Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 exptools 1/6/84; site ihuxp.UUCP Path: utzoo!linus!decvax!harpo!whuxlm!whuxl!houxm!ihnp4!ihuxp!gvw1 From: gvw1@ihuxp.UUCP (George V. Wilder) Newsgroups: net.micro.cpm Subject: Eco-C Compiler by Ecosoft, Inc Message-ID: <887@ihuxp.UUCP> Date: Thu, 8-Nov-84 17:38:59 EST Article-I.D.: ihuxp.887 Posted: Thu Nov 8 17:38:59 1984 Date-Received: Sat, 10-Nov-84 03:47:41 EST Organization: AT&T Bell Labs, Naperville, IL Lines: 61 A couple of days ago I made a request to the net for information on the Eco-C Compiler. I'm posting the following response that I received. I find myself usually asking instead of "giving." I hope that the following will be of interest to the net (I have edited out the sender's name). George V. Wilder ----------------------------------------------- In response to your query: Eco-C (ver 1.52) was reviewed in May 1984 Byte (p. 246), with follow-ups and comments in August 1984 (p. 310) and Sept. 1984 (p. 358). The latest version is 3.1, and is substantially improved over 1.52. I was using BDS-C on my Z-80 machine, but wanted a standard C compiler so I could move programs among the machines I work on (an IBM PC-XT, a Sage, and a Vax, besides the Z-80). I purchased Eco-C from the Programmers Shop ($225 including Purdum's "C Programming Guide") about a month ago, and have been satisfied with its performance. It is slow, compared with BDS-C (but then what isn't?) or DeSmet C on the XT, but about the same as Pascal MT+. I suspect it would be speeded up considerably with a hard disk or RAM disk (I have a 4 MHz Z-80 with 2 1MB 8" disks). The syntax of Eco-C is pure K&R - maybe a little too pure, as it complains about spaces between names of functions not yet declared and the opening parenthesis of the argument list. I'm not sure whether K&R really mean this or not (p. 186, paragraph 4, last sentence) so I can't complain too much (I'm going to post a query to get a consensus on net.lang.c). But I've found no flagrant modifications or extensions to the language. The library is good and quite standard. I counted 121 functions, of which 79 were common to Eco-C and Unix C (4.2 BSD), and 67 were common to those two compilers plus DeSmet C. A few minor annoyances: there is no abs, fread, fwrite, fputc, realloc, or rindex (although most of these can be supplied from other functions, such as read or write); some functions have been given different names (e.g. malloc is alloc, rand is irand, log is ln, log10 is log, and pow is power) - these can be fixed with #defines. Source for functions written in C is available for $10 - I haven't bothered yet. The manual is quite good - good reference section on functions, material on assembly language interface. You can't embed assembly code with #asm as in DeSmet C, but it's easy to link in assembly language functions - this is in keeping with the "keep it standard" idea. Overall, I am satisfied. I do most of my development on other machines, then transfer the source and recompile, but if you're patient, you could do all development with Eco-C quite reasonably (I think the relatively slow speed is to be expected for a full compiler on a Z-80 floppy-disk machine). I would recommend Eco-C for a standard C compiler for CP/M. -- George V. Wilder ihnp4!ihuxp!gvw1