Speaker Ca­bles

Can You Hear the Dif­fer­ence?

Sound & Vision - - CONTENTS -

In the early 1980s, es­o­teric high-end au­dio as we know it to­day was just tak­ing off as an al­ter­na­tive to the mass­mar­ket equip­ment of­fered in neigh­bor­hood Tv/ap­pli­ance stores. Fu­eled by an un­der­ground au­dio press that in­cluded mag­a­zines and news­let­ters such as Sound & Vi­sion sis­ter pub­li­ca­tion Stereophile, The Ab­so­lute Sound, In­ter­na­tional Au­dio Re­view, The Au­dio Critic, and oth­ers, a cot­tage in­dus­try emerged, one pop­u­lated by small man­u­fac­tur­ers of low-vol­ume, high-priced ex­ot­ica claim­ing greater faith­ful­ness to the mu­sic than the gear re­viewed and ad­ver­tised in the pages of Stereo Re­view, High Fi­delity, Au­dio, et al. Some of these claims were founded—true ad­vances were in­deed be­ing made by start-ups run by tech­ni­cians with first-class bonafides and good ears. But the High End also at­tracted its share of half-baked prod­ucts and at least a few char­la­tans look­ing to cash in sell­ing ac­ces­sories that had lit­tle higher performance than a dime-store en­gage­ment ring.

In the midst of all this, the pre­mium ca­ble busi­ness emerged, driven in no small part by the suc­cess of the early Mon­ster Ca­ble prod­ucts that fol­lowed the com­pany’s found­ing by en­gi­neer/au­dio­phile Noel Lee in 1979. The edi­tors of our pre­cur­sor Stereo Re­view were sus­pi­cious of the ben­e­fits of such speaker ca­bles and in­ter­con­nects, which were sud­denly be­ing prof­fered by an ever-widen­ing mix of high-end spe­cial­ists, of­ten at prices far higher than Mon­ster’s. The highly ob­jec­tive mea­sure­ment-based test­ing ap­proach em­ployed by Ju­lian Hirsch and his col­leagues al­ready ran counter to the high-end com­mu­nity’s sub­jec­tive re­views, which fo­cused solely on claimed sonic dif­fer­ences that SR’S in­stru­ments couldn’t de­tect. It wasn’t long be­fore Stereo Re­view be­gan po­si­tion­ing it­self as the skep­ti­cal voice of rea­son in what its edi­tors deemed an au­dio in­dus­try gone mad.

It was no sur­prise, then, that in 1983, the mag­a­zine jumped at the op­por­tu­nity to con­duct a dou­ble-blind lis­ten­ing test, which edi­torin-chief Bill Liv­ingston and his col­leagues hoped would re­veal, sci­en­tif­i­cally, that high-end ca­bles were in­deed a hoax and pro­vided no higher performance than the ev­ery­day lamp cord in com­mon use at the time. The pitch was made by ac­com­plished re­search psy­chi­a­trist Lau­rence Green­hill, an ac­tive mem­ber of the Westch­ester (county) Au­dio­phile So­ci­ety based in the New York City sub­urbs, and a re­viewer for the au­dio jour­nal High Performance Re­view. Green­hill’s sci­ence back­ground and per­sonal in­ter­est in acous­tic per­cep­tion and sub­jec­tive lis­ten­ing had led him to write some ar­ti­cles on these sub­jects, and even­tu­ally to fas­ci­na­tion with a new au­dio test­ing de­vice called the ABX Dou­ble Blind Com­para­tor Sys­tem. This box (as ex­plained in Green­hill’s ar­ti­cle) used logic cir­cuitry to ran­domly trig­ger re­lays that al­lowed lis­ten­ers to com­pare two un­known sources in­stan­ta­neously while elim­i­nat­ing the need for (and there­fore, the bias of) a test ad­min­is­tra­tor to se­lect the sources; hence the term dou­ble-blind. The sys­tem had been de­vel­oped by mem­bers of the

South­east­ern Michi­gan Woofer and Tweeter March­ing So­ci­ety, a DIY hob­by­ist group out of the Detroit area (of which for­mer long-time Stereo Re­view con­trib­u­tor Tom Nou­saine was an ac­tive mem­ber).

