Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site bonnie.UUCP Path: utzoo!watmath!clyde!bonnie!jww From: jww@bonnie.UUCP (Joel West) Newsgroups: net.micro.mac,net.lang.mod2,net.lang.pascal Subject: Re: SofTech/Volition Modula-2 for Mac? Message-ID: <477@bonnie.UUCP> Date: Tue, 7-May-85 11:52:31 EDT Article-I.D.: bonnie.477 Posted: Tue May 7 11:52:31 1985 Date-Received: Wed, 8-May-85 03:33:42 EDT References: <473@bonnie.UUCP> Organization: CACI -- La Jolla, Calif. Lines: 59 Xref: watmath net.micro.mac:1305 net.lang.mod2:208 net.lang.pascal:289 I have an update to my earlier posting, from fa.info-mac digest #42 > ...[the p-System software]got a reputation from program developers as > being particularly slow in execution, especially on the Apple, which, > some said, "ran slow as molasses." [Goodwin] agrees: "We lost contact > with the customer and input of what they wanted in a product." > But that's all changing, he said: "We're speeding up performance > on the Apple II family by a magnitude of 20 to 30 times." ... > -----------------personal comment-------------------- > Second, the UCSD Pascal for the Mac ("MacAdvantage") benchmarked > in my test between 30-100 as slow as a compiled C. This suggests > that the new UCSD Pascal would be not a p-system at all, but a > fully compiled language. > -- > Joel West (619) 457-9681 Date: 3 May 1985 18:11:29-EDT From: jcr@Mitre-Bedford Subject: Improvements to UCSD Pascal.... After reading the above, I called Softech to check it out. Having used Apple Pascal on the IIe for about a year now, I was pretty excited at the capabilities described in this article. The performance increase described in the article comes through the use of a native code generator (NCG). For those not familiar with the p-system, an NCG is a program that takes the p-code produced by a p-system compiler and translates in into native code for the processor it is to run on. The translation is not complete; some portions of p-code are left intact and still must be interpreted. However, the increase in execution speed is still quite significant. Alas, when I mentioned that I'd heard that an NCG (or some other improvement) might soon be available for the Apple II family, the man at Softech knew immediately what my source was. He said that the reporter owned a IIe and got a little overzealous in his reporting. The fact is that they currently have no plans for an improvement to pascal for the Apple II family, though he said they WERE keeping an ear open, and if demand was high enough, such a product MIGHT possibly be released in the future. However, NCGs are available (or soon will be) for several other processors. It would be interesting to compare the performance of Mac UCSD pascal using an NCG to Mac Object Pascal when both become available. Also, does anyone perhaps have statistics on UCSD pascal with an NCG on the IBM PC as compared to Turbo Pascal? Regards, --- Jeff Rogers jcr@Mitre-Bedford.ARPA -- Joel West (619) 457-9681 CACI, Inc. - Federal 3344 N. Torrey Pines Ct La Jolla 92037 jww@bonnie.UUCP (ihnp4!bonnie!jww) westjw@nosc.ARPA "The best is the enemy of the good" - A. Mullarney