[Info-vax] Chinese Alpha?

[David Froble]

Bob Koehler wrote:
> In article <jnq1mq$no$1 at solani.org>, Michael Kraemer <M.Kraemer at gsi.de> writes:
>> I don't get it what magical design features the Alpha has, which
>> would make it still attractive, even if it had been developed further.
>> It's a design more than twenty years old, patents would have expired,
>> and even by DEC's exaggerated claims it would reach its predicted
>> lifetime in five years from now.
>>
> 
>    DEC's claims were based mostly on the state of the art when Alpha was
>    new.  By the time alpha was abandoned DEC was putting things into
>    Alpha that I don't think DEC had considered when it was new.

When the 11/780 was introduced, I think the CPU was 4 large boards plugged into the 
backplane.  2 MB of memory worked.  The microprocessor, CPU on one chip, really changed 
things.  It was basically miniaturization.  Even though much smaller, it was much more 
powerful.  Shorter data paths with a constant on the speed of electricity is a big gain. 
Still is in today's continued process shrinks.

The continued shrinking of the die size allowed additional things to be placed on the 
chip, instead of on the motherboard, or additional boards, and finally to multiple cores 
on one chip.  DEC lead some of this with the on-chip memory controllers, inter-processor 
communications and such.  Then AMD used some of this technology and ended up dragging 
Intel (relatively) out of the stone age.

The things added to the basic CPU were not Alpha specific, though the Alpha developers 
seemed to "get it" ahead of others.  Most likely none of this was imagined when the Alpha 
design was implemented.  Such is progress.

The original 25 year target could not imagine what might be developed during that time. 
Some of those developments could affect the validity of the 25 year target, for better or 
worse.  For example, could an OoO design run in 22 nm process?  Got no idea.  Don't even 
know what I'm talking about.