https://gcc.gnu.org/bugzilla/show_bug.cgi?id=77776
--- Comment #27 from g.peterh...@t-online.de --- Hi Matthias, thanks for your benchmark. I still have 2 questions: 1) Accuracy The frexp/ldexp variant seems to be the most accurate; is that correct? Then other constants would have to be used in hypot_gp: scale_up = std::exp2(Type(limits::max_exponent-1)) scale_down = std::exp2(Type(limits::min_exponent-1)) 2) Speed Your benchmark outputs several columns (Δ)Latency/(Δ)Throughput/Speedup. What exactly do the values stand for; what should be optimized for? thx Gero