Date: Mon, 25 Jun 2001 14:13:31 -0700 (PDT) From: Matthew Jacob <mjacob@feral.com> To: Jordan Hubbard <jkh@osd.bsdi.com> Cc: <rbw@myplace.org>, <freebsd-alpha@FreeBSD.org> Subject: Re: Compaq's alpha unit being sold (off-topic) Message-ID: <20010625140615.L3157-100000@wonky.feral.com> In-Reply-To: <20010624235945G.jkh@osd.bsdi.com>
next in thread | previous in thread | raw e-mail | index | archive | help
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 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010625140615.L3157-100000>