Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site rtech.UUCP Path: utzoo!watmath!clyde!burl!ulysses!allegra!mit-eddie!think!harvard!seismo!lll-crg!dual!unisoft!mtxinu!rtech!daveb From: daveb@rtech.UUCP (Dave Brower) Newsgroups: net.lang.c Subject: Re: This Sentence is False (DEC-20) Message-ID: <659@rtech.UUCP> Date: Thu, 26-Sep-85 02:19:05 EDT Article-I.D.: rtech.659 Posted: Thu Sep 26 02:19:05 1985 Date-Received: Sun, 29-Sep-85 08:01:19 EDT References: <1118@brl-tgr.ARPA> <139200010@uiucdcsb> <188@graffiti.UUCP> <314@ihdev.UUCP> Organization: Relational Technology, Alameda CA Lines: 19 > On a Dec-20, you have a huge number of strange instructions, including : ... > skipn : skip never > jumpn : jump never I always assumed these instructions 'existed' out of dedication to orthogonality in the instruction set. The DEC 10/20 seemed to have about a bizzilion instructions that boiled down to NOP. The advantage was that it was easy to learn the mnemonics, since they were all constructed in a regular fashion. It was also very easy to create and examine object code: all the instructions are the same size, and the scheme for modifying operand types is also very regular. Stanford had an ungodly quick one-pass assembler called FAIL that worked because it always knew how big an instruction was (one 36 bit word), and because it had the linker try to resolve all of the forward references. -- {amdahl|dual|sun|zehntel}\ |"If his brains ran down, how could {ucbvax|decvax}!mtxinu---->!rtech!daveb |he talk?" ihnp4!{phoenix|amdahl}___/ |"Happens to people all the time...."