By chance, Green­hill lived a few miles away from Hirsch, and their ca­sual ac­quain­tance pro­vided the en­trée. The SR ed­i­to­rial team worked up the test pro­to­col with Green­hill and some Au­dio­phile So­ci­ety col­leagues and March­ing So­ci­ety mem­ber/abx of­fi­cer David Clark. To keep things from get­ting un­ruly, they set­tled on test­ing three speaker ca­bles at 30-foot lengths, in­clud­ing thin, 24-gauge of the type found at elec­tron­ics out­lets; 16-gauge lamp cord bought at a hard­ware store; and 11.5gauge Mon­ster Ca­ble to rep­re­sent the pre­mium seg­ment. As the most vis­i­ble and suc­cess­ful pur­veyor in the cat­e­gory, Mon­ster was an ob­vi­ous choice, though, as Green­hill told me re­cently for this ar­ti­cle, it’s note­wor­thy that Mon­ster’s early speaker ca­ble was es­sen­tially also twisted-fil­a­ment cop­per like the other test sub­jects, with noth­ing par­tic­u­larly ad­vanced in the ca­ble de­sign com­pared with later prod­ucts that came from Mon­ster and other man­u­fac­tur­ers. An 11-mem­ber lis­ten­ing panel was re­cruited largely among ex­pe­ri­enced high-end au­dio­philes from the Westch­ester Au­dio­phile So­ci­ety, with Green­hill’s own young son and an­other high-schooler fill­ing out the mix. The 50 hours of tests were con­ducted in Green­hill’s per­sonal lis­ten­ing room with equip­ment bor­rowed from man­u­fac­tur­ers.

The re­sult­ing ar­ti­cle cre­ated a firestorm. As you’ll read, the panel iden­ti­fied, to a sta­tis­ti­cally sig­nif­i­cant de­gree, the 24-gauge from the other two con­tenders with pink noise as the source. More crit­i­cally, they also iden­ti­fied, again with sta­tis­ti­cal sig­nif­i­cance, the Mon­ster Ca­ble from the 16-gauge with pink noise. But the lat­ter re­sults didn’t hold when cho­ral mu­sic was used, and none of the Mon­ster ver­sus 16-gauge re­sults passed the higher thresh­old of a 75 per­cent or greater de­tec­tion rate said to be psy­choa­cous­ti­cally sig­nif­i­cant.

In­ter­est­ingly, it was not the re­sults of the test (which were mixed, af­ter all), but the ar­ti­cle’s writ­ten con­clu­sion that set off so much heated dis­cus­sion. As orig­i­nally sub­mit­ted, Green­hill’s bal­anced con­clu­sion ac­knowl­edged that the panel had in­deed heard sta­tis­ti­cally valid dif­fer­ences, at least with pink noise, and that the test should be seen not as a blan­ket re­sult for all au­dio­phile ca­bles, but sim­ply as re­search on these three, largely sim­i­lar test sub­jects that were dis­tin­guish­able mostly by level changes eas­ily as­so­ci­ated with gauge.

SR’S edi­tors, how­ever, rewrote the end­ing to cre­ate some­thing akin to a blan­ket con­dem­na­tion of the cat­e­gory and pres­sured Green­hill to ac­cept the changes, a de­ci­sion he later re­gret­ted. “They felt my con­clu­sion didn’t grab the reader as it should, but that was my point,” Green­hill re­called. “I wanted to con­clude that this test wasn’t a fair judge of all ca­bles, but re­vealed you could tell level dif­fer­ences based on the size of the ca­ble. They wanted it to be a con­dem­na­tion of [high­end] au­dio­philes.” (You can read Stereophile’s de­scrip­tion of the con­tro­versy on stereophile. com; search for “The Horse’s Mouth.”)

by Lau­rence Green­hill

The ar­ti­cle as pub­lished was picked apart and at­tacked by the high­end press, and the Mon­ster Ca­ble com­pany, not un­rea­son­ably, felt un­fairly ma­ligned as ex­pressed in a let­ter to the au­thor from Noel Lee. Green­hill, for a time, be­came a pariah in the high-end com­mu­nity, though he con­tin­ued re­view­ing and has for many years been a reg­u­lar con­trib­u­tor to Stereophile. He now lives in Cal­i­for­nia and, at 76, is still a prac­tic­ing hospi­tal physi­cian and a teacher at the Univer­sity of Cal­i­for­nia, San Fran­cisco. To­day, 35 years later, the de­bate over au­dio­phile ca­bles re­mains as ac­tive as ever.— Rob Sabin

Some au­dio­philes be­lieve that the seem­ingly in­nocu­ous wires used to con­nect stereo am­pli­fiers to loud­speak­ers ac­tu­ally have a con­sid­er­able ef­fect on a sys­tem’s over­all sound qual­ity. Un­sat­is­fied with the performance of 16-gauge heavy-duty lamp cord or “zip” cord (“gauge”

is a mea­sure of thick­ness; the lower the num­ber, the thicker the wire), these purists in­stall exotic, ex­pen­sive, and phys­i­cally im­pos­ing ca­bles in­stead. A 30-foot run of spe­cial au­dio ca­ble may cost any­where from $55 (for a pair of Mon­ster Ca­bles) to more than $300 (for Levin­son wire). Af­ter pur­chase, these thick and mas­sive wires are ter­mi­nated with spe­cial lugs or pres­sure-fit­ting ba­nana plugs ($25 per pair), coated with a con­tact cleaner (Cramolin), and in­stalled with lov­ing care.

