[Info-vax] Some questions on software for VMS 7.3 VAX

[Stephen Hoffman]

On 2016-01-12 21:56:51 +0000, Bob Koehler said:

> In article <mailman.119.1452616123.14919.info-vax_rbnsn.com at rbnsn.com>, 
> lists at openmailbox.org writes:
>> 
>> I wasn't there so I really don't know.  I thought I read in a recent 
>> post here that VAX was still sold into the early 2000s (2002?). If so I 
>> would have thought there would have been an F90 compiler given the VAX 
>> apparently had very nice floating point hardware support and the amount 
>> of VAX hardware already out there where people didn't want to/need to 
>> migrate to Alpha.

Why would you spend time, effort and money updating an older platform, 
when that budget comes directly out of work on the newer platform?

Among the many issues with VAX, VAX lacked PCSI for OpenVMS 
installations.  VAX lacks a consistent console.  VAX lacks optical 
media requirements, and had way too many distribution media types.   
When the last of the magtape and the TK media finally became 
unavailable, the platform was far too gone to overhaul and upgrade to 
PCSI installs, for that matter.   VAX booted standalone backup from 
tape distro with 9-track and TK50 being the most common, and — other 
than optical, RC25 and maybe one or two other older media types — 
couldn't boot "full" OpenVMS off of distro kits.   VAX didn't use VCG 
for the compilers.   VAX was 32-bit, and folks were increasingly 
slamming into S0/S1 space limits.    You had to reboot the system once 
a year or remember to save the time manually between January and April, 
or the TOY clock would wrap, too.   VAX didn't have IEEE FP, either — 
which is part of why Java never got to VAX.

So... Why mess around with VAX, at the expense of the Alpha platform — 
the platform that DEC was trying to get accepted for Tru64 Unix / 
Digital UNIX / OSF/1, Microsoft Windows and OpenVMS, and was seeking to 
get other vendors — reportedly including Apple and other vendors — to 
use?

>    Long before VAXen stopped being sold, DEC stopped providing any 
> significant software updates.
> 
>    The excuse at the time was a claim that all the customers still on 
> VAX wanted the freeze.

The folks that wanted changes were moving to Alpha, or were porting off 
of VMS.   Those that didn't want to move — same as now — didn't want to 
change anything.

The OpenVMS VAX APIs were effectively frozen at V6.0, at both user-mode 
and kernel-mode.   The kernel-mode freeze was a then-new thing for 
OpenVMS developers.


-- 
Pure Personal Opinion | HoffmanLabs LLC