Compaq's alpha unit being sold (off-topic)
r...@myplace.org
you may even know about this already), but it is relevant:
http://www.theinquirer.net/22060114.htm
... so i guess it looks like Intel is buying Compaq's Alpha division.
--
--===-----=======-----------=============-----------------===================
| rbw aka bjp | god's final message to his creation: |
| r...@myplace.org | we apologize for the inconvenience. |
===================-----------------=============-----------=======-----===--
To Unsubscribe: send mail to majo...@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message
mja...@feral.com
If this happens and plays out, you can kiss the Open Source movements goodbye
within 5 years.
r...@myplace.org
>
> If this happens and plays out, you can kiss the Open Source movements goodbye
> within 5 years.
pucker up. =\
--
--===-----=======-----------=============-----------------===================
| rbw aka bjp | god's final message to his creation: |
| r...@myplace.org | we apologize for the inconvenience. |
===================-----------------=============-----------=======-----===--
To Unsubscribe: send mail to majo...@FreeBSD.org
A.Derg...@tn.utwente.nl
Matthew Jacob wrote:
> If this happens and plays out, you can kiss the Open Source movements goodbye
> within 5 years.
>
Hmm,
I wonder why do you think so ?
mja...@feral.com
> Matthew Jacob wrote:
>
> > If this happens and plays out, you can kiss the Open Source movements goodbye
> > within 5 years.
> >
>
> Hmm,
>
> I wonder why do you think so ?
If the amount of viable large scale platforms drops to just Intel, the only
way so-called 'Open Source' will live is as a relatively unpaid adjunct to
whatever Intel feels is appropriate.
That doesn't mean that there won't be Alpha, Sparc, Arm32, PowerPC and other
chips around. But they'll become so marginalized and niche oriented that the
main thrust of Open Source, including Linux (which will probably be co-opted
by IBM by the end of next year), will be lost.
IMHO. Wrong list for this rant anyway- sorry.
j...@osd.bsdi.com
> within 5 years.
Tsk tsk. As much as I admire the Alpha and its place in time, Matt, I
think it's an orphan child currently working its way down through a
procession of progressively crueler and less loving foster parents.
It will almost certainly end up the way they all finally do - running
away and living out its last days on the streets, sleeping in a
discarded hardware dumpster behind Fry's and turning tricks for
crack. :-)
Until the 64 bit architectures with more "mainstream backing" really
start appearing in quantity, however, it makes an excellent reference
and test platform. Don't get any more emotionally attached to it than
that and you'll be fine. As far as the "Intel's taking over the
industry" predictions are concerned, I'm certainly not as worried as I
was perhaps 4 years ago and I don't see why anyone else should be
either.
Intel has by no means got things all sewn up to its satisfaction, the
fact that AMD has come from way behind to give it serious heartburn on
the high-end being a good sign that any real opportunity for an
exclusive lock vanished somewhere around the time of the PIII, when
things started to seriously slow down there. The P4 hasn't exactly
been a sales success story in its own right and the Itanium is by no
means a sure bet either. I kinda like AMD's Hammer stuff, but we'll
just have to see.
Far more interesting to me are the truly interesting non-x86
generation processors which haven't even yet been invented and I
suspect a good number of them won't be created by Intel, either. That
company has a lot of inertia to fight and going to the next generation
of processor architectures is going to involve stuff a lot more
innovative than taking the same tired optimization tricks and doubling
or tripling them. They've just about run out of room on that
approach.
- Jordan
peter....@alcatel.com.au
>... so i guess it looks like Intel is buying Compaq's Alpha division.
It seems the deal's been done: http://www.theinquirer.net/2506alpha.htm
RIP Alpha. I'm glad the iA64 port is progressing nicely.
Peter
mja...@feral.com
On Sun, 24 Jun 2001, Jordan Hubbard wrote:
> > If this happens and plays out, you can kiss the Open Source movements goodbye
> > within 5 years.
>
> Tsk tsk. As much as I admire the Alpha and its place in time, Matt, I
> think it's an orphan child currently working its way down through a
> procession of progressively crueler and less loving foster parents.
> It will almost certainly end up the way they all finally do - running
> away and living out its last days on the streets, sleeping in a
> discarded hardware dumpster behind Fry's and turning tricks for
> crack. :-)
Hmm- I can't tell whether you are saddened or thrilled by this scenario.
I never suspected you of such Gibsonesque leanings before.
> Until the 64 bit architectures with more "mainstream backing" really
> start appearing in quantity, however, it makes an excellent reference
> and test platform. Don't get any more emotionally attached to it than
> that and you'll be fine.
I'm fine with whatever happens- I haven't had any sense of control of such
events (which was foolish even then) since I worked for Sun.
It'll be interesting. Since some large government labs have predicated their
supercomputing clusters on EV8, we'll see what they have to say about it.
I'm just noting that with the death of any serious corporate backing for
servers for Alpha, the only thing left that could be taken seriously as a
server alternative to Intel is the UltraSparc platforms, and that's unlikely
to be a win for Open Source here either as there's less support in Sun than
there was inside DEQ for Open Source. Oh well.
