Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site decwrl.UUCP
Path: utzoo!watmath!clyde!bonnie!akgua!mcnc!decvax!decwrl!dec-rhea!dec-viking!wasser_1
From: wasser_1@viking.DEC (John A. Wasser)
Newsgroups: net.puzzle
Subject: Z80 Code Puzzle (from Dr. Dobbs Journal)
Message-ID: <1059@decwrl.UUCP>
Date: Mon, 11-Mar-85 17:34:03 EST
Article-I.D.: decwrl.1059
Posted: Mon Mar 11 17:34:03 1985
Date-Received: Fri, 15-Mar-85 02:19:44 EST
Sender: daemon@decwrl.UUCP
Organization: DEC Engineering Network
Lines: 44


In the January 1985 issue of Dr. Dobb's Journal, on page 17 the following
article appeared:

	David S. Tilton sent us [DDJ] the following sequence of Z80 
	code.  It's an absolutely astonishing implementation of... 
	but you can figure it out.  Unfortunately, it requires such 
	narrowly defined conditions that it's probably useless, 
	despite being the fastest one of its kind. What does it
	do, and what are its narrow requirements for utility?

[Comments added by me -JW]

-------------------------------------------------------------------------
; A is an 8 bit register
; B is an 8 bit register
; HL is a 16 bit register made up of H (the high byte) and L (the low byte)
;
LOOP:
	CP	(HL)	; Compare A to memory at HL
	RET	Z	; Return if equal.

	RL	L	; Rotate carry and L register (bottom 8 bits of
			; HL pair) left (Carry goes into LSB of L)
	DJNZ	LOOP	; B--; if (B != 0) GOTO LOOP

	LD	L,0	; Load L with the value Zero.
	RET		; Return
-------------------------------------------------------------------------

	I have yet to hear a reasonable answer.  If you get an answer
	from DDJ, please don't post it for a couple of weeks... that
	would spoil the fun!

		-John A. Wasser

Work address:
ARPAnet:	WASSER%VIKING.DEC@decwrl.ARPA
Usenet:		{allegra,Shasta,decvax}!decwrl!dec-rhea!dec-viking!wasser
USPS:		Digital Equipment Corp.
		Mail stop: LJO2/E4
		30 Porter Rd
		Littleton, MA  01460