Hacker News Reader

Such a good writer. His Wolfenstein book had the best explanation I’ve ever seen for how floating point numbers are encoded, and that explanation is in the first several pages. If you don’t have an understanding of floating point encoding, that alone is worth the purchase price.

View on HN

Here the "floating point" excerpt from the book: http://fabiensanglard.net/floating_point_visually_explained/...

Am I alone in thinking that the sign, exponent and mantissa model makes more sense, and that (-1)^S * 1.M * 2^(E-127) isn't that hard to understand? Maybe because I already understand that model, but his explanation just made it more confusing for me.

Reply on news.ycombinator.com