Re: is Vax addressing sane today

Lawrence D'Oliveiro <ldo@nz.invalid> writes:

On Tue, 01 Oct 2024 16:26:25 GMT, Anton Ertl wrote:
>

In any case, the performance advantage of the RISCs vanished during the
1990s ...

>
Only for as long as Intel could afford to spend 10× as much on developing
each chip generation as the RISC vendors could. It could because it could
reap 10× the profits in return, but it can’t any more.

Nexgen and AMD were smaller companies than Intel, DEC, Sun, HP, or the
AIM companies, and yet managed to produce CPUs that were competetive
with Intel's CPUs despite suffering from the CISC baggage.  If the
RISC companies failed to keep up, they only have themselves to blame.
It seems to me that a number of RISC companies had difficulties with
managing the larger projects that the growing die areas allowed.

Another issue was the marketing.  The RISC companies did not want to
damage their existing high-priced workstation and server business by
providing cheap CPUs for the masses, and yet had to do that in order
to displace Intel, AMD, and Cyrix.  AMD and Cyrix did not have that
problem.

Which is why you
see ARM coming to the fore, and RISC-V appearing as the upstart
challenger.

ARM did not have the marketing problem, either, because they were not
competing in the workstation/server market.  They developed their
business model of selling cores (and more) for SoCs for portable
computing, and expand from that.

... the RISCs never had wide ISV support, and so WNT on RISCs
flopped.

>
As I said above, RISC is still around and dominating the computing world.
They’re not running Windows, because it was Windows that could not adapt
well to them. Instead, they are running Linux.

Dominating?  In the smartphone and tablet world, yes.  In the embedded
world, too.  In laptops, desktops and servers, no.

- anton
--
'Anyone trying for "industrial quality" ISA should avoid undefined behavior.'
  Mitch Alsup, <c17fcd89-f024-40e7-a594-88a85ac10d20o@googlegroups.com>