XPost: alt.folklore.computers   
   From: anton@mips.complang.tuwien.ac.at   
      
   Michael S  writes:   
   >On Mon, 4 Aug 2025 18:16:45 -0000 (UTC)   
   >Thomas Koenig  wrote:   
   >   
   >> Anton Ertl  schrieb:   
   >>   
   >> > The claim by John Savard was that the VAX "was a good match to the   
   >> > technology *of its time*".  It was not.  It may have been a good   
   >> > match for the beliefs of the time, but that's a different thing.   
   >>   
   >   
   >The evidence of 801 is the 801 did not deliver until more than decade   
   >later. And the variant that delivered was quite different from original   
   >801.   
   >Actually, it can be argued that 801 didn't deliver until more than 15   
   >years late.   
      
   Maybe for IBM.  IBM had its successful S/370 business, and no real   
   need for the IBM 801 after the telephone switch project for which it   
   was originally developed had been canceled, so they had no hurry in   
   productizing it.   says:   
      
   |The architectural work on the ROMP began in late spring of 1977, as a   
   |spin-off of IBM Research's 801 RISC processor (hence the "Research"   
   |in the acronym). Most of the architectural changes were for cost   
   |reduction, such as adding 16-bit instructions for   
   |byte-efficiency. [...]   
   |   
   |The first chips were ready in early 1981 [...] ROMP first appeared in   
   |a commercial product as the processor for the IBM RT PC workstation,   
   |which was introduced in 1986. To provide examples for RT PC   
   |production, volume production of the ROMP and its MMU began in   
   |1985. The delay between the completion of the ROMP design, and   
   |introduction of the RT PC was caused by overly ambitious software   
   |plans for the RT PC and its operating system (OS).   
      
   If IBM had been in a hurry to introduce ROMP, they would have had a   
   contingency plan for the RT PC system software.   
      
   For comparison:   
      
   HPPA: "In early 1982, work on the Precision Architecture began at HP   
   Laboratories, defining the instruction set and virtual memory   
   system. Development of the first TTL implementation started in April   
   1983. With simulation of the processor having completed in 1983, a   
   final processor design was delivered to software developers in July   
   1984. Systems prototyping followed, with "lab prototypes" being   
   produced in 1985 and product prototypes in 1986.  The first processors   
   were introduced in products during 1986, with the first HP 9000 Series   
   840 units shipping in November of that year."   
      
      
   MIPS: Inspired by IBM 801, Stanford MIPS research project 1981-1984,   
   1984 MIPS Inc, R2000 and R2010 (FP) introduced May 1986 (12.5MHz), and   
   according to   
    MIPS   
   delivered a workstation in the same year.   
      
   SPARC: Berkeley RISC research project between 1980 and 1984;   
    does not mention the IBM   
   801 as inspiration, but a 1978 paper by Tanenbaum.  Samples for RISC-I   
   in May 1982 (but could only run at 0.5MHz).  No date for the   
   completion of RISC-II, but given that the research project ended in   
   1984, it was probably at that time.  Sun developed Berkeley RISC into   
   SPARC, and the first SPARC machine, the Sun-4/260 appeared in July   
   1987 with a 16.67MHz processor.   
      
   ARM: Inspired by Berkeley RISC, "Acorn initiated its RISC research   
   project in October 1983"   
      
   "The first samples of ARM silicon worked properly when first received   
   and tested on 26 April 1985.  Known as ARM1, these versions ran at 6   
   MHz.[...] late 1986 introduction of the ARM2 design running at 8 MHz   
   [...]  Acorn Archimedes personal computer models A305, A310, and A440,   
   launched on the 6th June 1987."   
    Note   
   that the Acorn people originally were not computer architects or   
   circuit designers.  ARM1 and ARM2 did not include an MMU, cache   
   controller, or FPU, however.   
      
   There are examples of Motorola (88000, 1988), Intel (i960, 1988), IBM   
   (RS/6000, 1990), and DEC (Alpha, 1992) which had successful   
   established architectures, and that caused the problem of how to place   
   the RISC architecture in the market, and a certain lack of urgency.   
   Read up on the individual architectures and their predecessors to   
   learn about the individual causes for delays (there's not much in   
   Wikipedia about the development of the 88000, however).   
      
   HP might have been in the same camp, but apparently someone high up at   
   HP decided to replace all their existing architectures with RISC ASAP,   
   and they succeeded.   
      
   In any case, RISCs delivered, starting in 1986.  There is no reason   
   they could not have delivered earlier.   
      
      
   - anton   
   --   
   'Anyone trying for "industrial quality" ISA should avoid undefined behavior.'   
     Mitch Alsup,    
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)