No sooner had I gotten the "your message has been sent to a
million people" confirmation then in comes Payson's solution.
But then he's always chomping at the bit (of course the pun was
intended). His solution ties the hinted to 10 point solution.
Can anyone do better?

Return-Path: <supercat@Mercury.mcs.net>
Received: from Kitten.mcs.com by unix.SRI.COM (SMI-8.6/SMI-SVR4)
        id MAA23485; Tue, 2 Sep 1997 12:32:09 -0700
Received: from Mercury.mcs.net (supercat@Mercury.mcs.net [192.160.127.80]) by
 Kitten.mcs.com (8.8.5/8.8.2) with ESMTP id OAA06224 for
 <sdattalo@unix.SRI.COM>; Tue, 2 Sep 1997 14:32:06 -0500 (CDT)
Received: (from supercat@localhost) by Mercury.mcs.net (8.8.5/8.8.2) id OAA03475
 for sdattalo@unix.SRI.COM; Tue, 2 Sep 1997 14:32:05 -0500 (CDT)
From: John Payson <supercat@Mcs.Net>
Message-Id: <199709021932.OAA03475@Mercury.mcs.net>
Subject: Re: AND how fast?
To: sdattalo@unix.SRI.COM
Date: Tue, 2 Sep 1997 14:32:05 -0500 (CDT)
In-Reply-To:  <340C658D.DD@unix.sri.com> from "Scott Dattalo" at Sep 2, 97
 12:14:21 pm
X-Mailer: ELM [version 2.4 PL24]
Content-Type: text


> How fast can you logically AND two bits?
>
> Given:
>  bit_a  of register A
>  bit_b  of register B
>  bit_c  of register C
>
> The goal in C parlance is:

        bsf     C
        btfsc   A
         btfss  B
         bcf    C

>    1) Size: 2*(6 - # of execution cycles)
>    2) Isosynchonous: 3 points
>    3) Preserves W: 2 points
>    4) Preserves reg A and B: 1 point

I think that's 10pts?