by Keijo Tanskanen
Foreword
Many audio experts and enthusiasts have written about the benefits of DVD-A
and SACD. Nevertheless I would like to add my own thoughts about some of the
facts and points of views about these formats compared to Red Book. Why does
this debate deserve a reassessment? Because it seems the comparisons between
the formats are too often less than objective and in many cases comparing the
proverbial apples with oranges.
During several years now, I've had some doubts about the existence of any real audible differences between Red Book and high resolution formats. The reasons for this are obvious: I’m a telecommunications engineer; I’ve noticed that backgrounds of the listening tests and comparisons have been somewhat unclear and/or lacked the relevant reference points; and finally there have been undeniable proofs of “cheating”. After Mr. E. Brad Meyer and Mr. David R. Moran published their test [September 2007 issue of the Journal of the Audio Engineering Society (Volume 55, Number 9)] last year, I decided to write an article about my concerns and observations. In my opinion this test is so far the best done about the issue, although some questions remain open: Why the test was done in ABX mode and not in AB which would have possibly given more definitive results? Were the test participants familiar with the sound of the equipment used? Were all the test participants experienced audio enthusiasts? Anyway the result of the test is still very remarkable although the test arrangements could have been done even better. A good analysis of the test was done at theaudiocritic.com.
http://theaudiocritic.com/plog/index.php?op=ViewArticle&articleId=4&blogId=1
http://www.bostonaudiosociety.org/explanation.htm
When I looked at the
test diagram (not associated to the links above) I could see that in absolute
terms the test setup even gives a slight extra advantage to the SACD and DVD-A,
because the necessary amps and volume controls had been installed into the Red
Book line! Naturally the extra components, even very high quality ones, can
only distort the sound even more.
My Thoughts
Quality of standard
CDs
The sound quality of standard (= Red Book) CDs varies dramatically. Today there
are unfortunately several fashion crazes, especially in the mixing and mastering
processes of standard CDs. This, for example, means too much compression and
lack of equalization. I believe numerous recording companies are aiming for
good playback in average audio systems. The audiophiles and other serious listeners
are typically forgotten and I really don’t understand why? Why do audio
equipment vendors manufacture better audio equipment if the sound has already
been destroyed on the source disc? Luckily there are several recoding companies
which still make very high quality Red Book CDs, new or remastered ones.
It's also good to remember that if you compare badly made standard CDs to well done SACDs, the format comparison is naturally nonsense. The prominent aims to make more money in the SACD markets might have a negative impact on the mastering and dithering processes of standard layers!? Additionally one must remember that the hybrid discs are not Red Book discs! Even some of the commercially available “standard” CDs do not fulfil the Red Book standard when one inspects their thickness and other physical features!? This kind of careless CD production may cause more reading errors during the CD playback. In the worst case the results can be heard as a distortion if the error correction limits of the player have been regularly exceeded.
Listening experiences
and technical point of views
Basically I have had a similar level of enjoyment when listening to both very
high quality Red Book CDs and SACDs. That is one reason why I have my personal
listening based doubts about the real audible benefits of high resolution formats.
Additionally I know that very few analogue masters, on which most of the music
classics have been recorded, have any significant amount of information above
22 kHz and dynamics over 96 dB. Besides that, compression is widely used during
the mastering process. The average RMS power levels are typically between 10
- 20 dB among the very best standard rock CDs and probably 3 times that among
the best classical CDs. Naturally there are dynamic peaks which have clearly
wider dynamics, but still, especially in the case of rock music, where do we
need a format which gives over 96 dB dynamics?
Many enthusiasts argue that
the ultrasonic information, which only the high res formats can deliver, is
very significant. Those enthusiasts usually also present a couple of test results
which clearly show that ultrasonic information can be heard in certain situations.
That’s unquestionably true, but again there is other side of the issue.
