[log in to unmask] wrote:
: Item Subject: Re: PA-8000 chip / 64-bit Splash
 
[good stuff snipped]
 
: The ultimate would be for HP to provide 64bit C, Pascal, and COBOL
: compilers, recompile all of MPE/iX into 64bit code, provide 64bit
: APIs to replace or extend all the current intrinsics, and adopt some
: future 64bit posix standard to replace the current posix interfaces.
: This would be a "true" 64 bit MPE.  This will *NEVER* *HAPPEN* for a
: number of reasons, perhaps the biggest being that we DON'T WANT IT
: to happen.  Why you ask?  Because it is way too much work and we want
: HP to be working on something more important, i.e.:
 
: The HP/Intel architecture is coming.  It is not PA-RISC 2.0.  It's not
: even PA-RISC 3.0.  It's not even RISC!  It will be as different from
: PA-RISC as PA-RISC was from the classic (MPE/V) HP3000 architecture.
: PA-RISC 2.0 is new, and it's very cool, but it's the end of the line
: for PA-RISC.  If we make HP expend all their energy on 64bit PA-RISC
: MPE, then they probably will not be able to afford to do what we really
: want which is 64bit MPE on the HP/Intel architecture.  If this happens,
: then MPE will die along with PA-RISC in a few years.  If, however, we
: let HP do the *minimum* acceptable work to simply get MPE running on
: PA-RISC 2.0 hardware (so that MPE systems can keep up with the industry
: performance curve for the next couple years), then there is a chance
: that they will be able to scrape together enough resources to produce
: a version of MPE that runs in whatever passes for "Native Mode" on the
: HP/Intel systems, and is a true 64 bit MPE.  If THIS happens, then the
: life of MPE would probably be insured for another 10 years.  For this
: to happen though, HP will probably have to start expending all their
: energy on these projects pretty soon.  This means that we will have to
: give up virtually all other enhancements, probably for two years or so.
: I think that this is something that we *must* do if we want MPE to
: survive.
 
I totally agree with this.  For MPE to go to new HP/Intel architecture
is a massive investment.  If you want this, you need to get the MPE lab
moving NOW.  The problem for CSY is: making a business case for it.
 
Rick Ehrhart
[log in to unmask]                     408-553-3776