Does it make any dif­fer­ence? Is all this trou­ble and ex­pense nec­es­sary to achieve the best pos­si­ble sound? The edi­tors of Stereo Re­view have long main­tained that for nor­mal home ca­ble runs (in the 20- to 50-foot range) to 8-ohm speak­ers, 16-gauge zip cord pro­vides op­ti­mal power trans­mis­sion be­tween a sys­tem’s am­pli­fier and speak­ers. The “of­fi­cial” view has been that heav­ier and/or more ex­pen­sive au­dio ca­bles rep­re­sent—for or­di­nary home in­stal­la­tions—elec­tronic overkill.

Nev­er­the­less, equip­ment-ori­ented au­dio hob­by­ists have con­tin­ued to tin­ker and ex­per­i­ment with their sys­tems’ con­nec­tions. Quick to re­spond to a felt need, au­dio man­u­fac­tur­ers have filled the marketplace with a va­ri­ety of ex­ot­i­cally named wires: Dis­cwasher Smog-lifters I, Au­dioSource UHD, Live Wire 301X4, Mogami Ne­glex 2477, Oracle Leonis­che, Ful­ton Golds, Mon­ster Power Line, Big Red Live Wire, and so on. Is this just an­other case of the mys­tique and glam­our of high-end au­dio in­duc­ing oth­er­wise sen­si­ble mu­sic lovers to waste their money? Or is there some real sonic ad­van­tage to us­ing au­dio­phile speaker ca­bles?

Our view is that if there are au­di­ble dif­fer­ences among speaker ca­bles, they prob­a­bly de­rive from mea­sur­able dif­fer­ences in their re­spec­tive elec­tri­cal prop­er­ties. Nel­son Pass of Thresh­old Corp., for in­stance, claims that sub­tle sonic dif­fer­ences can be cor­re­lated with dif­fer­ences in elec­tri­cal re­sis­tance, ca­pac­i­tance, and in­duc­tance, which vary ac­cord­ing to over­all ca­ble length. And these fac­tors can re­sult in mea­sur­able dif­fer­ences in fre­quency re­sponse and sig­nal level, de­pend­ing on the par­tic­u­lar speak­ers and am­pli­fiers the ca­bles con­nect. But the mea­sur­able dif­fer­ences in elec­tri­cal char­ac­ter­is­tics and performance be­tween au­dio­phile ca­bles and cheaper, 16-gauge zip cord seem too small to ex­plain the ap­par­ently huge au­di­ble dif­fer­ences that are some­times re­ported.

Just how dif­fer­ent in elec­tri­cal char­ac­ter­is­tics do ca­bles have to be be­fore there are au­di­ble ef­fects on fre­quency re­sponse and sig­nal level? In or­der to an­swer this ques­tion, among oth­ers, we em­barked on a se­ries of lab­o­ra­tory and lis­ten­ing tests. Be­cause of lo­gis­ti­cal, time, and bud­get re­stric­tions, only three ca­bles were ex­haus­tively tested: (1) 16-gauge heavy-duty lamp cord, pur­chased from a sub­ur­ban hard­ware store for 30 cents a foot; (2) 30-foot lengths of New Mon­ster Ca­ble, cost­ing $55 a pair; and (3) 24-gauge “loud­speaker ca­ble,” avail­able from many sources at about 3 cents a foot.

Although the high elec­tri­cal re­sis­tance of 24-gauge wire makes it in­ad­vis­able for runs of more than a few feet, its low price and small di­am­e­ter make it ap­peal­ing to decorators, who use it unashamedly. It seemed the ideal “worst case” ca­ble—if any ca­ble would sound dif­fer­ent, 24-gauge should be the one. The 16- and 24-gauge wires were cut to the same 30-foot lengths as the Mon­ster Ca­ble, and all the ca­bles were ter­mi­nated with Mon­ster’s X-ter­mi­na­tors for con­sis­tency in the con­nec­tions.

