[Info-vax] Why it is a good idea that OpenVMS isn't on x86-64 just yet

[Johnny Billquist]

On 2016-01-18 23:16, Norm Raphael wrote:
>> On 01/18/16, John Reagan via Info-vax<info-vax at info-vax.com> wrote:
>   >
>> We distributed compilers on TU58 as well as 9-track so I had to make & verify the kits on the console.
>>
> Yes, I remember a COBOL release on TU58 that had a saveset split across 2 cartridges and the kitinstal.com
> did not handle that correctly so I had to copy the savesets onto disk and fix the kitinstal.com and rebuild the
> TU58's  (I believe that was using VMSINSTAL.COM Options G).  I know, at that point I could - and did - do
> the install from the disk, but I had to put the kit back together in case the next fool had to use it on another
> machine, and because I did, that was never necessary - if I had not, as sure as shootin' it would have been.
>
> So now all these years later I get to find out who never tested that release on TU58's.

:-)

> Oh and because of the VMSClustering, we had to use the TU58 to boot (load the microcode IIRC). S-L-O-W.

That makes sense (unfortunately). The VAX-11/750 was/is rather special 
when it comes to booting. Just about the only VAX I know of that don't 
have anything even closely related to VMB involved. Instead you have 
simple bootstrap roms that just read the boot block from disk (just like 
a PDP-11). Unfortunately, there was only four boot roms/routines, and a 
selector at the front to select which one to boot from. And I think the 
selection was/is pretty fixed. And CI is not one of the options. So I 
guess that meant in this case the TU58 was "it".

	Johnny

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