Re: is Vax addressing sane today

In article <vdnef0$3uaeh$5@dont-email.me>, ldo@nz.invalid (Lawrence
D'Oliveiro) wrote:

On Thu, 3 Oct 2024 23:49 +0100 (BST), John Dallman wrote:

Given all of IBM's missteps, it's mildly surprising they got that
one right. Even a stopped clock is right once a day ...

IBM doesn't often repeat a mistake. They're made all the ordinary ones,
so nowadays they usually invent new ones.

SGI decided to embrace the platform that was eating their market,
and try to sell Windows NT boxes. Trouble is, those NT boxes, while
only a fraction of the cost of an IRIX-based product, still cost
about 3� what other NT machines were going for.

SGI had a lengthy internal conflict about Windows NT. One group of pro-NT
people left and founded NetPower, whose idea was to build really fast
workstations running NT on MIPS. We had one for a while, and were
persuading Microsoft to fix a bug from the MIPSPro code generator for the
third time (we'd also had it on DEC MIPS/Ultrix, and SGI Irix) when the
Pentium Pro was released, and NetPower suddenly went very quiet.

Then there were the SGI Visual Workstations, which ran NT on x86. The
first generation of them were quite nice, but needed a very custom HAL,
and hence couldn't be upgraded to later versions of Windows once SGI
abandoned them.

The later generations were ordinary PCs - the one I had as a deskside for
a while was made by Mitsubishi - with an Nvidia graphics card. The only
SGI added value was their OpenGL driver, and that didn't seem to justify
the price if you were buying them.

By this time, SGI had a department of downsizing, whose job was to get
rid of departments and sites. Being an American company, this department
fought for power and budget share, and nobody inside the company seemed
to think that this would spell doom for SGI.

They could still have sold SPARC hardware running Linux. I can
remember comments saying Linux ran better on that hardware than
Sun's own SunOS/Solaris did.

They would not have faced up to that. There was an interesting incident
with Solaris on x86. Since the Linux and Solaris kernel interfaces are
somewhat similar, somebody at Sun decided to try making the Solaris
kernel capable of acting as a Linux kernel, so that they could run a
Linux userland and applications on the same machine as the Solaris
userland and applications.

So they hired some Linux people, but they didn't get good ones. A year
later, their Linux people came back to Sun with a huge set of patches
that amounted to patching a lot of the Linux kernel into Solaris, and
didn't do it at all well. The Solaris kernel people looked at it a bit
and said "Hell, no! This will destabilise Solaris!" They weren't
exaggerating.

So that year was wasted, and the project was restarted with some of the
Solaris people involved, to explain how their kernel worked. Quite a
while later you could install the 32-bit Red Hat Enterprise Linux 3.0
userland and most application would run, but not all. This was not a
success, and was dropped.

John