[Info-vax] x86 Update 4/22/19

[George Cornelius]

In article <qa107q$fdh$2 at gioia.aioe.org>, helbig at asclothestro.multivax.de (Phillip Helbig (undress to reply)) writes:


> Similarly, at some point EDT was VESTed or whatever.  But I recall 
> someone here saying that he had a new version, removing the 255 and 
> 65535 limits.

Had not heard of EDT being VESTed.  It was in Bliss, after all.

PITA on the PDP-11, though, because it was developed in Bliss-36
and then shoehorned into 16-bit address space via a pile of ODL
(overlay description language).  Much more reasonable if you had
11M+ with separated I and D space and memory resident overlays.

There are a number of EDT emulators.  JED comes to mind.  If you
go looking for one, and have an initialization file to set up
customized function keys, you'll probably want to be careful to
find a version with a faithful emulation of line mode and pre-v2
screen mode.

As far as the clever programmer comment is concerned, we are
back to the use of macros.  I believe TECO-11 was converted
from 12-bit (PDP-8) Teco using a lot of macros and clever
programming tricks, and the migration from that one to the
VAX just compounded the issues, partly because the '11 did
not have stack frames.  So the JSB's - lightweight subroutine
calls - and the coroutines may have had their roots in the
conversion to 32 bits.

Pretending to be something pretending to be something
else pretending to be something even further removed
is _hard_.

Running TECO-11 under SIMH would be OK with me, but most
would find the lack of 32-bit registers and the lack of
support for RMS and ODS-2 file naming conventions rather
off-putting.

George

P.S. From the second Google match on "Andy Goldstein teco" (the first had
to do with https://github.com/tesneddon/tecox, teseneddon being of course
Tim E. Sneddon):

 ...TECO-11! " And the TECO-11 code is butt ugly. Looking at it, you can certainly 
 believe that it would have been a brute force translation from PDP-8 code. On 
 and off I am considering if I would rewrite it so that I can at least get it as 
 a split I/D space program... > o. VAX with extra help from Andy Goldstein, and 
 as you rightly stated...

I for one am certainly not going to claim that Andy was too smart for his
(actually meaning our) own good.  I certainly benefitted from all his
efforts in keeping the venerable beast alive.