Re: Risc V greatly underperforms

On Mon, Sep 20, 2021 at 12:07 PM Vincent Lefevre wrote: [...] > And even the 6502 had an ADC instruction with carry-in and carry-out. > This is the most logical and intuitive behavior, as from a hardware > point of view, the carry-in and carry-out are handled *for free*. This is not completely

Re: hgcd1/2

On Tue, Sep 3, 2019 at 10:49 AM Torbjörn Granlund wrote: > > ni...@lysator.liu.se (Niels Möller) writes: > > And this on a laptop with an Intel U4100 (5 years old?), so I'd assume > it doesn't have a particularly fast div instruction. Should we just > delete div1 ? On which architectures

Re: nth_prime

On Tue, May 21, 2019 at 8:49 PM Niels Möller wrote: > > Seth Troisi writes: > > > 3. Use Legendre's Formula (github > > ) > > Any reference for this method? I've never read anything about efficient > computation of the nth prime. A web search turns up