Some examples are
(1) direct drive vs belt drive turntable - the few "best" sounding on the market are still belt driven ones (i.e. Linn Sondek, Oracle ...). These turntables
match the best direct drive turntables on paper and they do sound better.
For sure, direct drive turntables are, in general, very low in noise
figures and wow & flutter etc. However, men do not listen on specs.
As a matter of fact, it is widely known that direct drive turntables
generate a lot of "noise" in the sub-sonic region which is filtered
out during measurements. These sub-sonic noises do affect signals at
(2) we were told that the so called NEW CLASS A amplifiers would be THE
design for all future amplifiers since they are more efficient than
pure class A amps while they do not switch as class AB amps do.
However, the first NEW CLASS A amp (from Threshold) which was aimed
at the high end market is a flop and none of the NEW CLASS A amps
from Japan (from where 90% of the amps are using the NEW CLASS A design)
can make it in the high end market. I listened to some medium priced
New Class A amps about two years ago and I was not impressed.
Now, we have the digital technology on the horizon. Digital technology does
provide high dynamic range, low noise, low distortion ... etc. However,
will it replace analogue records, say, within 5 years? I don't think
so. Sure, in long term, digital technology will become the main stream
and analogue records will be completely wiped out, but that will take at
least 10 or more years.
There are still many obstacles and problems, to name a few:
(1) price, you can buy a phono record player for less than $200 but
I can't perceive the price of a digital record player can drop to such price (at least not in 5 years)
(2) standardization, take a look at 4 channel technology, it killed itself
by having more than 3 systems on the market. The same applies to digital
records, there are at least two systems from Japan and Philips Compact Disc. Although it seems that Philips will win out just as they did when cassettes
were introduced, the standardization on tape is not yet agreed upon.
(The standardization in Japan is not yet made international)
(3) D/A conversion is still the weakest link in the system, much has to be
improved in this respect. The same applies to the anti-aliasing circuit:
in theory, a wall filter is needed to filter out signals above the
frequency = 1/2 of the sampling rate. Obviously, this means a very high
order low pass filter is needed => high cost + accurate alignment.
Another disadvantage of higher order filters is that they ring like hell.
Philips tackled this problem by using a mixture of digital filtering +
a third order low pass filter. This seems quite effective, but digital
filter is quite expensive.
(4) the theoretical 96 db S/N (16 bits) is quite optimistic - noises are
generated from the A/D, D/A circuits and, as a matter of fact, white noise
is added to the signal before it is converted to digital. This is needed in
order to eliminate the problem in digitizing very low level signals. Thus
a S/N figure of 70 db is more realistic.
(5) home digital taping (from analogue) source can become quite frustrating:
you have to watch the recording level very carefully since
when a digital recording clips, it does clip like crazy. The increase in
distortion is almost exponential.
(6) the "fixed number" system being used has a characteristic which is
substantially different from analogue system - higher distortion at low
level and low distortion at high level. This may not be desirable since
most of the time, music signals are at a relatively low level (for
example, in a domestic environment, the amplifier is working at the region
of 5 to 10 watts at "normal" listening level while the output may
jump to 70 to 100 watts (provided the amp is powerful enough) during
the peak of the music)
To make my point clear, I am NOT against digital technology, I am only being
cautious. I think there is an urgent need in standardization and room must
be left for future development. For example, the Japanese standard called for
a single sampling rate of 44.something KHZ, this may not be acceptable by most
audiophile. The "fixed number" digitization is deficient,
I think a "floating number" system is much better (although more expensive).
I won't be the first to buy a digital player/recorder, but definitely,
I will move into it only when the technology is proven to be superior to
the analogue technology.
Replying to the article which bears this article's head.
1) Direct drive turntables do NOT have more low frequency noise than
belt drive turntables. What they DO have is "cogging" distortion, so
named for the stepping motor's tendency to provide an impulse every
t sub s seconds. This impulse does two things:
a) It propigates through the entire platter system, causing noise
pickup at a frequency which is NOT low, and
b) It causes a very slight FM modulation of the material being
reproduced. This effect "seems" to be the cause of most dislike of
direct drive turntable, in fact, it is responsible for the 'unique' sound
of several of the direct drive breed. There are a few turntables using
quadrature drive that do not have this problem. Strangely enough, they
sound a lot like a belt drive turntable. The IM and sideband properties of
this kind of modulation are mind boggling. Try it yourself.
Non-the-less, I still wonder what is wrong with a servo'ed BELT drive.
It has the advantages of both systems, and few of the problems. Its only
problem is its belt, and that is common to all belt drives (funny thing!).
2) "Class A" amplifiers have been touted in the high end market for a long
time. REgardless of advertising noise, class a is class a. If the output
device doesn't conduct linearly through the whole cycle, it ain't class
a. There have been several important new designs in the biasing of NON
class A amplifiers lately, but these amps are not in the high end market,
they don't need to be. There is NO reason why any amplifier has to be in
the high end market.
3) There are several untruths listed under this head. I will cover a couple.
a). The filter doesn't have to have an instantanious change from
on to off. In most cases, the sampling rates are chosen to be ABOVE! the
Nyquist frequency far enough that the filters are not that difficult to design.
In fact, that is usually the one thing that constrains the sampling rate.
b). D/A conversion, since the advent of the digital audio "threat"
has been pushed quite a ways. The D/A that cost 300 $ ten years ago now
costs $80 and works better. If a 16 bit D/A introduces noise above
-96dB, it isn't a 16 bit D/A, by definition. The A/D and D/A introduce
noise only as they quantize the signal to 16 bits. If they do otherwise,
they should be replaced with a relatively inexpensive unit that WILL work
right. So much for that. The signal coming into the A/D may not have 16 bits
resolution, but it isn't any worse than the same signal that is being used
for disc or tape, so that arugement is handwaving.
c). The idea that digital has a lower signal to noise ratio for lower
energy signals is true. So does everything else. If the Signal to Noise
Ratio (SNR) is better, then the noise will be less troublesome, if the noise
is of the same (white) type. Disc's, tape, and the like have exactly the same
behavior, except that the signal disappears at a higher level, indicating that
the noise is causing MORE damage.
As far as standardazation, I couldn't agree more. There are several
systems (Phillips has suporters in Japan, BTW) where there should be
one. The reasons don't even have to do with competition. I could flame
for hours about that.
On the other hand, quad doesn't have much to do with digital, as far as
public acceptance is concerned. I have listened to quad setups, and I don't
like them very much. A proper stereo system sounds more realistic to me
than a quad recording, even including ambiance. I'm not convinced
that quad is better. I am convinced that a record that doesn't lose high
frequencies when it's played, doesn't automatically introduce 3% distortion,
and is washable if it gets dirty, has much to offer. Especially since
it sounds so much better.
Now-- If we could only raise Stowkowski for a few recording sessions.
J.D.Johnston rabbit!jj BTL/MH