I knew someone was going to ask this! 

The above post was actually an oversimplification, in actuality the
impedance can vary with frequency. Its similar to saying a speaker is 8
ohms, very rarely is it 8 ohms at all frequencies in the audio range.
Some reviews will publish the full graph of speaker impedance VS
frequency, but how does a consumer use this? How can a consumer figure
out whether this speaker is going to sound  better than another using a
graph for each? 

The same sort of thing applies to S/PDIF systems. IF you want full
information you need a impedance VS frequency chart (covering at least
1MHz to 200MHz). So you have one of those for the source and one for the
connecting cable and one for the receiver. What do you DO with this
information? A human can probably do some gross interpretaions with
these, particularly if the spread for each one is not too large, but if
each one has a large spread, how do you make meaningfull interpretations
of how the system  as whole is going to behave? 

There ARE ways to  take this  information into a computer and simulate
what the total SYSTEM behavior is going to be, but then you have to know
how the particular DAC product is  sensitive to this, different ones are
sensitive in different ways. Once a group of people have discovered the
sensitivities of a particular DAC then this information might be useful.
(note that very few manufacturers even think about any of this stuff, so
information on this is NOT going to come from a manufacturer)

There certainly are test equipment that can measure this, the telecom
industry uses this stuff all the time, but they are not cheap, $20,000
to $100,000. It should be possible to design a piece of fairly
inexpensive test equipment that has a USB interface to a computer so the
computer can do all the heavy lifting and have it just do this one
function over just the frequency range we are interested in.
Unfortunately the market will be small, so even if technically it is not
too difficult, it will still be fairly expensive. Somebody could
probably market such a device for $4k-$5k. If a bunch of DIY types get
together to do the circuit design and programming it could be a DIY
project for a lot less. Note that DIY, open source, cooperative projects
for test equipment are quite rare!

It would be interesting to see what would happen in the industry  if
reviwers started publishing this information. My guess is that LONG term
it might  be usefull, manufacturers might actually start measuring their
products and put some effort into getting  things closer to spec. But
short term it might be a major  problem because most  consumers would
have no clue what it all meant, the market (audiophiles and reviewers)
would wind up fixating on some aspect to make comparisons on, which may
or may not have any relationship to something important. The important
thing would  be to get the  market to focus on what is primarily
important, low spread, and averaging around 75 ohms. This is something 
that is very obvious by  looking at the impedance VS frequency charts,
as long as all the charts are done the same way, have the same units
etc. 

For a particular peice of equipment built using surface mount parts on
PC  boards the impedance is going to be quite uniform from box to box as
long as the board design is not changed and the components stay the
same. Changing a chip from one manufacturer toanother CAN have a
significant change impedance, since that is usually not a parameter that
is "tied" to the part number.  


Yes it IS possible in  many cases to build a device that you can insert
between boxes to improve the match, but they would have to be engineered
for a specific combination, there is no such thing as one box that you
can use that does everything. I suppose it's possible to build a device
that measures the system and configures a network to optimize it, but
such a device would be exceedingly expensive. It would be way cheaper to
just build the sources and recivers in such a way that they don't care.


John S.


------------------------------------------------------------------------
JohnSwenson's Profile: http://forums.slimdevices.com/member.php?userid=5974
View this thread: http://forums.slimdevices.com/showthread.php?t=95614

_______________________________________________
audiophiles mailing list
audiophiles@lists.slimdevices.com
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to