Re: The joy of FORTRAN

John Levine <johnl@taugh.com> wrote:

It appears that Scott Lurndal <slp53@pacbell.net> said:

John Ames <commodorejohn@gmail.com> writes:

On Tue, 25 Feb 2025 13:43:06 -0700
Peter Flass <peter_flass@yahoo.com> wrote:
 

Assembler and COBOL were needed but avoided 

 
Assembler can be fun.

 
Especially PDP-10 assembler. Wonderful instruction set.

 
Need to play around with -10 assembler sometime.*

 
Take a look at HAKMEM, which includes some impressive snippets of PDP-10 assembler.
 
https://www.inwap.com/pdp10/hbaker/hakmem/algorithms.html
 

From what I've seen of the PDP-10, the VAX-11 instruction set
was superior.

 
It's not a meaningful comparison. The PDP-6, on which the -10 was based, was
designed in 1963 in an era when each transistor was in a separate can and 64K 36
bit words of hand-strung core was a whole lot of memory. It got excellent
performance out of a design that was practical to build in that era. It
addressed 36 bit words because most of the scientific machines in that era did,
since the IBM 704. In that expensive memory era, the way the -10 packed five
7-bit ASCII characters into a word with only one wasted bit was pretty slick.
These days we think that all bytes are 8 bits but in that era 6 or 7 or 9 were
all common. I don't ever recall writing code on a PDP-10 that used 8 bit bytes
other than maybe unpacking magtapes from IBM systems.

Multics used 8-bit bytes on a 36-bit Honeywell 6000. The developers tried
to discourage the use of 9-bit bytes - I was told that “nothing” used them,
but I found out WordPerfect used them for metacharacters.
 

 
The VAX was developed over a decade later, when they put thousands of
transistors on each logic chip and thousands of bits in each memory chip. It
suffered from a severe case of second system syndrome, where they started from
the elegant PDP-11 and added every feature a programmer could ever possibly
want, with less than fabulous performance to match. There's a reason that the
VAX inspired RISC systems.
 

--
Pete