Path: utzoo!mnetor!uunet!husc6!rutgers!sri-spam!ames!sdcsvax!ucsdhub!hp-sdd!hplabs!pyramid!voder!apple!dgold
From: dgold@apple.UUCP (David Goldsmith)
Newsgroups: comp.sys.mac
Subject: Re: Code Optimization
Message-ID: <7281@apple.UUCP>
Date: 28 Jan 88 17:33:38 GMT
References: <6252@cc5.bbn.COM> <10000@ccicpg.UUCP>
Reply-To: dgold@apple.UUCP (David Goldsmith)
Organization: Apple Computer Inc., Cupertino, USA
Lines: 16
Keywords: mac,68000,68020

In article <10000@ccicpg.UUCP> nick@ccicpg.UUCP (Nick Crossley) writes:
>Another significant difference is that the 68020 has several additional
>addressing modes, so that double indirection (ie for handles) and scaled
>indexing for arrays, amongst other things, can be done in one instruction.

Actually, it turns out it's both smaller and faster to dereference a handle
with two separate instructions, just like on the 68000.  The single 68020
instruction is larger and slower than the two separate 68000 instructions.
-- 
David Goldsmith
Apple Computer, Inc.

AppleLink: GOLDSMITH1
UUCP:  {nsc,dual,sun,voder,ucbvax!mtxinu}!apple!dgold
CSNET: dgold@apple.CSNET, dgold%apple@CSNET-RELAY
BIX: dgoldsmith