[SGVLUG] Why Coding At Fifty May Be Nifty

[bobjaffray at juno.com]

---------- Original Message ----------
From: Rod Morison <rod at morison.biz>
To: sgvlug at sgvlug.net
Subject: Re: [SGVLUG] Why Coding At Fifty May Be Nifty
Date: Mon, 05 Nov 2012 10:02:10 -0800



On 11/5/2012 9:55 AM, bobjaffray at juno.com wrote:
>   But FORTH
> was what I thought was outstanding not for its terrible syntax
> but it superior capability to create embedded applications.
I recall trying to write a FORTH compiler using a free Pascal compiler 
for the Amiga 1000. Got through most of the parser, but never finished 
code gen.

BJ: My processor was the Intel 8080/8086. The idea was to get the
number of primatives down to about 18 or so very short Assembler
code segments, and compile the rest with direct threaded code,
first on a host machine and then on the target processor for the
standalone embedded application running at full speed a bit at
a time without the need for an simulator/emulator. That was my 
goal, but I never got off the ground with it. I still think it
is the logical way to go.

BJ: The problem with FORTH is that it is a community of dilitantes,
with everyone creating his own wierd version of FORTH to suit
himself. But for speed it can't be beat IF the whole code is
cross-compiled to optimized machine code. But that is an extremely
small niche, because extemely few embedded applications 
require speeds anywhere near the best modern processor speeds.

Bob
____________________________________________________________
Woman is 53 But Looks 25
Mom reveals 1 simple wrinkle trick that has angered doctors...
http://thirdpartyoffers.juno.com/TGL3141/5098394a90fa5394a51dfst02vuc