>> Then we have the WD Pascal Microengine that
basically was the
>> implementation
>> of P-code in microcode.
>> There are machines that are coded for forth primitives directly.
>> I believe somewhere there was or is a a Java engine.
>> Memory says there was a Wang machine that directly executed Basic.
>> Generally it was not uncommon but most were lost to time.
> I get a little uncomfortable contemplating the possibility of a
> machine
> designed to execute
ASP.NET.
On Thu, 1 Jan 2009, Cameron Kaiser wrote:
Please, I'm trying to eat here.
On Thu, 1 Jan 2009, Dave McGuire wrote:
WHY did you go there??
Sorry for such an excremental comparison, this is an aspect of the field
that I am interested in, but know very little about.
But some systems, such as Java bytecode, and including ones that might
support multiple languages such as UCSD P-system, compile to a moderately
low level intermediate language.
IF that intermediate is low enough, then it becomes feasable to design
hardware that is closely tied to the language, such as the p-system
micro-engine, or SOAR ("Smalltalk On A Risc"), etc.
ASP appears to be designed that way, and I have been meaning to get and
read the ASP ILASM ("Intermediate level assembler"?) books.
Would MICROS~1 not be working on a dedicated hardware system?
For that matter (sorry for the mass indigestion), surely Intel is studying
and doing massive statistical analyses of the relative frequency of
different instructions, in order to optimize processors for what they will
actually be used for. If their data is real world, then such
optimizations would ultimately result in creation of processors that are
specifically optimized for the weirdities of Windoze!
I had hoped that this year would be better than last.
--
Grumpy Ol' Fred cisin at
xenosoft.com