From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 7762 invoked by alias); 27 May 2005 00:43:29 -0000 Mailing-List: contact gcc-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Archive: List-Post: List-Help: Sender: gcc-owner@gcc.gnu.org Received: (qmail 7736 invoked by uid 22791); 27 May 2005 00:43:23 -0000 Received: from smtpout01-04.mesa1.secureserver.net (HELO smtpout01-04.mesa1.secureserver.net) (64.202.165.79) by sourceware.org (qpsmtpd/0.30-dev) with SMTP; Fri, 27 May 2005 00:43:23 +0000 Received: (qmail 22660 invoked from network); 27 May 2005 00:43:22 -0000 Received: from unknown (24.96.113.81) by smtpout01-04.mesa1.secureserver.net (64.202.165.79) with ESMTP; 27 May 2005 00:43:21 -0000 Message-ID: <42966CF1.1020708@coyotegulch.com> Date: Fri, 27 May 2005 00:54:00 -0000 From: Scott Robert Ladd User-Agent: Mozilla Thunderbird 1.0.2 (X11/20050512) MIME-Version: 1.0 To: "Menezes, Evandro" CC: Richard Henderson , gcc@gcc.gnu.org Subject: Re: Sine and Cosine Accuracy References: <84EA05E2CA77634C82730353CBE3A843028F45C1@SAUSEXMB1.amd.com> In-Reply-To: <84EA05E2CA77634C82730353CBE3A843028F45C1@SAUSEXMB1.amd.com> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit X-SW-Source: 2005-05/txt/msg01471.txt.bz2 Menezes, Evandro wrote: > Besides, you're also comparing 80-bit calculations with 64-bit > calculations, not only the accuracy of sin and cos. Try using > -ffloat-store along with -mfpmath=387 and see yet another set of > results. At the end of the day, which one do you trust? I wouldn't > trust my check balance to x87 microcode... ;-) I wouldn;t trust my bank accounts to the x87 under any circumstances; anyone doing exact math should be using fixed-point. Different programs have different requirements. I don't understand why GCC needs to be one-size fits all, when it could be *better* than the competition by taking a broader and more flexible view. ..Scott