pmr@drufl.UUCP (Rastocny) (02/08/84)
RESULTS OF THE AT&T AUDIO CLUB SPEAKER WIRE EVALUATION Remember last November the proposed test of six different types of speaker wires? Well, the Club finally got together and performed the double-blind test. A quick summary of the test is outlined below. Wires (in alphabetical order, not in actual order during the test): A. Home-made 4-wire pseudo-Litz, 25' B. Kimber Cable, 15' C. Monster Cable, 12' D. Monster Cable Power Line, 15' E. Saxton, 15' F. #18 zip cord, 15' Sound pressure between the most resistive wire (#F) and the least resistive wire (#E) were measured at less than 0.3dB using a General Radio broad-band SPL meter and pink noise. Wires were switched manually in about 15 seconds. No switching system was used since the Rdc of the wires was far LESS than the contact resistance of the best available switches. (8-40 milliohms Rdc wires vs. 80 milliohms Rdc of 4 switch contacts.) Contact resistance of the push-type speaker connectors used in this evaluation was less than 3 milliohms and less than 1 milliohm on the amplifier 5-way binding posts for a total Rdc path of about 7 1/2 milliohms. There was no discussion during the entire test. All wires were listened to in order from A-F preceded by a wire letter announcement. Signal source was a Dolby C tape of "Growing Up In Hollywood Town", side 1 cut 2, played on a Nakamichi BX-2. The tape was rewound to the start of the song while the wires were switched. Electronics for the reference system were: Modified Hafler DH-500 Modified NAD 3140 (preamp section used only) Denon DP-30L table with Boston Acoustics MC-1E cartridge Ohm Walsh-2 loudspeakers Undoubtably, more differences would have been noticed with a better reference system but the point of the test was to evaluate the wires and not the system. Loudspeakers were raised off of the floor and placed on top of small tables to improve the sound stage. The first pass eliminated three wires right away. The next pass the signal source was changed to "Billy Jean" from the Michael Jackson "Thriller" album, an analog recording played on the Denon table. This pass eliminated one additional wire and then came the final decision: which wire did the group like the best? Signal source was again an analog disc. The preferences were documented and the evaluation forms turned in. The results were compiled and translated from the letter codes, this is the way the first three wires rated: Best -- Moster Cable Power Line 2nd best -- Moster Cable 3rd best -- Saxton There was not enough time to properly evaluate the last three wires but preliminary indications are summarized below: 4th best -- home-made 4-wire pseudo-Litz 5th best -- Kimber Cable Worst -- #18 Zip cord Some comments from people were "I came expecting not to hear any differences at all, but I really did." "Results came out about the way I had expected." "I was surprised at all of the differences I heard." "I liked one wire best but I'm not sure why." "There were large differences between the best and worst wires, but not as many between the top two." "I couldn't hear any difference in any of the wires." In later discussions, it was apparent that people with "educated" ears were those that heard more differences. People involved only causally in high fidelity didn't hear as many (if any) differences. 90% of the people present heard some kinds of differences. But some of these people couldn't tell you in words why they preferred one wire over another. The basic conclusion was that if you couldn't hear these differences, you shouldn't invest in esoteric wire. But if you could, it was well worth the extra money. Also, two people who listened to these wires through their own receivers that had a speaker selector switch (the A, B, A+B switch) pointed out that they couldn't hear as many differences through the receiver as they did in this test. Well, there you have it. Many thanks to those that participated in the test and those of you on the net that contributed suggestions to help this test be impartial. Also, many thanks to the two local Audio Salons, Listen-Up and Boulder Sound Gallery, for contributing the wires and terminators for this evaluation. ------------------------------------------------------------------- As a side note to this subjective evaluation, remember the article in August, 1983, of "Stereo Review" about a similar speaker wire evaluation performed by a group of eleven people and authored by Larry Greenhill? On page 51 of the SR article, it concluded that > "...after fifty hours of testing, scoring, and listening to > speaker wires...16-gauge lamp chord and Monster Cable is > indistinguishable from each other with music and seem to be > superior to 24-gauge...however, it was a measurable > characteristic -- higher resistance per foot -- that made > 24-gauge sound different from other cables. If the cable > runs were only 6 instead of 30 feet, the overall cable > resistances would have been lower and our tests would > probably have found no audible differences between the > three cables." Well, the Volume 6, No. 6 issue of "Stereophile" magazine reprinted a portion of Greenhill's original manuscript sent to SR. Apparently, SR used a different set of evaluation criteria in reporting Greenhill's results than he did. An excerpt from Larry's manuscript on page 7 of "Stereophile" reads > "The striking outcome is that the panel accurately heard and > named speaker cables in 5 out of 6 comparisons...The listeners > felt pleased after this listening test battery: they had heard > real differences! After about 50 hours of testing, scoring, > comparing, and just plain listening, they were exhausted but > felt accomplished. Both Monster Cable and 24-gauge wire could > be heard reliably under double-blind conditions." Larry [unfortunately] signed off on SR's rewrite and the rest is history. I guess you really can't believe everything you read. And to coin a phrase, sometimes you just have to trust your ears. Yours for higher fidelity, Phil Rastocny AT&T-ISL ..!drufl!pmr
mat@hou5d.UUCP (M Terribile) (02/15/84)
``Wire is entirely passive and can have no other characteristics besides resistance, inductance, and capacitance'' Has anyone found out if oxide coatings have any effect. At low levels into sensitive 8-ohm speakers the voltage remains below 1 volt. Oxiide coatings on such things as razor blades were used to (nonlinearly) detect AM signals. Mark Terribile
edhall@randvax.ARPA (Ed Hall) (02/16/84)
------------------------ Sorry, I never saw your proposals as randvax was having problems with its news feed about that time. I'd be happy to make suggestions in the future. Switching speakers or speaker cables for side-by-side comparison is a tough problem without very low contact-resistance switches, unless you resort to the connection-swapping I described. Perhaps someone on the net could help find a source of high-current, low- resistance switches for such testing. The switches used by most stereo dealers are probably abysmal in this regard. -Ed Hall decvax!randvax!edhall
edhall@randvax.ARPA (Ed Hall) (02/18/84)
----------------------------
That's not a double-blind test! It's hardly even a `single blind' test.
In a double-blind test:
1) The order in which the cables were listened to would be random.
This can include listening to the same cable twice in succession.
2) Neither the person running the test nor the subject would have
any way (other than by listening) of determining which cable
was in use. (This is the double-blind part, and has proven
important in evaluative studies of all kinds, since unconcious
cuing (e.g. the `Clever Hans' effect) is a definite posibility.)
3) The results would be checked for statistical signifigance. This
is tough for a small sample.
Admittedly, #2 is a tough order. It possibly could be done by placing
short, coded wire pigtails on each end of several samples of cable, and
somehow hiding which cable a given pigtail attached to. Or, cable-
changing could be done behind some sort of screen (not exactly double-
blind, but so long as no one in the listening room knew the sequence
to be used it should suffice).
I'm not denying that there can be a difference made by speaker cables.
But if you claim a study is `double-blind', you'd best know what you
are talking about.
-Ed Hall
decvax!randvax!edhall
P.S. `Clever Hans' was a horse around the turn of the century who
seemingly could answer all sorts of yes/no and small-number
questions by stamping his hoof. It was eventually noticed
that Hans could only answer questions his owner knew the
answer to, and then only when the owner was within eyeshot.
It seems that the horse was cued by his owner's posture and
breathing.