X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f From: Rugxulo Newsgroups: comp.os.msdos.djgpp Subject: "-mtune=pentium" still default?? Date: Mon, 16 Jun 2008 18:56:25 -0700 (PDT) Organization: http://groups.google.com Lines: 13 Message-ID: <1bdbff80-165b-467f-acfd-51cd36bd94c0@34g2000hsf.googlegroups.com> NNTP-Posting-Host: 65.13.115.246 Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-Trace: posting.google.com 1213667786 13737 127.0.0.1 (17 Jun 2008 01:56:26 GMT) X-Complaints-To: groups-abuse AT google DOT com NNTP-Posting-Date: Tue, 17 Jun 2008 01:56:26 +0000 (UTC) Complaints-To: groups-abuse AT google DOT com Injection-Info: 34g2000hsf.googlegroups.com; posting-host=65.13.115.246; posting-account=p5rsXQoAAAB8KPnVlgg9E_vlm2dvVhfO User-Agent: G2/1.0 X-HTTP-UserAgent: Opera/9.50 (Windows NT 6.0; U; en),gzip(gfe),gzip(gfe) To: djgpp AT delorie DOT com DJ-Gateway: from newsgroup comp.os.msdos.djgpp Reply-To: djgpp AT delorie DOT com Okay, first of all, I still have a Pentium I (166 Mhz, no MMX) that I run stuff on occasionally, but I have to admit that such isn't as common as it used to be. By now, most people have other, newer computers (even me). But even DJGPP's latest version of GCC (4.2.3) still uses "-mtune=pentium" by default. I think this is slightly less wise than it would be if more people used such machines. Anyways, not a big deal. BUT, then it seems that *compiling* with "-mtune=i686" is actually faster (!) than with i586, i486, or i386. Granted, I didn't test on too many files, and it may be cpu dependent (although I did test on a few different machines), but it *seemed* to be the case (independent benchmarks welcome). So, that's my take, and those are two reasons why I think DJGPP should use "-mtune=i686" by default in later versions.