Hi,

I am using a creative Audigy 2 that connects to my AV reciever via a
homebrew toslink io connector. It has worked perfectly for 2 years, but
recently, I have started noticing glitches in the sound, manifested as pops
and clicks in PCM content, and squeaks in AC3 content, as well as poor
detection of AC3 content on my reciever.

I am keen to try to diagnose the source of these problems, and would like to
try some loopback tests. The kX audio drivers that I am using allow me to
record audio from a digital input without any processing, so I was hoping to
be able to record the bit error rate for differing optical cables, processor
loads and running applications. Is anyone aware of any free, or trial
version software that will allow me to perform these tests?

My current toslink assembly uses a TOTX173 and TORX173. If I have to
replace this assembly, should I be using the same components again, or are
the Sharp ones any better quality / more reliable. Is there anything at all
to choose between these two, or indeed any other manufacturer's? Can you
recommend a source for single units of these parts within the UK?. The
current ones were supplied as a sample, and I can't obtain any more.


Many thanks,

Chris Key

"Christopher Key" wrote in message

Hi,

I am using a creative Audigy 2 that connects to my AV reciever via a
homebrew toslink io connector. It has worked perfectly for 2 years,
but recently, I have started noticing glitches in the sound,
manifested as pops and clicks in PCM content, and squeaks in AC3
content, as well as poor detection of AC3 content on my reciever.

I am keen to try to diagnose the source of these problems, and would
like to try some loopback tests. The kX audio drivers that I am
using allow me to record audio from a digital input without any
processing, so I was hoping to be able to record the bit error rate
for differing optical cables, processor loads and running
applications. Is anyone aware of any free, or trial version software
that will allow me to perform these tests?

For performing conventional audio technical tests of soundcards, little
beats the freeware "Audio Rightmark" program.

Hi,

Thanks for the ultraquick response.


For performing conventional audio technical tests of soundcards, little
beats the freeware "Audio Rightmark" program.

I had a play with Audio Rightmark. I ran the tests with an optical loopback
connection, and the drivers configured for bit perfect recording (proven by
recording and playing back AC3 audio). I get slightly varying results each
time I run the test, so I suspect some resampling is occuring in the chain
on the output side. It'd be nice if RMAA had a function to 'lie' to the
operating system, pretending that the audio it was outputing was non PCM, so
as to prevent any resampling (Much the same way as software DVD players do
with AC3 or DTS sound).

I supose my original question still stands; Can you recommend a way to
accurately measure the BER between the output from the software, and
recording, for preference with realtime reporting?


Many thanks,

Chris Key





Below are the results obtained by RMAA, two using MME, and two with Direct
Sound. Both had the normamisation option disabled.

Frequency response (from 40 Hz to 15 kHz), dB: +0.00, -0.00
+0.00, -0.00 +0.00, -0.00 +0.00, -0.00
Noise level, dB (A): -86.4 -85.4 -87.1 -85.3
Dynamic range, dB (A): 86.1 89.6 84.4 86.4
THD, %: 0.0013 0.0010 0.0010 0.0008
IMD, %: 0.0088 0.011 0.013 0.0091
Stereo crosstalk, dB: -87.3 -86.3 -87.1 -87.1

Christopher Key wrote:

My current toslink assembly uses a TOTX173 and TORX173. If I have to
replace this assembly, should I be using the same components again, or are
the Sharp ones any better quality / more reliable. Is there anything at all
to choose between these two, or indeed any other manufacturer's? Can you
recommend a source for single units of these parts within the UK?. The
current ones were supplied as a sample, and I can't obtain any more.

A red LED, a transistor, and a resistor ought to suffice...

A red LED, a transistor, and a resistor ought to suffice...

True, but the mechanical connection is pretty awkward.