Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 22 Dec 1997 23:14:17 -0800
From:      Amancio Hasty <hasty@rah.star-gate.com>
To:        Luigi Rizzo <luigi@labinfo.iet.unipi.it>
Cc:        steve@visint.co.uk, multimedia@freebsd.org
Subject:   Re: precise soundcard tuning ? 
Message-ID:  <199712230714.XAA00430@rah.star-gate.com>
In-Reply-To: Your message of "Tue, 23 Dec 1997 05:23:35 %2B0100." <199712230423.FAA03728@labinfo.iet.unipi.it> 

next in thread | previous in thread | raw e-mail | index | archive | help

After about 7 minutes or so:
3112000 388.919 8001.66
3120000 389.919 8001.66
3128000 390.919 8001.66
3136000 391.919 8001.66
3144000 392.918 8001.66
3152000 393.918 8001.66
3160000 394.918 8001.66
3168000 395.918 8001.66
3176000 396.917 8001.66
3184000 397.917 8001.66
3192000 398.917 8001.66
3200000 399.917 8001.66

In order to really measure my sound accuracy I have to find out what is
the clock drift on my PC.

Thats fine I don't think that 8000 is an exact divisior of XTAL1 or
XTAL2.

Your point that different frequencies may not be as accurate is well
taken . A lot depends on how accurate is the clock in the sound card.


	Cheers,
	Amancio




> > Strange, according to Jim's test program my guspnp is about
> > .002 accurate.
> > 520000 64.9858 8001.75
> > 528000 65.9856 8001.75
> > 536000 66.9854 8001.74
> 
> quite reasonable result (I assume that you have run the test for a
> much longer time,  or the 200us differences that you see are not that
> significant).
> 
> But the problem is that some sampling rates might not be so precise
> (due to the sample rate not being an exact divisor of the XTAL
> frequency) and these differences accumulate over time. The problem is
> often worse at high sampling rates (where you need a smaller divisor).
> Even in your case, you have a drift of roughly 1 second/hour.
> 
> 	Cheers
> 	Luigi





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199712230714.XAA00430>