mas@teddy.UUCP (Mark A. Swanson) (03/22/85)
IBM bought an add for the new PC AT/370 product. Apparently this is an IBM PC AT with the 370 CPU addin board set installed. It appears on page 28 of the April 1985 issue of SCIENTIFIC AMERICAN. The ad copy refers to the product as providing ~ "the power of the 64 bit architecture of the System 370." This strikes me as rather outrageous hype. True, some of the IBM mainframe implementations of the 370 architecture fetch 64 bits of data at a time (if not more.) But how the heck can one reasonably describe the 370 architecture (which I consider to be the programmer's model) to be a "64 bit" one? (Preferably, the argument should apply to the PC AT with its (probably) 16 bit data bus and not apply to the "32 bit" VAX 11/780.) Mark Swanson ...decvax!genrad!panda!mas
dww@stl.UUCP (David Wright) (03/26/85)
I always thought that the IBM 370 was a 12 bit architecture at heart? That was what I was told when I asked why programs could not address segments of more than 4K bytes, (i.e. the adress field in the instructions is never more than 12 bits long).
mat@amdahl.UUCP (Mike Taylor) (03/28/85)
> The ad copy refers to the product as providing ~ "the power of the 64 bit > architecture of the System 370." > > This strikes me as rather outrageous hype. True, some of the IBM mainframe > implementations of the 370 architecture fetch 64 bits of data at a time (if > not more.) But how the heck can one reasonably describe the 370 architecture > (which I consider to be the programmer's model) to be a "64 bit" one? It *is* hype, but presumably refers to the fact that the architecture can do floating point arithmetic on 64 bits. Or it is a pre-announcement. -- Mike Taylor ...!{ihnp4,hplabs,amd,sun}!amdahl!mat [ This may not reflect my opinion, let alone anyone else's. ]
jans@mako.UUCP (Jan Steinman) (03/28/85)
In article <408@teddy.UUCP> mas@teddy.UUCP (Mark A. Swanson) writes: >IBM bought an add for the new PC AT/370 product... The ad copy refers to the >product as providing ~ "the power of the 64 bit architecture of the System >370."... This strikes me as rather outrageous hype. What else could you expect from the company who started this micro "bit inflation" fad? Does anyone remember the the short article (InfoWorld?) when some Intel VP "discovered" that their newly introduced 8 bit processor, the 8088, was actually a 16 bit processor? It seems they "discovered" this by reading an IBM PC ad in the WallStreet Journal! -- :::::: Jan Steinman Box 1000, MS 61-161 (w)503/685-2843 :::::: :::::: tektronix!tekecs!jans Wilsonville, OR 97070 (h)503/657-7703 ::::::
jeff@alberta.UUCP (Curt J. Sampson) (03/30/85)
In article <669@mako.UUCP> jans@mako.UUCP (Jan Steinman) writes: >What else could you expect from the company who started this micro "bit >inflation" fad? Does anyone remember the the short article (InfoWorld?) when >some Intel VP "discovered" that their newly introduced 8 bit processor, the >8088, was actually a 16 bit processor? It seems they "discovered" this by >reading an IBM PC ad in the WallStreet Journal! Hmmm... Sounds just like the way people at Motorola discovered that their 16 bit 68000 was actually a 32 bit processer. They read the Macintosh ads. Incidently, I consider the 8088 a 16 bit processer because it has an address space larger than 64K and will do 16x16 bit multiplies and divides. The 68000, though, won't do a 32x32 bit multiply... -- Curt Sampson ihnp4!alberta!jeff "There is a theory which states that if every anyone discovers exactly what the Usenet is for and why it is here, it will instantly disappear and be replaced by somehing even more bizarre and inexplicable. "There is another theory which states that this has already happened."