From: dk4xp@arcor.de   
      
   Am 13.02.26 um 01:09 schrieb john larkin:   
   > On Fri, 13 Feb 2026 00:38:43 +0100, Gerhard Hoffmann    
   > wrote:   
   >   
   >> Am 12.02.26 um 23:12 schrieb john larkin:   
   >>> On Thu, 12 Feb 2026 21:05:53 +0100, Lasse Langwadt    
   >>> wrote:   
   >>>   
   >>>> On 2/12/26 19:19, john larkin wrote:   
   >>>>> The internal stack-oriented architecture of Focal-8 inspired the cool   
   >>>>> PDP-11 architecture which in turn inspired the 68K.   
   >>>>   
   >>>> how did stack-oriented inspire architectures with lots of orthogonal   
   >>>> registers, by showing how it should not be done?   
   >>>>   
   >>>   
   >>> Stack-oriented inspired stack-oriented. In the PDP-11, any of the   
   >>> registers could be used as a stack pointer.   
   >>   
   >> That does not make it stack-oriented. It had registers and nice   
   >> addressing modes with pre-decrement and post increment etc.   
   >> Nice for string handling, too.   
   >   
   > I called it stack oriented!   
   >   
   > A couple of people made real stack machines, hyper-CISC things with   
   > advanced instructions. HP and Intel as I recall. They were pig slow   
   > and didn't last.   
      
   That was the iAPX432. HP's brainchild, Intel prepared the   
   mass production, which never really happened. We had a   
   Multibus board with one 432 in our processor zoo at the Tech Univ   
   Berlin, but nobody adopted it to the point that it did useful   
   work. The entry cost was just too high. With today's cache   
   technology it might make more sense.   
   It had nothing to do with stack machines.   
      
   Western Digital had a stack machine for UCSD Pascal; that   
   was essentially the chip set of the LSI 11/23 with custom   
   microprograms.   
      
   Our above-mentioned PDP11/40e was also microprogrammable   
   (therefore the e suffix). There were only 5 of it's kind   
   in the world; we had 2 of them in the same unibus.   
   Someone at the institute wrote a microprogrammed   
   stack machine for Per Brinch Hansen's Pascal compiler that   
   ran like a scalded dog.   
      
   Luckily for Intel, they did not drop the X86. Internally,   
   a Pentiyummy has nothing in common with x86. There are 100s   
   of registers (called renaming registers) that can hold   
   multiple incarnations of eax, ebx and so on, so that multiple   
   sequential instructions can be executed at the same time, with   
   hardware that checks for dependencies over time.   
      
   When the 200 MHz Pentium appeared, that was the EOL for   
   most Riscs, including Mips, Fairchild Clipper, DEC Alpha,   
   IBM Power PC, 32032 and and and.   
      
   > On the PDP-11 there was a short program to fill every location in   
   > memory with zero, the halt opcode.   
   >   
   > And another program would fill all of memory with the NOP opcode, and   
   > it would cycle on that forever.   
      
   That filled some badly needed voids.   
      
   Gerhard   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)