[Info-vax] RealWorldTech on Poulson

[John Wallace]

On Jul 4, 3:08 pm, Paul Sture <paul.nos... at sture.ch> wrote:
> In article <iusece$76... at Iltempo.Update.UU.SE>,
>  Johnny Billquist <b... at softjar.se> wrote:
>
>
>
> > On 2011-07-04 09.22, Michael Kraemer wrote:
> > > Jan-Erik Soderholm schrieb:
>
> > >> About geometry shrinks. Not that it realy matters, but... :-)
>
> > >> I just checked these two pages :
> > >>http://en.wikipedia.org/wiki/DEC_Alpha
> > >>http://en.wikipedia.org/wiki/Itanium
>
> > >> It's interesting to compare the last Alpha (EV7z) and the
> > >> current Itanium (Tukwila).
>
> > >> The EV7z was released in 2004, used 180 nm and runs at 1.3 GHz.
> > >> Tukwila was released 2010, uses 65 nm and runs at 1.33-1.73 GHz.
> > >> Pulson (next year?) will use 32 nm.
>
> > >> EV8 was ment to use 125 nm and run at 2 GHz.
>
> > >> How would a EV7z/8 shrinked from 180/125 nm to 65 or 32 nm performed ?
> > >> And with 24 Mb on-die cache (as Tukwila) instead of 1.75 (3 for EV8) ?
>
> > >> The highest clock speed for an 180 nm Itanium (McKinley) was 1.0 GHz.
>
> > >> Ah well... :-)
>
> > > Whatever technical merits can be brought to the table,
> > > they don't answer the question whether the world
> > > needs yet another competing CPU design.
> > > I think the history of both latecomers, Alpha and Itanic,
> > > shows that there's not much commercial room
> > > beyond x86 and the classical RISCs.
> > > Iirc the last successfull introduction of a new architecture
> > > was IBM's Power, this was more than 20 years ago.
>
> > Well, the Alpha was not really a latecomer. It's also about 20 years old
> > now. And it was the first high speed CPU out there, as well as the first
> > clean 64 bit CPU. It had a lot going for it.
>
> Alpha did indeed have a lot going for it.  My PWS is 600 MHz and
> according to my interpretation of the serial number was manufactured in
> Feb/Mar 1997.  In March 1997 I was in the market for an Intel box, and
> my supplier recommended I go for a two x 133 MHz cpu configuration since
> that would be cheaper than a single "latest and greatest" 200 MHz cpu.
>
> Three times the speed of the fastest Intel offering.
>
> > However, it was tied to a company that was sinking, which is a part of
> > what made it fail.
>
> Prices cannot have helped.  I don't know what my Alpha PWS cost new, but
> we did find an old advert for one at USD 20,000 (spec unknown, so not
> necessarily a true guide).  My well specced Intel box cost a fraction of
> that (circa GBP 4K).
>
> --
> Paul Sture

I'm sure you already know this, but please don't compare based on MHz
alone.

By the time the DIGITAL Personal Workstation(433, 500 and 600MHz,
fwiw) came out, the basic prices were reasonably competitive compared
with similarly configured UNIX workstations from other vendors (sorry,
no references right now).

There was also an Intel PWSi in a similar NLX-format system which was
intended to compete with other vendors Intel-based workstations (from
266MHz, fwiw).

The entry level PWS/Alpha prices depended on whether you wanted to be
able to run just NT/Alpha, or pay a bit more (and probably install a
bit more memory and disk) and run VMS and/or UNIX (Tru64).  $20K
sounds well high end (huge quantity of memory? fast SCSI disks?).