Lab, Tests

For the first part of our tests, Ju­lian Hirsch mea­sured the three se­lected ca­bles’ re­sis­tance and in­ter­con­duc­tor ca­pac­i­tance. The re­sults are eas­ily sum­ma­rized: in 30-foot lengths, the 16-gauge zip cord had a re­sis­tance of 0.24 ohms, the Mon­ster Ca­ble 0.09 ohms, and the 24gauge “loud­speaker ca­ble” 1.8 ohms; their in­ter­con­duc­tor ca­pac­i­tances were 420, 600, and 400 pi­co­farads, re­spec­tively.

The ca­bles were then con­nected be­tween a high-qual­ity power am­pli­fier (a Per­reaux PMF 21508) and two pairs of high-qual­ity speak­ers (Spen­dor BC-1S and KEF 105.2s). A 1,000-Hz square wave was fed to the speak­ers through the ca­bles, and the ca­bles’ ef­fects on the sig­nal were mon­i­tored by sub­tract­ing the wave­form at the speaker ter­mi­nals from the wave­form at the am­pli­fier out­put. Pho­to­graphs were made of the os­cil­lo­scope dis­plays. There were no ob­serv­able dif­fer­ences be­tween any of the ca­bles driv­ing ei­ther speaker sys­tem. All the mea­sure­ments were then set aside and not shared with the lis­ten­ing panel un­til af­ter the lis­ten­ing tests.

Open Lis­ten­ing

In the ini­tial open (non-blind) phase of the lis­ten­ing tests, the lis­ten­ers in­di­vid­u­ally eval­u­ated the sound of seven dif­fer­ent mu­sic se­lec­tions played with each of the three dif­fer­ent ca­bles while know­ing which ca­ble was in use at each mo­ment. They were asked to fill out an elab­o­rate, eight-page ques­tion­naire. In ad­di­tion to ten ques­tions about the sound of the dif­fer­ent ca­bles, the form asked about the pan­elist’s age, oc­cu­pa­tion, de­gree and kind of in­volve­ment with au­dio, and any prior at­ti­tudes to­ward the spe­cific ca­bles be­ing tested. The pro­ce­dure was de­signed to col­lect in­for­ma­tion on the pan­elists’ var­i­ous bi­ases and to sug­gest how those bi­ases might change in a group-lis­ten­ing sit­u­a­tion. For ex­am­ple, a lis­tener’s pref­er­ence might be in­flu­enced by knowl­edge of a friend’s pref­er­ence or by the rep­u­ta­tion that a par­tic­u­lar brand or type of ca­ble has ac­quired through re­views in the “un­der­ground” hi-fi mag­a­zines.

Con­trolled Lis­ten­ing

The heart of our whole project was the set of con­trolled lis­ten­ing tests. These dif­fered from the ini­tial open tests in that the lis­ten­ing pro­ce­dure was de­signed to max­i­mize psy­choa­cous­ti­cal fair­ness. That is, in or­der to keep the lis­ten­ers’ bi­ases about the weight, ap­pear­ance, cost, or brand of the ca­bles from in­flu­enc­ing their sonic pref­er­ences, the tests in this part were dou­ble blind—nei­ther the lis­ten­ers nor the test ad­min­is­tra­tor knew which ca­ble was be­ing lis­tened to. Switch­ing be­tween the dif­fer­ent ca­bles was prac­ti­cally in­stan­ta­neous (less than

50 mil­lisec­onds) in or­der to make sub­tle sonic dif­fer­ences as ap­par­ent as pos­si­ble (hu­man be­ings have no­to­ri­ously poor long-term mem­o­ries for sounds).

To make pos­si­ble such a dou­ble-blind, in­stan­ta­neous-switch­ing test, we used a lab­o­ra­tory-grade au­dio com­para­tor de­vel­oped and sold by the ABX Com­pany. It con­sists of con­trol cir­cuits and re­lays that can rapidly switch be­tween dif­fer­ent in­puts. The lis­tener com­pares sounds and de­cides whether a par­tic­u­lar source, des­ig­nated X, is son­i­cally the same or dif­fer­ent from each of two other sources, des­ig­nated A or B (hence ABX). The trick is that source X is ac­tu­ally ei­ther A or B; each con­nec­tion is made at ran­dom by the ABX com­para­tor and is not iden­ti­fied for ei­ther the lis­tener or the tester. The lis­tener writes his re­sponse on a test form. Dig­i­tal me­mory cir­cuits store the se­quence of con­nec­tions for re­trieval and anal­y­sis at the end of the test.

