bwhite@inmet.inmet.com (02/16/91)
I am looking at models of floating point numbers. There are apparently three candidates, or at least three that I have looked at: 1.) The Brown model (Brown, W.S., A Simple but realistic model of floating point computation, ACM TOMS, 1981) 2.) The Ada Manual's version of the Brown Model (Ada Language Reference Manual, sections 3.5.6, 3.5.8, various places in 4.5 and subsections.) 3.) The LCAS model (Payne, Mary, Schaffert, Craig, and Wichmann, Brian, Proposal for a Language Compatible Arithmetic Standard, SIGPLAN Notices, Jan 1991) I am looking for citations and references of analyses of numerical algorithms and software using any of these. Particularly interesting is the first and third. If anyone has succeeded in doing sensible error analysis using the second, or has published a failed attempt along with the reasons for the failure, that is interesting also. I have a copy of the IEEE floating point standard, as well as other standards for particular hardware implementations of floating point numbers. These standards, while fine and useful documents, are not at the level of detail I am interested in. Thanks in advance. Peace, Bill White <bwhite@inmet.inmet.com>