Path: utzoo!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!aplcen!mef
From: mef@aplcen.apl.jhu.edu (Marty Fraeman)
Newsgroups: comp.lang.forth
Subject: Re: ANS TC Magnet for LOOPS, EXIT, Term.
Message-ID: <4805@aplcen.apl.jhu.edu>
Date: 1 Mar 90 02:36:12 GMT
References: <578.UUL1.3#5129@willett.UUCP>
Reply-To: mef@aplcen (Marty Fraeman)
Organization: Johns Hopkins University
Lines: 33

In article <578.UUL1.3#5129@willett.UUCP> ForthNet@willett.UUCP (ForthNet articles from GEnie) writes:
>
> Date: 02-26-90 (08:19)              Number: 2974 (Echo)
>   To: STEVE PALINCSAR               Refer#: 2965
> From: IAN GREEN                       Read: NO
> Subj: DO LOOPS IN FORTH             Status: PUBLIC MESSAGE
>
>    The reason for using FOR-NEXT is that I saw a comment comparing FOR 
> vs. DO in the RTX2001 data sheet. It said FOR runs several times faster 
> so I assumed that this was a standard word in Forth.
>
Nope.  My impression is that the designers of the original Novix chip
had a cute way of doing the FOR-NEXT loop with only a minimal amount of
hardware.  Rather than make the hardware more complicated (amazingly,
the whole thing fit into a 4K gate array) they just invented a new loop
construct.  After all we are talking about Forth. -)  

While we were designing the Hopkins 32 bit Forth engine that later
became the SC32, we also struggled with this issue.  We finally came up
with a way to implement Forth 83 style do-loops in 2 cycles.  I believe
the Novix (and hence the Harris RTX 2000) design can do the FOR-NEXT
loop in just a single cycle.  However in all but the most trivial loops
this small difference is neglible.  

	Marty Fraeman

	mef@aplcen.apl.jhu.edu
	301-953-5000, x8360

	Room 13-s587
	Johns Hopkins University/Applied Physics Laboratory
	Johns Hopkins Road
	Laurel, Md. 20723