Level match­ing be­tween ca­bles to com­pen­sate for their dif­fer­ing elec­tri­cal-loss char­ac­ter­is­tics was not per­formed for ev­ery com­par­i­son in this part. The ba­sic premise of these tests was that mea­sur­able dif­fer­ences among ca­bles would be­come au­di­ble if they were large enough. Match­ing all mea­sur­able pa­ram­e­ters—in­clud­ing sig­nal loss—would make the hy­poth­e­sis untestable. For one com­par­i­son, how­ever, of Mon­ster Ca­ble with the 24-gauge speaker wire, we did com­pen­sate for level dif­fer­ences by switch­ing in a pre­ci­sion at­ten­u­a­tor be­tween the pream­pli­fier and the power am­pli­fier when­ever the lower-re­sis­tance Mon­ster Ca­ble was se­lected by the ABX sys­tem.

The pur­pose of this was to de­ter­mine if there were any au­di­ble dif­fer­ences be­tween the two ca­bles other than those re­sult­ing from a vol­ume-level change.

Some au­dio­philes refuse to ac­cept the va­lid­ity of this kind of con­trolled lis­ten­ing test. They ar­gue that ei­ther the test pro­ce­dure or the switch­ing sys­tems will mask im­por­tant sonic dif­fer­ences. The whole pro­ce­dure and test set­ups are de­signed to in­crease lis­ten­ers’ sen­si­tiv­ity to small dif­fer­ences and to by­pass some of the lim­i­ta­tions of the hu­man hear­ing sys­tem so that the re­sults will be ap­pli­ca­ble to all lis­ten­ers, not just those par­tic­i­pat­ing in the tests. Let’s re­view three of the main fea­tures of this phase of our tests: (1) The abil­ity to switch al­most in­stan­ta­neously from one ca­ble to an­other with a pre­ci­sion set of low-re­sis­tance re­lays makes per­ceived sonic dif­fer­ences as vivid as they can be given our limited sonic mem­o­ries. (2) The dou­ble-blind lis­ten­ing setup re­lieves the tester and the lis­ten­ers from hav­ing to com­pen­sate for pre­con­cep­tions about the dif­fer­ent ca­bles. (3) The use of nu­mer­ous lis­ten­ers mak­ing a large num­ber of com­par­isons pro­duces suf­fi­cient data for mean­ing­ful sta­tis­ti­cal anal­y­sis and gen­er­al­iza­tion.

The pro­ce­dure for the con­trolled lis­ten­ing tests can be quickly sum­ma­rized: six ca­ble com­par­isons were car­ried out us­ing the ABX com­para­tor and ei­ther pink noise or a cho­ral mu­sic se­lec­tion. Mon­ster Ca­ble was com­pared with 24-gauge wire, gain matched and un­matched; next, Mon­ster Ca­ble was com­pared with 16-gauge zip cord; then the 16-gauge and 24-gauge ca­bles were com­pared with each other. Each com­par­i­son con­sisted of a se­ries of fif­teen tests; in each test, as ex­plained pre­vi­ously, the lis­tener was asked to iden­tify which of two constant sources, A and B, was the same as the switched source X. The choice and or­der of ca­bles be­ing com­pared and pro­gram sources be­ing used were de­ter­mined by a ta­ble of ran­dom num­bers, with switch­ing ran­domly con­trolled by the ABX com­para­tor.

Re­sults were ex­am­ined only af­ter each lis­tener com­pleted the se­ries of com­par­isons, and re­sults were held con­fi­den­tial so that each panel mem­ber knew only his own score—af­ter all, the egos of these “golden ears” were on the line. More­over, in our ac­count of the re­sults we will

con­sider mainly the performance of the group as a whole; in­di­vid­ual pan­elists will be iden­ti­fied by code let­ter—to pro­tect the guilty! Fur­ther de­tails of the test setup can be found in the box la­beled “Test Meth­ods and Equip­ment.”

The lis­ten­ers were all males with an av­er­age age of about thir­ty­nine years (rang­ing from thir­teen to sixty). Although two were high­school stu­dents, most of the re­main­ing nine were mid­dle-aged pro­fes­sion­als with a se­ri­ous in­volve­ment with au­dio. Six pan­elists owned ex­pen­sive es­o­teric twin-lead speaker ca­bles, two owned in­ter­wo­ven au­dio­phile ca­bles, and one even used 14-gauge zip cord. Seven thought that con­trolled, dou­ble-blind tests like those we used were valid, but the oth­ers thought such tests missed the boat.

Half the pan­elists thought that sonic dif­fer­ences be­tween speaker ca­bles could not be ex­plained by mea­sur­able dif­fer­ences in the ca­bles’ elec­tri­cal char­ac­ter­is­tics. Be­fore the lis­ten­ing tests even be­gan, ten out of the eleven lis­ten­ers ex­pected the “bet­ter” ca­bles to show im­proved bass, “punch­i­ness,” and fre­quency re­sponse.

