Date: Sun, 16 Sep 2012 12:01:09 -0400 (EDT) From: Stuart Barkley <stuartb@4gh.net> To: freebsd-multimedia@freebsd.org Subject: Re: Sound system developement question Message-ID: <alpine.BSF.2.00.1209161053130.46761@freeman.4gh.net> In-Reply-To: <50547CA3.2010209@gmail.com> References: <4B739CF4-5D1D-4FE0-83FD-6987DCB40866@gmail.com> <CABzXLYNvsJm6QKPJga2FYFm_ssDW2cz9ZVrDfx5-DeMFW=0LAg@mail.gmail.com> <201209151453.41118.hselasky@c2i.net> <50547CA3.2010209@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
This message is in MIME format. The first part should be readable text,
while the remaining parts are likely unreadable without MIME-aware tools.
--4280880523-1718368019-1347809008=:46761
Content-Type: TEXT/PLAIN; CHARSET=ISO-8859-15
Content-Transfer-Encoding: 8BIT
Content-ID: <alpine.BSF.2.00.1209161124391.46761@freeman.4gh.net>
On Sat, 15 Sep 2012 at 09:03 -0000, Roberth Sjonøy wrote:
> Bitperfect function does that and much more, just what I want but
> bitperfect is one thing bits are perfect even if there is a lot of
> jitter.
Do you think you are hearing a problem? Is your D/A indicating
problems with jitter?
Your initial question is good. It would be a good regression test to
send data between two systems via a digital audio connection and
ensure that the transfer is bit-perfect. This would prove that the
system can operate without bit loss. However, this does not address
hardware quality issues in other systems.
I think you are either over thinking things or have been reading some
pretty old information. Jitter should not be a problem for most
modern digital audio connections.
This is more information to not take religiously. The details can be
complex and some is speculation on my part.
Older systems (> 10-15 years) could have problems and occasional
buffer underruns where more common (careful system tuning could
address most issues). On a properly functioning modern system FreeBSD
will keep the byte stream full to the audio chip. Buffer underruns
might still occur with modern systems with incorrect tuning but should
not happen with default system parameters.
Once at the audio chip, FreeBSD has little control over things and you
are at the mercy of your specific hardware.
The jitter is in the analog wave form of the digital signal. You can
be subject to jitter on the clocking source (quality of sound 'card'),
jitter on the transmission cable (quality of cable and connectors),
jitter in the D/A (quality of hardware), etc.
It sounds like you have an older D/A converter (not an oversampling
converter) where jitter on the S/PDIF cable matters. Be sure to use
quality cables with solid connectors. If you are very concerned about
this, buy "monster" brand cables. Consider replacing the D/A if
jitter is still causing you problems.
You only talked about jitter. If you are also seeking low latency
(delay between analog in to analog out), that is a different story and
can still cause issues as you reduce the audio buffering throughout
the system.
Stuart Barkley
--
I've never been lost; I was once bewildered for three days, but never lost!
-- Daniel Boone
--4280880523-1718368019-1347809008=:46761--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1209161053130.46761>