What is your bet in 2025?
What tech, product, or project are you betting on for 2025?
This player has its own OS, no WiFi or only for firmware upgrades. Has no apps and boots up under three seconds and large capacity from SD Card. The best is it just plays.
What's even cooler is that it's a DAC so I can have my headphones plugged in to it and then use Spotify/Service to stream to the player.
Expensive though at $1000 but just how I want it. My headphones decoupled from my mobile phone.
Not so serious ATM but looks promising so far.
Unfortunately all official sipeed 3" touch displays are sold out, probably i need to go I2C or DPI...
We're just scratching the surface on Generative AI and I'm already tired of it. I got a tech book that was obviously written with the aid of ChatGPT.
I am favoring old non-fiction books for my reading pleasure - the ones that were bourne from human toil.
- Retro game engines
- Doom engine
- Wolfenstein engine
- Raycasting
- Console/PC emulators getting more popular
- New search engines doing okay (Kagi etc)
- Linux distros
- Programming
- Anything low-level
- C , C#, Rust
- Anything without the term "AI" in it
- Retro phonesI think they will start migrating to lower level programming (to escape AI to an extent), and the most enjoyable lower level programming projects are:
- Writing your own language
- Retro game emulators
- Operating Systems
- Graphics programming
shameless plug: if you like retro-games I'm working on https://pixelbrawlgames.com
A hallucination could not just be career/license ending but land the nurse in jail.
Even with 100% accuracy, is not entirely obvious to me how a nurse could leverage a language model to make their job that much easier.
A language model is not going to bath or dress someone.
Once upon a time Assembly developers thought optimizing compilers would never be good enough.
That’s why I’m putting my bet on building a product that helps manage it all better called Saner.AI
Like Faraday's first electric motor, it'll have no practical use, at first. If I'm right, it'll be somewhat useful in reducing the power consumption of AI and let us ride Moore's law for a few more decades.
The guts of an FPGA are a lot of D-flip-flops in chains, that shift in the bitstream, and then send control signals to LUTs, switching fabric, etc. What I propose is to remove most of the switching fabric, so that every cell is identical, and add latching to every cell's output, to eliminate timing (especially race conditions) as a concern.
Every cell only has short connections to neighbors, so the capacitance of high speed lines that cause such grief in large VLSI designs is mostly removed (you do still need to clock those latches, so those would take some good drive signal).