The left-hand por­tion of Ta­ble I sum­ma­rizes the pre-test at­ti­tudes of the lis­ten­ers. In terms of at­ti­tudes, Mon­ster Ca­ble scored sig­nif­i­cantly higher on ap­pear­ance and rep­u­ta­tion than ei­ther 16- or 24-gauge ca­ble. Although 16-gauge scored high­est on fre­quency of use and cost-ef­fec­tive­ness, the pref­er­ence was not sta­tis­ti­cally sig­nif­i­cant in these cat­e­gories. Gen­er­ally, the panel seemed to prefer Mon­ster Ca­ble from the start, sug­gest­ing that it would come out on top in the non-blind open lis­ten­ing tests, which is just what hap­pened.

The Open Tests

When the lis­ten­ers knew which ca­ble they were lis­ten­ing to, Mon­ster Ca­ble scored sig­nif­i­cantly higher, as the data in the right side of

Ta­ble I show. It was pre­ferred to 24-gauge in seven out of ten cat­e­gories. Mon­ster Ca­ble also scored sig­nif­i­cantly higher than 16-gauge in three cat­e­gories. In the pink-noise por­tions of the test, 16-gauge scored higher than 24-gauge.

Sub­jec­tive comments from the pan­elists were con­sis­tent with the nu­mer­i­cal scor­ing. They re­ported that Mon­ster Ca­ble con­veyed deeper bass, more im­pact, more am­bi­ence, and a fuller, lusher sound with greater trans­parency and that the pro­gram could be played louder than with the other ca­bles. In com­par­i­son, the sound with 16-gauge was said to be “con­gealed and ho­mog­e­nized,” re­veal­ing less “space and ex­panse” in the mu­sic. Lis­ten­ers also noted that 24-gauge pro­duced a drop in level and rolloffs in fre­quency re­sponse at both ends of the au­dio spec­trum. The “clipped and com­pressed” qual­i­ties of 24-gauge were even said to give it a “New Eng­land” sound!

It is im­por­tant to em­pha­size again that these rat­ings and comments came from lis­ten­ers who knew which ca­bles they were lis­ten­ing to. The pan­elists’ strong pre-test bi­ases in fa­vor of Mon­ster Ca­ble could have in­flu­enced their re­ac­tions dur­ing the open lis­ten­ing tests no mat­ter how con­sci­en­tiously they tried to ig­nore them. Elim­i­nat­ing the lis­ten­ers’ aware­ness of which ca­ble was be­ing au­di­tioned was, there­fore, a ma­jor con­sid­er­a­tion in our de­sign of the con­trolled lis­ten­ing tests.

The Con­trolled Tests

The most strik­ing re­sult of the con­trolled lis­ten­ing tests (see Ta­ble

II) was the se­lect­ing out of the 24-gauge wire. All the com­par­isons in which it was not matched in gain level with the other ca­bles show a high de­gree of sta­tis­ti­cal sig­nif­i­cance. With pink noise, ev­ery pan­elist was able to dis­tin­guish non-level-matched 24-gauge wire from ei­ther of the two heav­ier ca­bles in at least twelve out of fif­teen tries, or 75 per­cent, which is an ac­cepted thresh­old of sig­nif­i­cance in psy­choa­cous­ti­cal test­ing. Even when the lev­els were matched be­tween 24-gauge wire and Mon­ster Ca­ble (Com­par­i­son No. 2), four lis­ten­ers still heard a dif­fer­ence. While one lis­tener might have just been mak­ing lucky guesses, it is very un­likely that four out of eleven were do­ing so.

It is rea­son­able to as­sume that be­cause Mon­ster Ca­ble is thicker and has lower elec­tri­cal re­sis­tance than 24-gauge wire, it is a more elec­tri­cally “ac­cu­rate” sig­nal con­duc­tor. The au­di­ble dif­fer­ences be­tween them in our con­trolled lis­ten­ing tests strongly sug­gest, there­fore, that 24-gauge is too thin for op­ti­mal au­dio performance—in 30-foot lengths, at least.

