This is pretty interesting. I've seen a similar technique for handling a single time series when the number of data points far exceeds the number of pixels, doing a two-dimensional histogram rather than connecting subsequent dots with line segments. I hadn't thought about how to do that for the case of having an extremely large number of time series overlaid on each other. This approach seems to produce results similar to what you'd see on an oscilloscope or an eye diagram.

Author of this paper here. Cool to see this show up on HN. Little story about this paper: Danyel and I did this while I was on an internship at Microsoft Research. The algorithm was too simple to get it published at a conference, but the technique has been adopted in a few places now, including other papers. We did some follow-up work to more efficiently compute the density and also support interactive filtering. See https://www.domoritz.de/papers/2021-KDBox-VIS.pdf.