Hacker News Reader

In the first several examples, the input seems to have a mantissa with all zeros beyond the first 15 bits. Why isn't the author using float32 for this instead of float64?

View on HN

Not having to make that kind of decision in the first place is inherently valuable in of itself, which I believe is what's being showcased here.

I guess, but most compression algorithms like DEFLATE or zstd would compress those big sequences of zeros basically to nothing.

Reply on news.ycombinator.com