Date: Sun, 25 May 1997 10:47:34 -0700 From: Sean Eric Fagan <sef@Kithrup.COM> To: dave@persprog.com Cc: chat@FreeBSD.ORG Subject: Re: Intel Pentium II released Message-ID: <199705251747.KAA17807@kithrup.com> In-Reply-To: <338726E2.6A9D29F4.kithrup.freebsd.hardware@persprog.com> References: <199705231702.KAA29056@george.lbl.gov> <19970523142235.11747@ct.picker.com>
next in thread | previous in thread | raw e-mail | index | archive | help
In article <338726E2.6A9D29F4.kithrup.freebsd.hardware@persprog.com> you write: >The other problem with this design is that it will be short lived since >the Deschutes will require a revised socket (called "Slot II" I >believe). Actually, the Deschutes is supposed to use the "Slot One" as well. But this lets me get on a rant: I am sick of Intel right now. They have been worrying me for quite a while, as they have a monopoly on the computer market. (More than 85% of all processors in the world currently in use are Intel processors. That constitutes a monopoly by most economists definitions.) And that's not good enough: now they're trying to use the patent system to prevent anyone else from making x86 clones. First, the one that really got me worried: Intel has patented some instructions. Yes, *instructions*. Specifically, they have applied for patents on some instructions, apparantly for the Merced, all to deal with moving back and forth between x86 mode and IA-64 mode. (A couple of instructions to move data back and forth between IA-64 registers and x86 registers, and a couple of instructions to jump between x86 mode and IA-64 mode.) The semantics for these instructions are fairly obvious, I think -- if you accept that you are going to want to be able to switch back and forth between two instruction sets, there are only so many ways you can do it. *But*: since the patent describes the instruction format, that may be enough for the PTO decide it is "novel" and patentable. Meaning nobody will be able to make a compatible processor. Next, there's the "Slot One." Why did Intel drop the Socket 7 and similar? Because they couldn't get a patent on those. Instead, they came up with this "Slot One," which is different enough that they can patent it. And they have. This means two things: nobody can make a motherboard without Intel's permission, and nobody can make a card for the slot without Intel's permission. Previously, if Intel had tried to do this, the motherboard manufacturers would have told Intel to go fly a kite -- and Intel would have had some problems. But now Intel is a major manufacturer of motherboards -- so any other manufacturer who says that, Intel can say, "Fine, we'll just make more of our own." And, of course, Intel is certainly not going to agree to any licenses with Cyrix or AMD. So nobody but Intel and HP will be able to make new processors. I'm beginning to think that the Java Virtual Machine isn't such a bad thing after all -- not because it's any threat to uSoft, but because it means Intel may not be as important. I'm very frightened of Intel these days. And... I don't think there's anything that can be done about it. With uSoft, forcing it to split into an applications company and an OS company would handle most of the problems, I think. But Intel just has the x86. Without that, it's nothing; and, conversely, you could get rid of everything but the x86 line at Intel, and it wouldn't make a dint in their profits or revenues. That has been my rant for the day. Sean.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199705251747.KAA17807>