Redish Labadredish@neuromatch.social
Aug 17, 2026, 4:49 PMYou laugh, but I had a project where we were doing early wireless neural transmission and discovered (to our surprise) that a lot of the reported bandwidths were based on (a) massive compression locally, which was very hard to do at the time, and (b) different transmission sizes for 1s and 0s. If I remember right, 0s were shorter. So, even if they didn't weigh more, they were shorter to say.
The upshot was that the reported bandwidths couldn't actually handle neural data at the time, even thought it had sounded like they could.