Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!linus!decvax!harpo!seismo!hao!hplabs!sri-unix!broome@brl-bmd From: broome@brl-bmd@sri-unix.UUCP Newsgroups: net.micro Subject: Re: Lisp for CP/M?? Message-ID: <4883@sri-arpa.UUCP> Date: Mon, 5-Sep-83 20:50:41 EDT Article-I.D.: sri-arpa.4883 Posted: Mon Sep 5 20:50:41 1983 Date-Received: Thu, 8-Sep-83 12:45:21 EDT Lines: 35 From: Paul Broome (CTAB) The Little Big LISP system is mentioned in an article by John Fitch and Jed Marti, "NLARGEing a Z80 Microprocessor." This paper is in Computer Algebra, Proc. EUROCAM '82, LNCS #144. All I know about it is in the paper but from what I see it sounds impressive. The Little Big LISP system is implemented for Z80s under TRS-80 DOS or CP/M. Modules available are a trace package, a simple editor, vector and matrix packages, a META translator system (?), a compiler, an RLISP parser and the NLARGE algebra system. NLARGE also looks very interesting, quoting from the paper, "NLARGE is a rational polynomial algebra system written in RLISP, providing multivariate polynomials and with greatest common divisor (gcd) capabilities between such polynomials. The operations of addition, subtraction, multiplication, division, differentiation and restricted integration are provided. ... The programming style of implementation is one of function application, with minimal use of assignment and sequential operations. ... The source is 500 lines of RLISP, defining 43 functions, which compiles into approximately 11,000 bytes as a FAP (fast load) file. ... it is quite possible to develop, translate into LISP and recompile the system with no other computer support than a Z80 with floppy disks, as indeed has been done." There are some comparisons with muMATH-79, written in muLISP. (muLISP is interpreted.) Two expressions were expanded: (1+x)^20 and (x1+x2+ .. +x9)^2 both of which took about a minute on a 48Kb computer for muMATH. (This might have been an 8080.) Anyway using the 4 mh Z80A, 48 Kb, the equivalent times for NLARGE were 9 seconds and 3 seconds. Anybody know more? Paul Broome broome@brl