[Info-vax] OT: IA-128 ???

[Neil Rieck]

On Oct 17, 4:49 am, JF Mezei <jfmezei.spam... at vaxination.ca> wrote:
> Marc Schlensog wrote:
> > A 128bit file system would be an entirely different story. But you
> > don't need a 128bit CPU for that.
>
> Lets say they were to produce an 8086 with 64 bit addressing, and 128
> bit registers and data paths.
>
> Are there many operations that were really benefit from that ?
>
> Moving long strings from one location to another would be faster since
> half the fetch/strore operations would be needed to move data around.
>
> But for integer math, would it really make a difference ? How many
> applications need to process numbers greater than 2^64 ? I think that
> even Bill Gates's net worth fits comfortably within 64 bits.
>
> Does IEEE already have a 128 version of the floating point
> representation ? Would creating 128but floating point registers provide
> advantages for high end computtional work (or video games) ?
>
> The move from 32 bit to 64 bit was made, as I recall, mostly to allow to
> go beyond the 4 gig of ram limit in 32 bit, and mostly to allow atabase
> engine to store more of s database in memory to increase performance.
>
> Perhaps going 128 bits addressing  might allow mapping
>  of very big files to memory ? In other words, you may not have more
> than 64 bit's worth of phyical memory,  but your whole database of
> terabytes of records would be mapped to memory (think global section in
> VMS terns).
>
> Google might be a driver for this, when you consider it has huge
> databases (images for google erth street vie etc).

I am now convinced John Reagan's post is the correct interpretation
("perhaps some 128 multiply instructions will appear in some future
x86-64 platform" rather than "some new IA-128 architecture"). Why?

1) Up until now, Intel always pre-announces new technologies at IDF
and they have never used the phrase "IA-128" even though one person at
Intel did.
2) Although both AMD and Intel added 64-bit extensions to their
respective x86 chips, the majority of the x86 operating systems and
applications currently in use only use 32-bits. The bulk of the
industry has not yet expanded into a 64-bit address or data space.

You people are bang on with your comparisons between 32-bit
(4294967296) and 64-bit (18446744073709551616) integerers. Not many
apps require integers of this size, but if they did, they could always
be constructed by using multiple memory locations (just like we did
with 4-bit, 8-bit, 16-bit, and 32-bit processors).

I'm sure everyone here already knew that some SSE2 instructions
already operate on 128-bit registers (which is why I have always had a
problem believing Intel's claim that a 32-bit Pentium-4 was a RISC
chip). But check out the following link:
http://en.wikipedia.org/wiki/Streaming_SIMD_Extensions#Later_versions
Both Intel and AMD are working on 256-bit SSE instructions called AVX
(Advanced Vector Extensions) which will appear in 2011.

So I am going to return to my original definition which came right out
of the mouth of DEC instructors in Boston in the early 1980s: "A
processor with 16-bit GPRs is a 16-bit processor"

NSR