🚀 go-pugleaf

On 2018-08-16 10:41, Questor wrote:
> On Mon, 13 Aug 2018 12:36:49 +0200, Johnny Billquist <b...@softjar.se> wrote:
>> On 2018-08-13 09:56, Questor wrote:
>>> On Sun, 12 Aug 2018 15:20:18 -0000 (UTC), John Levine <jo...@taugh.com> wrote:
>>>> In article <0897e059-0b26-4334...@googlegroups.com>,
>>>> Quadibloc  <jsa...@ecn.ab.ca> wrote:
>>>>>>> sorry when it was canceled, although I suspect it would have been difficult
>>>>>>> to expand the architecture to larger memory sizes.
>>>>>
>>>>>> We don't have to guess.  They did expand the addressing for the DEC-20
>>>>>> and it was pretty ugly. ...
>>>>
>>>>> I think it was ugly too; but I suppose that it could still be said that if the
>>>>> memory expansion method actually used was a kludge, that doesn't prove th at
>>>>> something better might not be possible.
>>>>
>>>> The people who did it were pretty smart.  An obvious alternative would
>>>> be a mode bit, like 16 vs 32 bit segments on the 386, and different
>>>> instruction formats, but if it's not backward compatible with 18 bit
>>>> code, you might as well switch to a Vax, which is what everyone did.
>>>>
>>>> The Vax was a mistake, too, beautifully optimized for a hardware and
>>>> software environment that was immediately obsolete, but that's a
>>>> different argument.
>>>
>>> Possibly a result of trying to maintain a bridge to the PDP-11 world.
>>
>> I'd disagree. The reason for most of the things in VAX was a result of
>> memory being expensive, and people programming in assembler. So, you
>> optimized the machine for being able to write programs that used as
>> little memory as possible. There are lots of things in the VAX that
>> comes from this. And it's definitely targeted at people still writing
>> programs in assembler.
> 
> I can't speak to the intentions of the designers of the VAX instruction set, but
> what you said isn't reflected in DEC's own practices.  Most of VMS and its
> system software was written in BLISS.  And certainly by the early 1980s, the
> overwhelming majority of programmers in Spit Brook (home of the VMS software
> groups(s)) were BLISS programmers.  Lots of bits were spent in discussions on
> internal message boards about all manner of BLISS arcana; VAX assembly, not so
> much.

True. VMS is a lot of BLISS, and quite some MACRO, and some sprinkling 
of other languages.
So, DEC at least partially did not so as much assembler as was probably 
expected when the design was made.
There is an awful lot of MACRO in there though, which forces every port 
to this day to first of all create a compiler for the new target, for 
MACRO, which is too much to rewrite in some other language.

However, I suspect that with regards to memory use, they probably also 
anticipated that a compact machine code would also be beneficial for 
code generated from compilers, if the compilers were just clever enough. 
Memory prices were definitely a factor no matter what language you might 
be writing in.

>> But, as mentioned, this was just at the time when this was all becoming
>> obsolete. Memory prices continued to drop, while sized grew. Which meant
>> that the whole effort at being memory efficient was moot. And people
>> started writing less and less in assembler. Instead it started becoming
>> more relevant how to make things faster and simpler, memory be damned.
> 
> Coming to the VAX from the PDP-10 world, my own view was that they were
> influenced by what seemed like "big" from the PDP-11 point of view.  DEC
> certainly had a kind of near-sightedness when it came to imagining the future of
> computing which proved fatal over the long term.  That myopia was a contributing
> factor to the failure of the Jupiter.  Much of this was hashed out and discussed
> ad nausem in alt.sys.pdp10 about fifteen years ago, as you well know.

Yup. :-)

   Johnny

-- 
Johnny Billquist                  || "I'm on a bus
                                   ||  on a psychedelic trip
email: b...@softjar.se             ||  Reading murder books
pdp is alive!                     ||  tryin' to stay hip" - B. Idol