running, as well as the average knee shape upon foot strike
- Counting chin-ups reps and measuring each rep duration
The code is open-source on GitHub:
Started a few different things: 1. A robotics information and news website. Any feedback is appreciated. https://robotica.dev/ 2. Applied DBSP theory to Postgres https://github.com/getravi/pg_dbsp and DuckDB https://github.com/getravi/duckDBSP . The idea is to have incrementally maintained materialized views.
I've been looking for a site like robotica.dev that pulls together news and events from multiple sources. The funding page is awesome and unexpected, a really useful mini-crunchbase for robotics.
It has LSP support, debugging, Git and GitHub integration, multiple cursors, customizable keymaps, and live previews for things like Markdown, PDFs, and diagrams (and lot's of bugs! too). It runs on macOS, Windows, and Linux, with no accounts or telemetry.
It’s a hobby project I work on in my spare time, and it’s still very beta. No big announcement or anything, I just thought I’d share it here.
If anyone tries it, I’d be interested to hear what you think.
Its basically a tool I made to make my own own business research easier, then decided to formalize it after a friend showed interest. What is it? its a way to evaluate a business idea from multiple angles and perspectives - similar to what a VC would do, only via a simple wizard that asks you to score the business from different angles. The next version will let one or more AIs answer the wizard so you can compare your evaluation with theirs.
Two scores come out the other end, kept deliberately separate — how good the opportunity actually is, and how good a fit it is for you specifically.
https://easel.games A programming language which lets you code multiplayer games like singleplayer games. Every program written in Easel is guaranteed deterministic and snapshottable, which is how it can automatically make your game multiplayer automatically using rollback netcode. My hope is that teenagers who makes games in Scratch might like to make games in Easel because it means they can make games they can play…
What I don't understand about your idea is that netcode as far as I understand is way more complex than you make it seem. The netcode for something like a MOBA or MMORPG is drastically different than the netcode for a much simpler turn-based game. They have elaborate synchronization mechanisms to make the illusion not break-down in various ways due to latency. And then even for simple turn-based games, there has to b…
The multiplayer architecture is rollback netcode, which I think is great because it performs client-side prediction by running your full game simulation step forward on the full game state and not some subset or stripped-down version of the simulation or state. It’s the most accurate client-side prediction model you can have.
Then it achieves eventual state synchronisation by rolling back and rerunning your code when there is a prediction error. It performs automatic rubberbanding to smooth over the error corrections. All the bodies get rubberbanded, so as long as you attach your sprites to bodies (which is what you would normally do), the rubberbanding is automatic.
Rollback netcode is underutilised because it’s difficult to implement on other game engines. It requires your entire codebase and all your dependencies to be deterministic and snapshotable, and so if you’re using another engine like Unity or Godot, it won’t meet these requirements because it wasn’t designed for this. Products like Photon Quantum for Unity get around this by reimplementing all the systems like physics, RNGs and even just ArrayLists in their own deterministic manner. But you have to follow all these rules about writing your code in a multiplayer safe way and there are all these traps to avoid.
Putting the rollback netcode into the programming language means it always works 100% of the time and you can’t make a mistake, which means even a beginner programmer on their first day can make a multiplayer game. That was my ultimate goal and I think it is true that if someone who is an experienced programmer wants to hand-code a specialised rollback netcode implementation for their game then they might be able to do a better job on their game. But the defaults are really good when the rollback netcode is in the language itself and I’ve had teenagers make multiplayer games that they play with their whole class and they’ve all had a lot of fun, and that’s the main thing.
projects per year in "Ask HN: What are you working on?" year projects avg/month ---- -------- --------- 2008 123 10.2 2009 104 8.7 2010 666 55.5 2011 86 7.2 2012 46 3.8 2013 389 32.4 2014 146 12.2 2015 104 8.7 2016 471 39.2 2017 456 38.0 2018 242 20.2 2019 246 20.5 2020 882 73.5 2021 837 69.8 2022 363 30.2 2023 425 35.4 2024 2122 176.8 2025 6220 518.3 2026 6093 677.0 (Jan-Sep) Top five topics per year (% of that year…
What's interesting (and kind of sad) here is that while AI drastically increased product count, more than 50% are now composed of dev tools and AI Some kind of golden shovels situation. I wish there were more games, personal apps, and open hardware – more fun stuff that became affordable due to AI
Works nicely with my existing Signoz, costs nothing extra, but I’ve yet to test it against Honeycomb and Clickstack. Apps are in private TestFlight for now.
https://github.com/Lightless-Labs/willikins: a tool to deterministically provision new projects’ git repository, secrets vault / config, build pipelines, etc, based on the project’s needs (cli tool, web service, mobile app,…).
It’s getting tiring to wire all those by hand or with cobbled together script. And I don’t trust agents to give them read and write access to all the resources they’d need to do it. Very much in progress, but it already forced me to clean up some config. I guess it’s like a bad Ansible, for project set up?
I’m also experimenting with a team / project / task-scoped blackboard (https://martinfowler.com/articles/exploring-gen-ai/an-accide...) tool for agents, to be used by multiple users’ agents (eg different devs working on different sides of the same fix or feature), and a self-hosted Q&A forum for agents intended to serve as an authoritative self-grooming source of truth on technical, domain, and business matters for a team or org’s agents.