> As far as the "Intel's taking over the industry" predictions are
> concerned, I'm certainly not as worried as I was perhaps 4 years ago and I
> don't see why anyone else should be either.
Really? Well, we'll see. My personal guess is that FreeBSD will become the
personal overseas outsourcing for cheap OS work from Intel, but w/o anything
to keep Intel honest and on their toes. I hope I'm wrong.
>
> Intel has by no means got things all sewn up to its satisfaction, the
> fact that AMD has come from way behind to give it serious heartburn on
> the high-end being a good sign that any real opportunity for an
> exclusive lock vanished somewhere around the time of the PIII, when
> things started to seriously slow down there. The P4 hasn't exactly
> been a sales success story in its own right and the Itanium is by no
> means a sure bet either. I kinda like AMD's Hammer stuff, but we'll
> just have to see.
>
> Far more interesting to me are the truly interesting non-x86
> generation processors which haven't even yet been invented and I
> suspect a good number of them won't be created by Intel, either. That
> company has a lot of inertia to fight and going to the next generation
> of processor architectures is going to involve stuff a lot more
> innovative than taking the same tired optimization tricks and doubling
> or tripling them. They've just about run out of room on that
> approach.
Sure. There'll be a lot of niche processor space to work with. The PowerPC and
the high end MIPS chips do well here.
-matt
jam...@what.net
> It'll be interesting. Since some large government labs have predicated their
> supercomputing clusters on EV8, we'll see what they have to say about it.
I thought about the same earlier in the weekend.
> I'm just noting that with the death of any serious corporate backing for
> servers for Alpha, the only thing left that could be taken seriously as a
I think this may be the fatal flaw in your argument, that is assuming
there's been serious corporate backing for the Alpha at any point
in the last 5 years. Let's face it, DEC was closer to death than
anyone really wants to admit. Further, just given how Compaq has been
marketing themselves to even current Alpha customers has given, at
least, me the impression that not only were they never really all
that committed to the Alpha, but it's also pretty clear that they
don't even understand the Alpha and why it worked so well for DEC
for so long.
The not understanding part is pretty easy and can be summed up in
one word: Linux. Compaq has spent every waking moment since
discovering that they really did own the Alpha trying to position
it as a faster Linux (than that copy of Red Hat you've got installed
on your LETNi machine). Digital mortgage their future on the Alpha
because they saw the processor as more of a platform to provide
stability and scalability via VMS and Digital UNIX on a platform
they not only control, but write the compiler for.
With that in mind, I really don't see OSF (Digital UNIX) as a viable
solution when the Alpha is abandoned in favor of IA-64, which is
exactly why Dec abandoned the OSF port to IA-64 in the first place.
You have to ask yourself a question: Why would I but a Compaq
branded IA-64 box in favor of any number of cheeper operating
systems with will probably work just as well? The answer is, I
wouldn't.
VMS, on the other hand probably is just as viable on IA-64 as it
is on Alpha. (Reference: VMS/MIPS -> OpenVMS)
> server alternative to Intel is the UltraSparc platforms, and that's unlikely
Tell that to IBM and Apple. I think you'll see both companies be
a bit more forceful with their positioning in the large server
market. Which is a good thing, as I'd be more than happy to buy
a large Apple PPC based servers when I can't buy Alphas anymore.
> to be a win for Open Source here either as there's less support in Sun than
> there was inside DEQ for Open Source. Oh well.
So, you're arguing that Open Source works better when there's more
platforms to port your OS to? I'm not sure I agree. Just on the
surface, I'd say that's it's somewhat easier since you know who
your friends are, and what they are trying to hide.
You are also assuming that Intel is like Microsoft. In several
ways they are (take a very close look at this agreement) but they
really aren't that interested in controlling what software you run.
Further, Intel realized a while ago that Microsoft isn't the only
game in town and really has almost no standing in the large
server market. Linux is the BSDs are far more important in this
segment and Intel knows this.
[1] Though it generally is, but only because Dec never could get
away with the kinds of crap that LETNi does.
j...@osd.bsdi.com
Subject: Re: Compaq's alpha unit being sold (off-topic)
Date: Mon, 25 Jun 2001 14:13:31 -0700 (PDT)
> Hmm- I can't tell whether you are saddened or thrilled by this scenario.
Saddened to be sure, but saddened in a way that someone who's watched
a someone live purely on life support equipment for years and then finally
die might be saddened. Sad, but long-since resigned to the inevitability
of the event once it finally came.
Call me an optimist, but I think the future is actually rather bright
for new processor architectures. While the x86 remained viable, it
set the computing industry back years and convinced a lot of people
that it just wasn't worth competing with legacy designs. Now that
Intel has more or less tacitly admitted that it's run out of steam by
going for a substantially different design with Itanium, it
legitimizes the effort for more nimble companies. Itanium is a good
example of a fine idea in principle but not particularly attractive in
implementation. To my thinking, at least, that leaves the door pretty
wide open.
- Jordan