From: "M. Sanders \(PA3BSH\)" <misan@xs4all.nl>
To: rsgb_lf_group@blacksheep.org
Subject: LF: RE: Sound cards.
Date: Sat, 8 Jan 2000 10:33:24 +0100
Message-ID: <000001bf59bb$691e19c0$67976dc2@v3q4v1>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1; format=flowed
Content-Transfer-Encoding: 8bit
Importance: Normal
In-reply-to: <c8.c8ef2e89.25a80935@aol.com>
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org
Sender: <majordom@post.thorcom.com>

Hello David,

The 5.5k mode workes with my PII-350 and 'SB-live' soundcard but there is a
problem with the analysis. The sample+processing time needed for 2048
samples or more exceeds the real time! therefore the computer probably
calculates with data from memory (previous calculated samples). The results
are instable and look like oscillating or reappearing sounds on screeen that
should not have been there. Not unlike the behaviour of ringing filters or
IIR digital filters with critical parameters. So there is a limit for the
analysis with the method used by R. Horne in Spectogram. You computer
apparently does not cope with the amount of calculations needed and crashes.
I use 11k /16384 samples/FFT size because it seems to be the most effective
mode in Spectogram.

Greetings,

Michael Sanders, PA3BSH


> -----Oorspronkelijk bericht-----
> Van: majordom@post.thorcom.com [mailto:majordom@post.thorcom.com]namens
> G0MRF@aol.com
> Verzonden: zaterdag 8 januari 2000 4:30
> Aan: rsgb_lf_group@blacksheep.org
> Onderwerp: LF: Sound cards.
>
>
> Hello All.
>
> Does anyone have a "fix" for a fault that causes spectrogram (or
> windows) to
> crash when it's started up on 5k5 sample rate?
>
> It appears to happen irrespectively of the number of FFT points
> selected. It
> may be a soundcard problem but I've seen this on a couple of
> computers now
> and it's always the most sensitive setting that causes the problem.
>
> 73
>
>
> David  G0MRF
>
>