Return-Path: Received: (qmail 1883 invoked from network); 8 Aug 2002 21:57:17 -0000 Received: from marstons.services.quay.plus.net (212.159.14.223) by mailstore with SMTP; 8 Aug 2002 21:57:17 -0000 Received: (qmail 7258 invoked by uid 10001); 8 Aug 2002 22:02:11 -0000 Received: from post.thorcom.com (193.82.116.70) by marstons.services.quay.plus.net with SMTP; 8 Aug 2002 22:02:11 -0000 X-SQ: A Received: from majordom by post.thorcom.com with local (Exim 3.33 #2) id 17cwzZ-0000yK-00 for rsgb_lf_group-outgoing@blacksheep.org; Fri, 09 Aug 2002 00:46:57 +0100 Received: from fep02.swip.net ([130.244.199.130] helo=fep02-svc.swip.net) by post.thorcom.com with esmtp (Exim 3.33 #2) id 17cwzY-0000yF-00 for rsgb_lf_group@blacksheep.org; Fri, 09 Aug 2002 00:46:56 +0100 Received: from oemcomputer ([213.101.37.22]) by fep02-svc.swip.net with SMTP id <20020808215445.SSFX8099.fep02-svc.swip.net@oemcomputer> for ; Thu, 8 Aug 2002 23:54:45 +0200 Message-ID: <005601c23f26$3cd69c40$162565d5@oemcomputer> From: "Johan Bodin" To: rsgb_lf_group@blacksheep.org References: <5.1.0.14.0.20020808100313.00b00a88@gemini.herts.ac.uk> Subject: LF: Re: Re: Jason Tests / Reasons for different soundcard sensitivity Date: Thu, 8 Aug 2002 23:37:45 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1; format=flowed Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2600.0000 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Precedence: bulk Reply-To: rsgb_lf_group@blacksheep.org X-Listname: rsgb_lf_group Sender: Jim Moritz wrote: > The curious thing is that if clipping occurs before the signal reaches > the sound card, the results are better than if clipping occurs in the > sound card - I would expect the effect of clipping would be similar > wherever it occurred between the IF filter and software FFT algorithm, > but apparently not so. Hmm... This is surprising, interesting and confusing... Could it be a strange kind of ADC or some DSP stuff on that soundcard that "wraps around" instead of clip at 0x7FFF and 0x8000? I was playing with an ADSP2181 DSP a few years ago and I before I got the filter coefficients scaled properly, I had problems with overflow in FIR filters. The "summing point" overflowed resulting in a _horrible_ sound. Fortunately I found a bit called something like "enable saturation" which made the MAC accumulator clip instead of wrap around. 73 Johan SM6LKM