Re: [?] Why should Distros be called as i386 for a 32-bit PC, and as amd64 for a 64-bit PC, when Intel Core PCs are also 64bit systems
>>So it was a great move on the part of AMD: cheap to implement but with
>>an enormous marketing impact.
> It had much more than a marketing impact, because x86 was a PITA for more
> than 2GB of RAM and that was getting cheap and becoming a common problem by
> 2003. Switching to opteron for 8G or 16G servers was a huge win vs x86, with
> better scaling for multiprocessor configurations. (These were becoming more
> common as well, and intel was still using an old (obsolete?) flat SMP bus
> whereas AMD arrived on the scene with a far superior NUMA architecture based
> on hypertransport--designed in partnership with what was left of the old DEC
> alpha team.) It was simply the right product at the right time.
I think the performance of the Opteron would have been
sufficient to make it quite successful even if limited to 32bit.
And Microsoft took its time before releasing a version of Windows for
amd64, so most of the machines sold between 2003 and 2005 were running
in 32bit mode, AFAICT.
So I think the marketing impact of Opteron's support of the new amd64
ISA during the 2003-2005 window was more important than the technical
impact. But you're right that 64bit support was really becoming
important right around that time: PAE was not as satisfactory a solution
(which is why AMD went ahead with amd64: the situation was becoming
untenable).
> At the time ia64 was announced alpha & MIPS processors were in some of
> the largest and most sucessful systems in the world. With further
> development they could have remained there, but their management was
> convinced that ia64 was going to have an unbeatable performance
> advantage and that they couldn't compete with the R&D money intel was
> pouring in.
They could have survived a few years more, definitely. But SGI only had
a good presence in supercomputers and computer graphics which were
pretty small markets where the CPU didn't matter that much, so it was
very costly for them to have to keep developing new top-of-the-line
processors additionally to top-of-the-line GPUs and interconnects.
They were already financially in poor health and they needed to start
designing their systems around someone else's CPU.
DEC was even worse because they didn't actually own any particular
segment of the market (besides from the VMS segment which was not
getting very many new customers) and PCs running Pentium Pros (and
successors) were taking over the workstation market.
In retrospect maybe DEC and SGI should have merged and then partnered
with AMD (as you note above some of DEC's processor design team indeed
ended up at AMD on the Opteron project), but I think it would have taken
a crapload of foresight and/or faith to do that.
[ Not sure what part HP could have played there, I wasn't very familiar
with their products (beside drooling over the idea of a 2MB L1 cache,
that is). ]
Stefan
Reply to: