[Simh] VUPs?

[Norman Lastovica]

> -----Original Message-----
> From: simh-bounces at trailing-edge.com
[mailto:simh-bounces at trailing-edge.com]
> On Behalf Of John H. Reinhardt
> Sent: Friday, April 04, 2008 9:44 AM
> Cc: simh at trailing-edge.com
> Subject: Re: [Simh] VUPs?
> 
> Norman Lastovica wrote:
> > in at least one case that I can think of, a VAX was purposefully slowed
> > down in the microcode.  that is, at least, the rumor while I worked
there.
> > I'll bet Bob Supnik has some insight and stories along these lines.
> >
> > Care to share Bob?
> >
> 
> I'm not Bob, but from things that I've read I believe that it's the VAX
> 8600 and 8650 that you're thinking of.  The tale is that the 8600
> microcode had a multitude of NOPs which were removed for the 8650 and
> were responsible for a large portion of it's speed increase.  IIRC the
> claim was that the NOPS were used much like NULs after carriage return
> characters in the old teletypes - i.e.  to create a certain amount of
> slack time to allow hardware operations to finish (or something similar).

	The 8650 had a significantly faster clock speed than the 8600
(55ns vs 80ns cycle time).  Perhaps the microcode did get some tweaks
as well.  I believe that at the time the 8600 was supposed to be a flag
ship and I suspect probably didn't have many feet burdened by ball and
chain.
	The story that I am aware of had to do with the 8500.  It was
released at about the same time as the 8700/8800 but needed to be slower 
presumably for marketing reasons.  Supposedly it shared the same hardware 
and clock rate.  I understand that at some point in time there was a new 
version of the microcode code released to fix or enhance something or
another.
Users of the 8500 noticed a performance increase with this new code
that lacked the foot dragging of the prior code.  So it was then given
to all of the 8500 customers as an "upgrade".  I do not remember the
rest of the story though as related to the 8500, 8530 and 8550 (which
from my memory were more or less the exact same computer).  Perhaps
they were basically the same machines with different amounts of streamlined 
microcode?   My memory has been lost to the sands of time.