Path: utzoo!attcan!uunet!tut.cis.ohio-state.edu!pt.cs.cmu.edu!dsl.pitt.edu!pitt!willett!ForthNet
From: ForthNet@willett.pgh.pa.us (ForthNet articles from GEnie)
Newsgroups: comp.lang.forth
Subject: Conventions and "tricks" used in Forth
Message-ID: <1883.UUL1.3#5129@willett.pgh.pa.us>
Date: 21 Oct 90 01:58:31 GMT
Organization: String, Scotch tape, and Paperclips.  (in Pgh, PA)
Lines: 23


 Date: 10-13-90 (13:58)              Number: 4048 (Echo)
   To: GARY SMITH                    Refer#: 4038
 From: RAY DUNCAN                      Read: NO
 Subj: CONVENTIONS AND "TRICKS"      Status: PUBLIC MESSAGE

 The hash-cache strategy that Mitch describes is quite effective.  LMI
 has had this in its older-generation PC/FORTH, PC/FORTH+, and Z-80 FORTH
 products for about 5 years; Micromotion also had it in its last line of
 Forth-83 products before they disappeared.  We found that we could just
 about double the compilation speed with a 256-slot hash cache.  We got
 an additional 2-3X improvement in compilation speed when we went to the
 totally hashed dictionary in the newer UR/FORTH implementations.

 The biggest problem with the hash-cache approach is in interpreting
 numbers;  you must probe the cache AND exhaustively search the linked
 dictionary before you can conclude that the token is a number and not a
 Forth "word."  The completely hashed dictionary eliminates this problem.

 NET/Mail : LMI Forth Board, Los Angeles, CA (213) 306-3530             
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: dwp@willett.pgh.pa.us or uunet!willett!dwp