herbie@watdcsu.UUCP (Herb Chong [DCS]) (06/17/85)
this article is abstracted from the August, 1980 issue of Audioscene
Canada and because of copyright restrictions, the number of graphs and
charts, and the size of the article, was paraphrased to reduce the
number of words without losing content and hopefully not biasing the
report. this part describes the second in a series of 3 tests
performed.
the full reference is
Floyd E. Toole, "A Comparison of Images and Realities",
Audioscene Canada, Volume 17, Number 8, August, 1980,
McLean-Hunter Publishing Company.
Dr. Toole works for the National Research Council in Ottawa and is a
contributing editor to Audioscene. he is a frequent contributor to the
JAES (for a more complete biography, see issue number 4 of this year's
JAES). many people consider the NRC audio lab to be the finest in the
world. it certainly has the largest and best anechoic chamber.
i will place the paraphrased material with no indenting and my comments
indented and enclosed in square brackets []. direct quotes will be
withing quotation marks "". hopefully, some useful discussion and
insight can be obtained from this posting. the original section titles
are used where possible. the article contains the results from the
last two of 3 tests on MC and MM cartridges. i do not have the first
article and i have never read it, but i am lead to believe that this
one is the more important of the two.
--------------------------------
[summary of report and previous report]
in the beginning, a small group of audiophiles were gathered together
to listen to cartridges under controlled conditions. many cartridges
were auditioned covering a wide variaty of price ranges. the
conclusions were
"...there were no audible differences that could not be
accounted for by differences in measured frequency response and
tracking ability, and nothing emerged that seemed unique to
either moving coil or fixed coil cartridges. several moving
coil devices exhibited energetic resonances at very high
frequencies, a feature that made them sound relatively bright
and added 'definition' and 'detail'... ".
the equipment
modified Sony TTS-4000
special hard rubber platter mat
3 SME 3009 III's used without damping
special turntable base weighing over 150 pounds with extremely
high isolation and damping
technics SU-9070 pre-amps
amcron (crown to you US types) DC-300A power amplifier
specially matched yamaha NS-1000M speakers
[i have detailed specs on the SME, techics, and yamaha units.
i am not familiar with the sony table used, but i gather that
is was one of their high end units, comparable to the esprit
series now in production.]
overall frequency response was within 0.2 dB for the electronics and
the speakers. the cartridges were set up using the recommended
techniques for doing so when a full audio test system is available to
do so and all cartridge loads were set to manufacturer recommended.
when the cartridges were mounted, they were adjusted to be within 0.1
dB at 1 kHz. listeners were seated in progressively higher chairs
placed along the line of symmetry between the speakers. the speakers
were placed to minimize the room effects. noise was extremely low and
room was designed as a stereo listening room.
the procedure
the listeners were instructed to not talk to each other, to listen for
differences only, and to remain as stationary as possible. listening
groups of up to 3 were auditioning at one time. 3 cartridges at a time
were listened to, with the listeners unable to tell which was which. a
separate person hidden from the rest switched between the cartridges
every 5 to 10 seconds. the selections were cued with slight time
delays between the cartridges and the first compared with the last
always. each listening round lasted about 30 minutes and then the
cartridge order was switched until all cartridges had a chance to be
lead cartridge. the listeners were told that only 3 cartridges were in
the test.
listeners graded the cartridges on various things such as sound quality
and imaging. though the listeners were told to listen for differences
and mark them that way, listener preferences could not be completely
eliminated.
before the tests, a list of 9 cartridges were handed out and the
listeners were to comment on performance and also to rank those
cartridges from poor to excellent, and for the special few, State of
the Art. from these 9, 3 would be chosen for the tests.
the products
based upon reputation, design, and measured excellence, 3 cartridges
were selected for evaluation. two were MC cartridges listing at over
$500 CAN each (over $1000 CAN with matching transformers) and one MM
listing at just over $200 CAN. the cartridges will be refered to
as MC-1, MC-2, and MM, respectively.
[if i remember correctly, the exchange rate for CAN/US was
about 0.85 or so then.]
the images
the results of the questionaires showed both MC cartridges classed as
excellent or SOTA by all but 1 of the listeners. the MM rated from
poor to SOTA in an almost symetric distribution. there were severe
comments associated with many of the poor ratings. given this, one
would expect the differences between the MC and MM cartridges to be
easily discernable.
[this is in line with my experiences with my Dynavector DV-23R,
my Yamaha MC-1s, and my Shure V15 Type IV. the five ratings
were poor, fair, good, excellent, and SOTA.]
the realities
sixteen listeners spent 40 man-hours to come up with no significant
differences between the products, based upn preference scores. yes,
there were differences, and yes there were preferences, but they all
balanced out in the end.
MC-1 produced a noncommittal response. There was a strong
equal-preference response from the listeners with almost equal
accepting and rejecting responses. MC-2 were biased toward the
rejection side of the equal preference point, with the exception of two
listeners who consistently prefered it. MM reponses were biased toward
the acceptance side of the preference scale who consistently rejected
the product.
with regard to stereo imaging, even less differences were indicated
between the cartridges, with the exception of two who maintained their
strong opinions of MC-2 and MM.
the comments of the listeners were examined for clues as to the reasons
for the differences in perceived sound quality. they could be broken
down into two classes: bass and treble. over 90% of the responses
refered in some way to high frequency performance. the two types of
reponses for high frequency differences were generalize to 1) too much
high frequencies, and 2) not enough high frequencies. some comments
indicated that the differences were more than just comparative
(brighter) but were excessive (too bright). the listeners were asked
to indicate on a small-medium-large scale the magnitude of the
differences. none were classified as large, so an index of the
difference was taken to be the percentage of the responses that were
classified a medium.
