@geerlingguy@mastodon.social
The RAM shortage comes for us all: https://www.youtube.com/watch?v=9rbz0akyLyQ
The RAM shortage comes for us all: https://www.youtube.com/watch?v=9rbz0akyLyQ
clippy appreciation post
Anyone trying to position Meshtastic as a secure protocol is nuts.
The profile poisoning but is alarming, but I never assume anything broadcast over LoRa to be secure. It's like Twitter, Bsky, FB, etc. don't ever assume a DM is secure.
https://partyon.xyz/@nullagent/115005493598214309
Also in tandem with my Framework Mainboard cluster testing, I've released a new tool for testing open source clustered AI solutions like Exo, llama.cpp RPC, and distributed-llama: https://github.com/geerlingguy/beowulf-ai-cluster
I tested four Framework Desktop Mainboards with AMD Ryzen AI Max+ 395 (what a mouthful!) in a private AI cluster.
I've published ALL my test results (including testing the new gpt-oss-20b and 120b models, as well as giant LLMs across the full cluster), links are in the video description: https://www.youtube.com/watch?v=N5xhOqlvRh4
Having a 5 Gbps NIC using standard cat6e cabling is... actually really nice, not gonna lie. A lot of media can't even keep up at ~400 MB/s.
10 Gbps has still taken so long to trickle down, and has so many caveats, maybe 5 Gbps will be the new standard in a few years?