mpx (mpx@nomail.pl) on 2/24/09 wrote:
---------------------------
>anon (anon@anon.com) on 2/24/09 wrote:
>---------------------------
>
>>to finishing manouvering IPF to exactly where it seems to >have been planning for a long time? (RISC replacement)
>
>RISC CPUS have have already been replaced by x64 where most CPUs were sold - workstations
>and few-socket servers. Because of both cost and software compatibility.

I said RISC replacement, not x86 replacement. So x86 replaces RISC in volume markets, and IPF is intended to replace it in margin markets. What's the problem? Any denials?

>If software compatibility was not important new ISAs would be introduced and dumped
>frequently. But it is, and it means there's less sense for Itanium to exist than some people imagine.

And yet it continues to grow, despite years and years of disadvantages versus x86 (and against continual predictions of naysayers -- how can you continue claiming you are right when you predicted it would die several years ago?).

Note (and note to Linus), I'm not talking about technical comparisons. Obviously x86 has indeed had great advantages in design and manufacturing, and IPF has been badly executed for years. And yet with all that, IPF has continually been growing, and has become significant enough that Intel seems willing to give it a cutting edge process soon.

And no, I'm not going to claim that single fact will make it faster than contemporary x86 on specint. That's wasn't my point when mentioning process parity and head in sand. My point is that Intel is continuing to take IPF seriously, and it doesn't behave like something it hopes will go on life support and shrink away.

I think Intel is quite happy for the moment with the low-end crowd dismissing IPF. I'm sure Intel doesn't want to push for commiditizing its highest margin and most exclusive markets. Why would it want to push x86 upward any faster? To help carve out niches for AMD to enter?