When 16-gauge wire was pit­ted against Mon­ster Ca­ble us­ing pink noise as the pro­gram ma­te­rial (Com­par­i­son No. 3), three of the pan­elists cor­rectly iden­ti­fied it in twelve out of fif­teen tries. Again, it is very un­likely that this could have oc­curred by chance (the odds are less than 1 in 999). But when cho­ral mu­sic was used in­stead of pink noise (Com­par­i­son No. 5), none of the pan­elists cor­rectly dis­tin­guished 16-gauge from Mon­ster Ca­ble to a psy­choa­cous­ti­cally sig­nif­i­cant de­gree (75 per­cent of the time). This sug­gests that both 16-gauge and Mon­ster Ca­ble are fine con­duc­tors for mu­sic signals of typ­i­cal com­plex­ity, but a very sen­si­tive lis­tener un­der the best con­di­tions might find an au­di­ble dif­fer­ence—not nec­es­sar­ily a pref­er­ence— be­tween them on wide-band (pink-noise) test signals.

In sum, five out of the six sets of con­trolled lis­ten­ing com­par­isons gave re­sults that are sta­tis­ti­cally sig­nif­i­cant us­ing the bi­no­mial dis­tri­bu­tion (see caption for Ta­ble II). In only two sets of com­par­isons, how­ever—both test­ing 24-gauge wire against a heav­ier ca­ble us­ing pink noise—were the re­sults above the stricter 75 per­cent thresh­old usu­ally con­sid­ered nec­es­sary to es­tab­lish an au­di­bly sig­nif­i­cant dif­fer­ence.

Comments

While it may not be news that the thin 24-gauge speaker ca­ble sounded dif­fer­ent from the other ca­bles tested, what does this re­sult mean in terms of ca­ble mea­sure­ments and their au­di­ble ef­fects? A 1- to 2-db de­crease in sound level was mea­sured for the 24-gauge wire dur­ing the pink-noise lis­ten­ing tests, and a num­ber of panel mem­bers re­ported a loss of tre­ble and bass re­sponse with 24-gauge dur­ing open lis­ten­ing. Was the wire also chang­ing the per­ceived fre­quency bal­ance?

Ju­lian Hirsch’s mea­sure­ments of the 24-gauge wire showed it to be much higher in re­sis­tance than Mon­ster Ca­ble or 16-gauge wire. Its 1.8-ohm re­sis­tance re­sulted in a 1.76-db in­ser­tion loss with an 8-ohm re­sis­tive load, which means that fully one-third of the am­pli­fier’s out­put was be­ing dis­si­pated by the ca­ble over the 30-foot runs we used. This ac­counts for the unan­i­mously au­di­ble de­crease in level. Fre­quency re­sponse was also af­fected be­cause the 24-gauge ca­ble’s rel­a­tively high se­ries re­sis­tance in­ter­acted with the sys­tem im­ped­ance of the KEF 105.2 speak­ers used in the tests to pro­duce 1.25-db fre­quen­cyre­sponse de­pres­sions at 100, 1,000, and 4,000 Hz. This per­haps ex­plains the sub­jec­tive im­pres­sion that with the 24-gauge wire the sound was “duller.” Note that even in the dou­ble-blind, matched-level test of 24-gauge vs. Mon­ster Ca­ble (Com­par­i­son No. 2), four panel mem­bers heard dif­fer­ences. The level-match­ing pro­ce­dure cor­rected only for an over­all level mis­match, not for fre­quency-re­sponse changes caused by the 24-gauge wire.

The pan­elists’ ini­tial pref­er­ence for Mon­ster Ca­ble was only partly vin­di­cated by their abil­ity to dis­tin­guish it from 16-gauge ca­ble in the con­trolled pink-noise test (Com­par­i­son No. 3). What the pan­elists no­ticed dur­ing this com­par­i­son was ei­ther the re­sult of the mea­sured 0.16-db in­ser­tion-loss dif­fer­ence be­tween the two ca­bles or the cor­re­spond­ing 0.04-db fre­quency-re­sponse vari­a­tion when they were con­nected to the KEF 105.2 speak­ers. The for­mer is far more likely.

It seems that tests with real mu­sic signals de­crease the abil­ity of lis­ten­ers to dis­tin­guish small sonic dif­fer­ences be­tween ca­bles. Although three pan­elists heard a dif­fer­ence be­tween 16-gauge and Mon­ster Ca­ble with pink noise, the panel as a whole was un­able to hear any sig­nif­i­cant dif­fer­ence be­tween them with cho­ral mu­sic

(Com­par­i­son No. 5). Even when Mon­ster Ca­ble was com­pared with 24-gauge us­ing cho­ral mu­sic (Com­par­i­son No. 6), only three panel mem­bers had psy­choa­cous­ti­cally sig­nif­i­cant scores (twelve out of fif­teen or bet­ter), though the group’s score in this non-level-matched com­par­i­son did reach the less strict level of “sta­tis­ti­cal sig­nif­i­cance.” It seems likely that the dif­fer­ences be­tween 24-gauge and Mon­ster Ca­ble would be even harder to hear (if au­di­ble at all) if the lev­els were matched and typ­i­cal mu­sic, not pink noise, were used.

