Phil Leigh;578558 Wrote: 
> A -90dB sine wave is just "not there" in normal music. I'd like to hear
> JA's justification of why he thinks this is meaningful?

Just saw this question. If you are talking about the undithered tone at
-90.31dBFS I show in all my DAC measurements, I use this signal, not
because it will be typical of music, but because it is diagnostic for
DAC problems. In 2s-complement PCM encoding as used on CD, the change
in level from digital "0" to +1 LSB actually involves all bits in the
16-bit word changing value. By contrast, the change from "0" to -1 LSB
involves just the LSB changing value. So you can immediately see from
the symmetry of the waveform, and how well-defined the 3 voltage levels
are, whether or not the DAC is sufficiently monotonic and the analog
noisefloor is sufficiently low to allow those data to be correctly
decoded.

In other words, it's a thumbnail indicator of DAC quality.

In general, the test is not as useful in these days of sigma-delta DACs
as it was when all we had were multi-bit DACs. However, you still come
across pathological designs that fail to reproduce this signal properly
- see, for example, fig.8 in the Cary review in the September 2010 issue
of Stereophile.

John Atkinson
Editor, Stereophile


-- 
Stereoeditor
------------------------------------------------------------------------
Stereoeditor's Profile: http://forums.slimdevices.com/member.php?userid=40770
View this thread: http://forums.slimdevices.com/showthread.php?t=82050

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/audiophiles

Reply via email to