From mboxrd@z Thu Jan 1 00:00:00 1970 From: lm at mcvoy.com (Larry McVoy) Date: Fri, 5 Feb 2021 06:56:18 -0800 Subject: [COFF] [SPAM] Re: Architectures -- was [TUHS] 68k prototypes & microcode In-Reply-To: References: <202102030759.1137x7C2013543@freefriends.org> <202102030858.1138wuqd011051@freefriends.org> <27567.1612399305@hop.toad.com> <20210204013356.GA16541@mcvoy.com> Message-ID: <20210205145618.GE13916@mcvoy.com> On Fri, Feb 05, 2021 at 09:36:20AM -0500, Clem Cole wrote: > FWIW: Intel >>does<< know how to make a $20 SOC, but the margins will > suck. The question is what will management want to? I really don't > know. So far, we have liked the server chip margins (don't forget Intel > made more $s last year than it ever has - even in the pandemic). I think Intel is sort of in the same place Sun was. Fat, dumb, and happy with the profits they are making and can't see what is coming. It just didn't make sense to have $20,000 Sun workstations when a $2,000 PC was at least half as good. I advocated for SunOS on x86, to me, it was the operating system that delivered the value, everything just worked on SunOS, for any other OS you were doing the configure dance. Offer SunOS on x86 and capture the low end market. The East coast Sun did the road runner but West coast Sun sneered at it, patches for x86 were not processed very fast, if at all. It's a shame. If Intel doesn't want to make money off of the cheap, but very high volume, $20 SOC, Apple has shown that it has the chops to make a cheap, fast, and power sipping M1 chip. Pretty impressive and if I were Intel, I'd be nervous. Apple has shown they can switch architectures pretty painlessly repeatedly. The x86 lock in isn't much of a lock in these days.