[the frequency response comments are relative responses. thus
1 vs 2 means that 2 was assumed as the reference and 1's
difference from 2 was plotted in the graphs he refers to.]
MC-1 vs MM - the difference index was 0.18, indicating small
differences between the two cartridges. only two comments about bass
response were made, that MC-1 was tighter than MM. this contradicts
the frequency response measurements made on the cartridges. the
relative rise in MC-1 was only 2 dB at 20 Hz, so not easily audible.
differences in high frequencies was split almost evenly between
too-bright and not-bright-enough.
MC-2 vs MM - the difference index was 0.47, so about half the people
thought the differences were moderate. the few comments on bass
reponse were split evenly between tighter and looser. the measured low
frequency response was within 0.1 dB until 1kHz. 82% of the remaining
comments indicated that MC-2 was brighter than MM and that the
brightness was excessive. on the other hand, there were two listeners
who prefered the brightness and said that MM sounded dull. the
dominant characteristic of MC-2 is a rising high end, relative to
MM, which was criticized by some of the listeners. the two opinions
probably reflect personal taste, listening experience, or high
frequency hearing loss. the two listeners who prefered the brightness
were among the most experienced listeners in the test and one
acknowledged a preference for "bright" sound even though he knew it was
not always realistic.
MC-2 vs MC-1 - the difference index was 0.56, indicating that half the
listeners felt that the differences were moderate. the three comments
on bass response indicated that MC-2 had tighter bass than MC-1. this
agrees with the relative response curves. MC-1 is slightly bass-heavy
compared to MC-2, which was almost flat. 88% of the remaining comments
noted that MC-2 was brighter than MC-1 and of these, 3/4 found the
brightness to be excessive. the repsonse measurements indicate that
MC-2 has a slight trough between 3k and 10k and a shallow rise above
10k relative to MC-1.
stereo imaging
"it is an interesting trait of the human ego that most of us, most of
the time, know how certain things *should* be. stereo is no
exception." despite multi-miking techiques, "many people persist in
believing that two speakers in a room are capable of rendering
positions, dimensions, and ambience that, to them, *should* be
perceived as realistic. fortunately, humans have good imaginations,
because occasionally this miserably imperfect system works amazingly
well."
listeners had problems and there were few comments. for the most part,
differences were simply not heard, as indicated by the many
equal-preference ratings. the differences noted fell into three
categories: depth, width, and imaging. the difference index for all 3
comparisons was about 0.4, so about 40% of the listeners judged the
differences to be moderate.
MC-1 vs MM - responses to depth were about equally divided between
shallower and deeper. width slightly favored MC-1. imaging was
about equal between better and worse.
MC-2 vs MM - equally divided between shallower and deeper, slightly
favored wider, and equal for better and worse.
MC-2 vs MC-1 - slight differences only. less depth, wider, and worse.
imaging judgements were inconsistent and small. wider image was usually
given to the cartridge with the higher output above 10k, although difference
was small.
[this correlates with other psychoacoustic results where the
direction a tone comes from is purely a function of frequency.
the higher the tone (starting at about 400 Hz), the higher it
appears to be coming from. for some people, a 8kHz tone will
appear to be coming from directly above, no matter where the
actual sound source is. directly horizontal and straight ahead
for most people is about 1kHz.]
a statistical comment
preference scores were computed for the listeners who expressed a
preference. the computed Z values were -0.03, +0.03, and 0.0.
statistically, these results could have occurred of the listeners had
not shown up and mailed their responses in without knowing what
cartridges they were going to audition.
[the Z values are the ordinates on the normal or gaussian
curve.]
reflections
most people who took the test believed that there would be changes in
sound quality and imaging that would be easy to hear and classify.
such was not the case for the three cartridges chosen. the differences
were audible only with certain kinds of music, only on certain
recordings, and in magnitude, were hardly the stark contrasts that
several listeners anticipated. in fact, a few listeners were simply
staggered to find that they had been complimentary to a product that
previously they had held in low regard.
naturally, there were a few indomitable spirits who remained
unconvinced, alluding to the choice of speakers, amplifiers, or
tonearms as reasons for the unexpected result.
[i don't agree completely with these conclusions. there were
two people who could consistently pick the MC cartridges. at
least to some of the people, the differences are there and
clearly audible. it should be noted that their preferences
were for the more flawed reproduction rather than the more
accurate one though.]
[this concludes part 1 of the MC vs MM cartridge test. part 2
will be posted when i get a chance to type it in. i will not
reveal the cartridges tested now except to say that i own one
of them and have auditioned one of the others.]
Herb Chong...
I'm user-friendly -- I don't byte, I nybble....
UUCP: {decvax|utzoo|ihnp4|allegra|clyde}!watmath!water!watdcsu!herbie
CSNET: herbie%watdcsu@waterloo.csnet
ARPA: herbie%watdcsu%waterloo.csnet@csnet-relay.arpa
NETNORTH, BITNET, EARN: herbie@watdcs, herbie@watdcsu