Actually, in live performances the ultra sonic tones attenuate very abruptly
(attenuation is directly proportional to the distance of listeners). Still,
it is very questionable who really is able to hear ultrasonic harmonics even
in the near field.
Recording technology
I’m quite familiar with recording technology and I know the phases of
the recording processes as well. Basically it’s both the analog and digital
technology which set the limits to the final results of the audio discs. There
are numerous stages where the recordings can be spoiled: numerous items of audio
equipment (microphones, microphone amps, recorders, mixers, equalizers and other
editing tools, amplifiers, compressors and limiters, and tens of meters of cables
and power cords) must be traversed before the signal reaches the final format.
One can guess that noise only increases when going through the whole process!
Although the very first recordings are nowadays done with high resolution formats
(96/24 or 192/24 PCM), we really have only 20 bit resolution left for the music,
because the 4 LSB bits sample the noise which lies at -120 dB level! This is
why the S/N of the masters has such a dramatic effect on the real benefit of
high resolution discs. So, if faults occur during the recording processes the
sound quality game has already been lost and no format can save your day.
Still 20 bits is a lot and
4 bits more that 16 bits, but engineers are clever. They have come up with very
nice converters, dithering units and other methods. Using the finest dithering
units the audio engineers have even managed to get a signal at -120 dB level
listenable and clean! Considering this, where do we need high resolution formats?
I didn’t mention that the noise floor on bandwidth from 16 kHz to 22.1
kHz rises quite heavily because of the best quality dithering, but this is quite
acceptable because ears are not sensitive to that. So, a very low price is paid
for a dramatic improvement!
Is DSD technology better than PCM? That’s a tough question which I’m not able to answer in full. The jury of engineers is still out on that. DSD at least has some theoretical advantages over the Red Book format: its dynamic range is 120 dB and it has better resolution. When the whole recording process has been carefully done by DSD and the final master data has been written on a SACD disc, it gives an unquestionable advantage to the SACD layer. But, as mentioned before, this can give practical benefits only when the spectrum of music in master tapes and all the recording processes, especially mixing which still seems to be a tough challenge to DSD, makes it possible!
Quality of CD players
The differences may come also from the playback equipment. Every serious enthusiast
knows that the quality of standard CD players is uncertain, probably even more
than among SACD players. The listening results are crucially dependent on the
quality of the CD player. Additionally, more advanced technologies have been
developed for the sound improvements of standard CD-players. Especially upsampling
technology and better DA-converters have brought remarkable improvements, without
forgetting the other technologies and units. Naturally, I’m not saying
that there has been no development with regard to SACD technology! Surely there
has been, but I think greater improvements have been reached in the standard
technology.
Conclusion
This article is surely
not a definitive one, but it does try to reveal that the issue is complex and
that an objective comparison between the formats needs a significant amount
of knowledge, correct equipment and processes. I wish that people who are debating
the issue would be properly qualified and that comparisons and/or test backgrounds
would be objective.
Additionally I hope that the recording companies will always do their best during standard CD production (so far the standard CD capacity has not been fully used, for whatever reason) and standard CD player vendors will continue their production and development in the future. I imagine a great number of audio enthusiasts and audiophiles share my wishes.
In my opinion it is a secondary issue what format discs one owns and plays. The important matter is that one enjoys the music and this can undeniably happen when high quality Red Book CDs are played by a high quality CD player!
Some appropriate links for
further reference:
http://en.wikipedia.org/wiki/Red_Book_(audio_CD_standard)
http://en.wikipedia.org/wiki/PCM
http://en.wikipedia.org/wiki/Upsampling
http://en.wikipedia.org/wiki/Direct_Stream_Digital
http://www.aes.org/events/112/papers/x.cfm
http://www.smr-home-theatre.org/surround2002/technology/page_07.shtml
http://en.wikipedia.org/wiki/Dithering
http://www.positive-feedback.com/Issue25/dsd_olson.htm
http://mixonline.com/recording/mixing/audio_emperors_new_sampling/