IRC channel logs

<oriansj>muurkha: completely valid point.

<oriansj>My method was going to https://cdn.netbsd.org/pub/pkgsrc/packages/NetBSD/earmv7hf/10.0/All/ and https://cdn.netbsd.org/pub/pkgsrc/packages/NetBSD/vax/10.0/All/ and just comparing a handful of packages to get an approximate value

<muurkha>oriansj: I think it's plausible that VAX would win in all of these for handwritten assembly

<oriansj>muurkha: well; when it comes to handwritten assembly by people who know what they are doing, VAX should win by a good margin but in terms of compiler generated assembly, it just depends on how much time was spent including optimizations for the given architecture and entirely explains i386's position

<muurkha>plausibly, but I don't know if you're right

<muurkha>also though there are some kinds oof instructions that are much more difficult for a compiler to take advantage of and others that are much easier

IRC channel logs

2023-07-06.log