[Sonnova]

On Thu, 10 Sep 2009 08:38:47 -0700, Arny Krueger wrote
(in article ):

"ST" wrote in message


3 days ago I decided to hook up my old Theta DS Pro Gen
III DAC to my Marantz SA11s2 just for the fun of it. To
my surprise it sounded better (more to my liking) then
SA11s2.

The most obvious difference among DACs is their output voltage. It often
varies over a range of several dB or more. A fractional dB difference in
level may not be perceived as a difference in loudness, but rather as some
other kind of quality difference.

Yep. When making any kind of sonic evaluation, one must be careful to match
all levels exactly. If one device under test is even subliminally louder than
another, that's the one that the ear will favor as being the better of the
devices under evaluation. This holds true for signal sources, amplifiying
devices as well as speakers.

Have you compared specs or done some bench checks to see that the output of
the two devices was the same within 0.1 dB?

That's difficult. Under real-world circumstances, I find that 0.25 dB is
about all one can realistically hope for.

Theta used Motorola 56001s,
30.0000mhz Rubidium clock, PCM63K and AD841 Op-Amp.

The PCM63K being one of the last non-oversampling, non sigma-delta DACs ever
to be used widely in high-performance equipment.

The PCM64K spec sheet says:

"More recently, DACs employing a different
architecture which utilizes noise shaping techniques and
very high oversampling frequencies, have been introduced
("Bitstream", "MASH", or 1-bit DACs)."

Which is true in a way. The copyright on the PCM64K spec sheet I have is
1990 which is probably when the part was introduced. By then oversampling
DACs had been around for almost a decade - the original Philips/Magnavox CD
player was oversampled.

If you are talking about the little Magnavox FD-1000, you are correct. It
used 4X oversampling and digital filtering and even though it was only only
14-bit, it sounded much better than the Sony CDP-101 or the Kyocera, or the
Denon or any of the other first generation players - it was also about half
the price of these other players and was built like a tank.

The spec sheet goes on to say:

"These (oversampled) DACs overcome
the low level linearity problem, but only at the expense
of signal-to-noise performance, and often to the detriment of
channel separation and intermodulation distortion if the
succeeding circuitry is not carefully designed."

If you understand where the oversampling technology has gone since then
(1990) and why, then you understand that oversampling *always* had the
promise of improved performance with vastly reduced costs, such as we see
today. The above paragraph seems to suggest that there were inherent
limitations with the basic oversampling technology, which we now know to be
completely untrue.

One of the other issues with the PCM64K is that it does not seem to include
any reconstruction filtering. Reconstruction filtering could even be
omitted. Reconstruction filtering, particularly that done in the analog
domain is generally very expensive if done well. There are even serious
questions whether or not it is feasible to do reconstruction filtering in
the analog domain as ideally as can be done economically in the digital
domain.

So my question is how could an old DAC out perform a new
reasonably well designed Marantz?

I question the evaluation technique that was used. It does not seem to
control *any* variables at all.

I also question whether all equipment involved was in fact in good operating
order and performed as well recently as it did on the best day of its life.
Bench testing would provide insights related to this issue.

Is it possible CD format technology already matured long
time ago and there's nothing new can really improve the
sound?

By the time the second generation of CD players such as the CDP 701 hit the
market, the audio side of the technology was pretty well perfected. Since
the mid-1980s, all that happened is that performance with suboptimal discs
and price/performance improved.

The first generation players such as the CDP 101 had very minor technical
failings that were very benign compared to the general level of technical
performance of audio equipment when they were introduced. For example, the
analog filters in the CDP 101 were subject to production variations and
could cause very mild dulling of high frequency sounds with certain program
material. The CDP 101 also shared the same DAC chip both temporally and
dynamically. The latter was not a problem the former could cause additional
minor dulling if an electrically-summed center channel was used.

CD players have improved considerably since those early units. But what has
improved more is the recording side. Getting rid of those awful Sony
PCM-1610, 1620 and 1630 processors that were practically de-riguer for early
CD production helped a lot. Those things were AWFUL!