Be­cause of our short au­di­tory mem­o­ries, notic­ing very sub­tle sonic dif­fer­ences re­quires a more or less constant sig­nal as well as in­stan­ta­neous switch­ing be­tween the com­po­nents be­ing com­pared. One must won­der how real are the vast sonic dif­fer­ences re­ported in the un­der­ground au­dio press among such com­po­nents as ca­bles, pream­pli­fiers, and power am­pli­fiers, since the re­sults de­rive from un­con­trolled, non-lev­el­matched lis­ten­ing tests us­ing only wide-dy­nam­i­crange mu­si­cal ma­te­rial.

The fi­nal sig­nif­i­cant con­clu­sion one can draw from our data is that at least one gen­uine “golden ear” does ex­ist. Lis­tener J on our panel had 80 per­cent “hits” (psy­choa­cous­ti­cally sig­nif­i­cant scores) with an av­er­age score of 12.7 (out of 15). Ob­vi­ously, cer­tain lis­ten­ers—whether through tal­ent, train­ing, or ex­pe­ri­ence—can hear small dif­fer­ences be­tween com­po­nents. But the ma­jor­ity of the pan­elists, although by no means poor or un­skilled lis­ten­ers, heard only dif­fer­ences re­sult­ing from the rel­a­tively large, eas­ily mea­sured vari­a­tions in sig­nal level and fre­quency re­sponse in the 24-gauge com­par­isons. Note, how­ever, that J was the only panel mem­ber to score a hit for both mu­sic and level-matched pink-noise com­par­isons be­tween Mon­ster Ca­ble and 24-gauge (Nos. 2 and 6). In these two tests only a third of the lis­ten­ers achieved psy­choa­cous­ti­cally sig­nif­i­cant scores, show­ing that the au­di­ble dif­fer­ences in these tests were not very ob­vi­ous de­spite the over­all sta­tis­ti­cal sig­nif­i­cance of the group’s scores. Our tests were cer­tainly not ex­haus­tive, since only three dif­fer­ent ca­bles were com­pared. Yet the re­sults demon­strate that while Mon­ster Ca­ble and 16-gauge lamp cord are both au­di­bly dif­fer­ent from and prob­a­bly su­pe­rior to 24-gauge wire, 16-gauge is good enough to be in­dis­tin­guish­able from Mon­ster Ca­ble when play­ing mu­sic. An es­o­teric ca­ble would have to be sub­stan­tially bet­ter than Mon­ster Ca­ble in or­der to be demon­stra­bly su­pe­rior to 16-gauge wire. One of the lis­ten­ers on the panel ran a quick but con­trolled lis­ten­ing test of Mon­ster Ca­ble against high-ca­pac­i­tance Mogami Ca­ble (with its “damper” re­moved) and 8-gauge Levin­son HF-LOC twin-lead, both prod­ucts more costly than Mon­ster Ca­ble. He did no bet­ter in dis­tin­guish­ing Mon­ster Ca­ble from the other two than chance would al­low.

So what do our fifty hours of test­ing, scor­ing, com­par­ing, and lis­ten­ing to speaker ca­bles amount to? Only that 16-gauge lamp cord and Mon­ster Ca­ble are in­dis­tin­guish­able from each other with mu­sic and seem to be su­pe­rior to the 24-gauge wire com­monly sold or given away as “speaker ca­ble.” Re­mem­ber, how­ever, that it was a mea­sur­able char­ac­ter­is­tic—higher re­sis­tance per foot—that made 24-gauge sound dif­fer­ent from the 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. This project was un­able to val­i­date the sonic ben­e­fits claimed for exotic speaker ca­bles over com­mon 16-gauge zip cord. We can only con­clude, there­fore, that there is lit­tle ad­van­tage be­sides pride of own­er­ship in us­ing these thick, ex­pen­sive wires.

Lau­rence Green­hill is a re­search psy­chi­a­trist who is cur­rently study­ing the ef­fects of drugs on per­cep­tion. He has been an au­dio en­thu­si­ast since the late Six­ties and has writ­ten a num­ber of ar­ti­cles on per­cep­tion, au­dio, and sub­jec­tive lis­ten­ing tests.

Left to right: pen­cils, Mon­ster Ca­ble, 16-gauge zip cord, 24-gauge “speaker wire.”

Newspapers in English

Newspapers from USA

© PressReader. All rights reserved.