(Comp.sys.hp48) Item: 82 by detlef at mwhh.hanse.de Author: [Detlef Mller] Subj: REMOTE3: debugged IR remote routines Date: Sun Oct 27 1991 [These routines can be considered upgrades to REMOTE and REMOTE2 on previous Goodies Disks. Don't use them; use these. Also see REMOTE4 on this disk for another programmer's approach. -jkh-] Hi, everybody. There was a bug in the previous routines. Fortunately I decompiled the programs before I used them. XMITREM masked out unused bits from every nibble in the string (read-modify-write in a loop) before sending it to the IR-LED. The loop was set up incorrectly: it modified six nibbles BEHIND the string. If the string used up all of the room left in the TEMPOB area, it is possible that the topmost address in the RPL return stack would be modified. This 'sometime'-error (I hate them :-) could cause a Memory Clear. I have fixed this bug. Here are the corrected and modified programs : Program : XMIT3 Interface : ( $ --> ) Description : Transfer a string (sampled by Look !) to the IR-LED. Note : Originally written by David J. Marsh . Modified by DM. Feed only with strings sampled by Look !!! Chksum/Size : #7AB1 / 84 Program : LOOK Interface : ( $ --> $ ) Description : Sample data from the redeye into the given string. Note : Originally written by David J. Marsh . Modified by DM. Chksum/Size : #CB14 / 103 Program : MKSTR Interface : ( % --> $ ) Description : Quickly generate a % length empty string. Note : Rewritten by DM. Chksum/Size : #FC2B / 30 Have fun with it, 8-), Detlef. P.S. Our Phillips-TV needs a character string which is at least ~325 chars long to 'see' Xmit'ed commands (max. distance: with card bay closed - ~2 m, with card bay opened - ~3.5 m). DM ÚÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÂÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ¿ ³ `You mean this isn't ³ Detlef Mller ³ ³ the afterlife ?` ³ detlef@mwhh.hanse.de ³ ³ Arthur in THGTTG, part II, p. 78 ³...!uunet!mcsun!unido!mcshh!mwhh!detlef³ ÀÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÁÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÙