Michael S (already5chosen.delete@this.yahoo.com) on February 5, 2013 2:51 am wrote:
> Richard Cownie (tich.delete@this.pobox.com) on February 4, 2013 5:52 pm wrote:
> >
> > Conversely, porting to Itanium (which was narrowly avoided ...) would have meant
> > spending money on a relatively expensive workstation, not useful for anything else,
> >
>
> The key word is "relatively".
> According to my understanding, by now you can find decent Itanium gear (HP rx2620, or even
> rx2660, if you got lucky) on eBay for less than the price of new developer-grade PC.

The SPARC64 port was around 1999/2000, x86_64 as soon as it became available in 2003.
It was lucky for me that Intel adopted x86_64, since they were also a customer
for the product and wanted to run it on some Intel platform, not SPARC or Opteron.

> > and then having to port all the performance-sensitive executables, even those which
> > didn't need >4GB, to have anything shippable.
> >
>
> After you ported to Linux/x64, shouldn't porting to Linux/IPF be close
> to "just recompile"? Esp, if the same code already runs on SPARC?

In theory, yes. Except that the result of "just recompile" would probably have
been something very slow, since the code was mostly about traversing many GB's
of complex data structures in thoroughly unpredictable ways, which an in-order
static-scheduled cpu would have struggled with.

I also heard from people who did tackle Itanium that compilation was slow, which
in itself would have made "just recompile" quite painful back in those days
(we implemented a distributed-compile strategy on a farm of x86_64/linux nodes
later on, around 2005, but even that wouldn't have helped if you only had a
single Itanium machine).