On Wed, 2003-01-29 at 04:56, Nicholas Clark wrote: > On Wed, Jan 29, 2003 at 06:53:25AM +0100, Leopold Toetsch wrote: > > Garrett Goebel wrote: > > > Many thanks for the links (searching for this stuff is a pain, there are > > too many results ;-) > > > > Anyway: > > - are there platforms with quad precision floats out there? > > sparc Solaris and Irix both have 16 byte doubles. I think that they are > both done in software, and I don't think either is to any IEEE standard.
Actually, Sparc and MIPS are pretty good about adherence. (There may be some breakdown in the specialized representation cases, as well as underflow representation.) In fact, most hardware these days use IEEE (or EEEI) representation. VAXen and CRU-based Crays are a brash exception. > > > - or should 12 byte long doubles get converted to 8 byte IEEE doubles. > > This doesn't feel like a good idea. It may be the only practical solution, > but I doesn't mean that I have to like it. Let me dig through my notes. -- Bryan C. Warnock bwarnock@(gtemail.net|raba.com)
