Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!cs.utexas.edu!usc!orion.oac.uci.edu!easu021 From: easu021@orion.oac.uci.edu (Jason Goldberg) Newsgroups: comp.sys.amiga Subject: Re: Manx C 5.0 Keywords: Manx C 5.0 update Message-ID: <3821@orion.cf.uci.edu> Date: 17 Dec 89 08:13:30 GMT References: <871@trlluna.trl.oz> <1989Dec16.094755.22491@csusac.csus.edu> Reply-To: easu021@orion.oac.uci.edu (Jason Goldberg) Organization: University of California, Irvine Lines: 59 The following is a reply to my request about Manx 5.0 on the Manx support BBS. Mike Spille is not only the SYSOP of the Manx BBS but also incharge or tech support for MANX. Bassically I complained about their upgrade policies and asked about Manx 5.0. =============================================================================== Message 1778 DATE/TIME: 12-13-89 1:44am From : MIKE SPILLE (SYSOP) -- PRIVATE -- To : JASON GOLDBERG Subject: Comment @ 12-13-89 12:12am Folder : Z, "Comments to the Sysop" I'm afraid I have to agree with you that our upgrade policies in the past have been atrocious. However, for the past 6-months or so we've been in the process of revamping our customer database (as well as the general update procedures). In the future you should receive regular update notices w/out a problem. If you have a problem where you believe you are due an update (you implied that you had paid for a year of upgrades and never got any), then you should definitely call our update dept. and complain. Their number is (201) 389-0290. The two major benefits of upgrading to our 5.0 version, in my opinion, would be ANSI compatibility and improved performance. If you don't think ANSI can help you, even for writing "machine-specific" code, then I think you'll be pleasantly surprised when you start using an ANSI greatly enhance compile-time error checking w/out tying the programmer down or sacrificing performance. And if you ever do have to port any of your code, ANSI makes it that much easier. The other main advantage of 5.0 is performance - both compile time and run time. At compile time, 5.0 is roughly 25% faster than 3.6a, even though additional features have been added to the compiler. This translates into something like a 10-fold improvement over Lattice 5.04 (I'm not kidding - our beta testers have had compile times of 10 minutes for us vs. over an hour on Lattice w/ very large source files!). In addition, many optimizations have been made to the code generated by the compiler which make it run much faster than that made by 3.6a. Overall, we're roughly equivalent w/ Lattice in the arena of code generation, with our compiler doing better in some areas, and Lattice better in others. I don't have 'official' benchmarks at this time, since the compiler is still officially in beta test (we don't believe in releasing benchmarks for beta products, unlike some other amiga compiler companies...). However, the final release is only one or two weeks away (we're really waiting for our printer to finish the new documentation - once they give it back to us, we'll be ready to ship), and then I'll be able to -Mike ---------------------------------------- Mike seems like he is being straight with me and is obviously knowlagable about the product. He is also sincerly interested in helping with any Manx related problems when he can. -Jason-