There are some difficulties with the large buffer approach. You mention
using a local clock that is the long term average of the input data
rate, again this is some form of adaptive.

If you use a fixed frequency clock you have the problem that you need a
fairly large buffer to handle an hours worth or so of music. That takes
very creative buffer management when people want to skip songs, change
playlists, fast forward etc. With a large enough buffer to handle
significant frequency deviations from external to internal clock AND
long playtimes before re-synching the lag in user interface could be
very objectionable to many people.

Its certainly an approach that should be looked into as an option, but
it has not been done very often. I tried this several years ago and the
buffer management issues became so messy I gave up.

And even the companies that have implemented such an approach have
still had interface jitter affect the output sound. It seems that even
with what should be theoretically completely immune getting the details
perfect is not easy.

My take as a designer i sI'm not going to aim for perfection. I'm going
to do my best to implement a technically robust interface method and do
my best to implement it well and let the chips fall where they will. I
have never achieved a DAC that is completely immune to outside
conditions and I probably never will, so I'm not going to worry about
it. The best done DACs sound so good I'm not going ot worry about a $500
cable making it sound slightly different. I would like to design
something that is not the case, but I've given up on that as not really
necessary for achieving  excellent sound. 

John S.


-- 
JohnSwenson
------------------------------------------------------------------------
JohnSwenson's Profile: http://forums.slimdevices.com/member.php?userid=5974
View this thread: http://forums.slimdevices.com/showthread.php?t=70626

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

Reply via email to