gideon@taux01.UUCP ( Gideon Intrater) (11/26/89)
I heard that NEC is s going to announce a new microprocessor (think its called - V80). Does anyone know how fast it is? Is it already available ? Is there any available technical data about it ? Gideon Intrater gideon@nsc.nsc.com National Semiconductor, P.O.Box 3007, Herzlia B 46104, Israel Phone: +972-52-522303, Fax: +972-52-558322
pec@necntc.nec.com (Paul Cohen) (11/27/89)
In article 12595 of comp.arch: gideon@taux01.UUCP ( Gideon Intrater) asks: > I heard that NEC is s going to announce a new microprocessor (think its > called - V80). Does anyone know how fast it is? Is it already available ? > Is there any available technical data about it ? This part exists now, in internal engineering samples. The 25 MHz part executes at 12.5 Mips and there will soon be a 33 MHz. part. It features 2K of on-chip instruction and data caches, and a seven stage pipeline with branch prediction. The V80 is a successor to the V60 and V70, and is essentially a faster part with this same architecture. Like these earlier parts, it supports demand paged virtual memory through an internal MMU, internal floating point, support for fault tolerance, and a V30 (i.e., 8086) emulation mode. These 32-bit V-series processors have some popularity in Japan, but like these earlier parts, the V80 may not be aggressively promoted for the U.S. market (this is my own perception, not a policy statement by NEC). In any event, we have at present very little literature available for the device. It is mentioned in our data book, but little detail is provided. There was an article about the V80 in the February 13, 1989 issue of EE Times.
pec@necntc.nec.com (Paul Cohen) (12/01/89)
I have received some follow-up correspondence to my posting in comp.arch about the V80, which I summarize below. > And the question foremost in everyone's mind... is it an 80386 clone? > -- Chuq Von Rospach, chuq@Apple.COM No! Not at all! It is NEC's own architecture. In native mode, the V60/70/80 carry none of the bagage from the old 8088/86. The architecture is roughly similar to that of a VAX. The only similarity to the 8086 architecture is in emulation mode (where the V60/70/80 execute 8086 object code). Also, in contrast to the 80X86's, entering and leaving emulation mode is painless. > I hope the V60 is popular in Japan, because it doesn't look like > anybody uses it in this country. . . . > Too bad, the V60 was a delightful chip to program. I wish it could > have replaced the 386/486/etc. Sounds like the V80 is hot stuff. > -- Mark Alexander (amdahl!drivax!alexande)
peter@ficc.uu.net (Peter da Silva) (12/02/89)
In article <28388@necntc.nec.com> pec@necntc.nec.com (Paul Cohen) writes: > > And the question foremost in everyone's mind... is it an 80386 clone? > > -- Chuq Von Rospach, chuq@Apple.COM That sounds like a question I asked. I'll have to be more careful with my cute little .signature quotes... because some people are not at all careful about attributions. > No! Not at all! It is NEC's own architecture. Pity. I'd have liked to pull the replace-the-x86-with-a-NEC-chip trick again. -- `-_-' Peter da Silva. +1 713 274 5180. <peter@ficc.uu.net>. 'U` Also <peter@ficc.lonestar.org> or <peter@sugar.lonestar.org>. "The basic notion underlying USENET is the flame." -- Chuq Von Rospach, chuq@Apple.COM
cth_co@tekno.chalmers.se (CHRISTER OLSSON) (12/04/89)
> > Pity. I'd have liked to pull the replace-the-x86-with-a-NEC-chip trick > again. Yeah! Why not? Change my old slow 10Mhz 80286 to a 80386-instruction compatible Vxx with 8K cache and an inbuilt 80387-clone :-)