U1DF1@wvnvm.wvnet.edu (John Neubert) (09/30/89)
In vol 8: issue 135, Chuck Williams writes of "The Sun is Setting!" He relates *rumors* that Sun is attempting to sell off the 386i line *and* the Sun3 line *and* any "non-SPARC Sun4 machine". Now, firstly the operative word is "rumors". Yes, I've heard the one about the 386i (and "sub"rumors that the Cambridge crowd got too independent (with the interface, etc) and that the 386i sold like bags of ice in the artic -- but, then, I've also heard that the 386i was selling like the proverbial "hotcakes" -- aren't rumors wonderful; they're just like Unix standards, there're so many of them, one's bound to find one he/she likes). I've also heard of "bad blood" between Moto and Sun since Sun wouldn't wait for the 88K -- Moto was so miffed that they didn't invite Sun to some Moto-sponsored big bash of 68xxx developers. (Then again, business makes for strange bedfellows -- two companies can publicly go at each others throats and still wheel and deal together; look at the Apple-Microsoft law suit vs the font/pdl deal.) *BUT*, just what is meant by "non-SPARC" Sun4s in the note? As far as I know *all* 4s are Sparc-based. Remember, rumors are just that... and it ain't over 'til (blah)(blah). Anyway, I would actually like to see Sun concentrate on just SPARC and producing a wide line of products using that technology -- coprocessors (hard and soft) could get the rest. As for lead times, I haven't heard anything remotely close to 240 days! But long lead times (up to 90 days, or more) are what happens when you come out with a great product (*and* have an internal snafu at the same time). Heck, I waited as long when I got my Osborne I and my Honda car! ;-) I wrote the above just before I saw the note from Gregory Miller (8:143) responding to Chuck's note. In Gregory's article he only refutes the 386i rumor and only references the EE-Times article. EE-Times prides itself on being early with stories in the electronics industry. Unfortunately, this same *rumor* has surfaced in a number of other publications and in personal conversations. I realize there's that thing called "pack journalism" (and that many journalists use each other as "sources" -- isn't incest wonderful!), but sometimes "where there's smoke, there's fire". I still see a concentration on SPARC as Sun's best strategic plan. P.S. Anyone know anything about *rumored* multi-processor work (ala Solbourne) at Sun; or a rumored multi-i860 board for Suns? Not rumored (at least I haven't seen *this* one, but we could always start it!) is any work on an actual parallel architecture machine at Sun. Anyone heard of such a high-end project -- or even an advanced NCS-type network application? Heck, it seems Sun, and most of the workstation/computer vendors looked at the Transputer and dismissed it (too small and unstable a source and one that the British would not allow to be lost -- some US companies did actually try!). Now, it looks like the other microprocessor vendors are either trying to emulate the commun- ication features of the T-800 (ala Moto's 88K) or the systems manufacturers are taking another look at the Transputer -- perhaps also because of the rumored (there's that word again) H1 follow-on that will not only have greater performance that the SPARC architecture by 1991 (1991's SPARC architecture!) but increased interprocessor communication connections than even the current Transputer. Ah, isn't the rumor mill wonderful. Or, maybe the *rumored* '040 NeXT with four 88Ks is even more interesting?!