Path: utzoo!news-server.csri.toronto.edu!rutgers!att!pacbell.com!ucsd!sdcc6!sdcc13!bdietz From: bdietz@sdcc13.ucsd.edu (Jack Dietz) Newsgroups: sci.space.shuttle Subject: Re: New Shuttle Computers Summary: Shuttle language, address space Message-ID: <17412@sdcc6.ucsd.edu> Date: 12 Mar 91 11:34:56 GMT References: <1991Mar10.164459.5216@rodan.acs.syr.edu> <1991Mar10.210059.26743@nowhere.uucp> <1991Mar11.200136.7329@casbah.acns.nwu.edu> Sender: news@sdcc6.ucsd.edu Organization: University of California, San Diego Lines: 31 As I recall from the very early flights and a dog-eared copy of the Space Shuttle Operator's Manual, the original shuttle programs were written in Very High-level Shuttle Instruction Code (VHSIC). This came up during those T-00:31 launch aborts when the announcers were trying to cover programmers frantically debugging on national television... Also, in the instructions from the SSOM, several times during the launch sequence the instructions tell the commander to punch in a command to load another software package. [OPS] 105, for example, loaded the package used between T-19 minutes and T-5 minutes, I believe. A suggestion: If the instruction space is limited to 128K words or so, bank switching sounds like an answer -- copying the routines directly to ROM images with an added instruction to switch banks at the end of each. Howabout a milspec SPARC processor? (just kidding) Actually, the XT-370 computers which IBM made for a while used a 68000 with the microcode replaced to run 370 programs on an expander card. Perhaps NASA could convince a couple of companies to update that chip in a radiation-hardened manner? -- The Army Axiom: | Jack Dietz (bdietz@ucsd.edu) Any order that can be misunderstood | Looking Forward to a Future has been misunderstood. | without Bellamyites -- Segal's Law: | Jack Dietz A man with one watch always knows what time it is. | (bdietz@ucsd.edu) A man with two watches is never sure. | Engineer-To-Be