Your audio system is terrific. How good are your ears?
by Howard Ferstler, "$ensible Sound" magazine, USA
In a series of articles in issues 71 through 74, Bob Thompson gave a detailed overview of what does and does not matter when it comes to compact-disc player performance, and as part of that short series he reviewed an error-correcting check disc for compact disc players. The name of it was the CD-Check disc (stock number DR-2002), which was and still is available from Digital Recordings, in Canada. (The company has a web site at: www.digital-recordings.com.) Bob concluded that the CD-Check disc could be very helpful in determining just what is mostly important with the playback performance of typical players, including DVD versions.
I have only recently gotten a copy of this remarkable piece of test software, myself, and as best I can tell it may be the most important audio-related check disc in history. It is called the AUDIO-CD
I have used that disc myself when reviewing DVD players for this magazine, and I periodically pull it out of my record stack to check the tracking-ability status of my seven DVD/CD and DVD/CD/LD combi players. I have also hauled it over to the homes of friends to check out their players. It is a very useful disc, even though the only thing it checks the error correcting and tracking capability of CD and DVD players.
Elsewhere, Bob had also favorably commented upon another disc, one that has the potential to have considerably more practical (and psychological) impact than the earlier one ever could. I have only recently gotten a copy of this remarkable piece of test software, myself, and as best I can tell it may be the most important audio-related check disc in history. It is called the Audio-CD (stock number 25172-02001) and it basically contains a hearing-threshold test that many of you will find much more interesting than any other audio-related test you could possibly think of, including even the most arcane ABX tests of amps and wires.
Most audiology exams are designed to evaluate an individualís ability to function in a world where conversation is important. As such, the upper frequency limit of an "audiogram" printout created by such a test is usually 8 kHz - which is rather low in frequency by high-fidelity sound-reproduction standards. The lower frequency limit is usually in the neighborhood of 250 Hz. There is a reason for these general limits. If you can hear that high up and that low with reasonable effectiveness it is likely that you will be able to hear what people say without having to interlace your assorted responses and comments with the words "What did you say?" or "Huh?" And out there in the real world the most important thing for most people is the ability to clearly hear what other people are saying.
The Audio-CD hearing test, on the other hand, is a whole different ball game. Rather than just test for conversation-related hearing acuity, it tests for high-fidelity sound reproduction hearing acuity. The disc allows a careful participant to evaluate their threshold hearing ability over an 80-dB range, and do so at 24 frequency points from 20 Hz to 20 kHz. It does this with a degree of precision that should be able to separate the golden ears from the tin ears in a very serious hurry.
OK, letís be realistic. This is a very dangerous test disc for two main reasons. First, if the person using it is not careful, erroneous results will ensue. Second, the disc has the potential to pull the psychological rug out from under just about any golden-eared enthusiast who cares to fool with it. There is no baloney with this disc if it is used properly: you either have the ability to hear well or you do not.
Note that I said "hear well" and not "listen well." There is a difference. There are people out there with superb hearing who would be indifferent to the nuances delivered by a good audio system. On the other hand, there are also individuals out there who might not have the hearing acuity of an 18-year-old, but who have worked hard over the years to recognize what is and what is not important in high-fidelity sound reproduction. And while in the best of all worlds a true audio connoisseur would combine both abilities into one ear/brain combination, I think that those with decent hearing and seriously good experience can still survive pretty well in the world of high fidelity.
Still, there is no denying the value of a disc like this as a test tool, provided the individual who obtains it and uses it properly is able to live with what it reveals. This is a dynamite piece of software and only the most intrepid audiophile, audio salesman, or product reviewer will be courageous enough to play around with it.
The disc comes with a good introductory fact sheet and set of instructions, and it is easy to set up the procedure. First, however, you do need a good pair of headphones. Junk versions will not do the trick.
The fact sheet mentions three models that should work satisfactorily: the AKG-K270, Sony MDR-7506, and Sony MDR-484. I used a pair of Sony MDR-V6 models (reviewed by me in issue 74), and I feel good about them, because a Sony representative told me that they are pretty much identical to the 7506 model. One on-line reviewer who evaluated the disc used Grado SR325 models with good results, and Bob Thompson has told me that the Sennheiser HD-280 phones are a good choice. Another, possibly best of all choice would be ear-insert headphones, such as those made by Etymotic Research. If you use a pair of headphones that do not have the ability to deliver uniform frequency response, or use a good pair improperly, the results you get with this disc may be compromised.
The company offers an audio-calibration device, the DR1-R acoustical coupler, that allows you to check the linearity of a set of phones with a Radio Shack SPL meter. You can check the web site for information on that tool. It will help the really serious enthusiast insure the best overall accuracy with the disc.
Initially, you have to test the performance of your CD player, to make sure that it handles the test procedure without any glitches. The disc has tracks on it to do that evaluation, and that is the first thing you should do before getting on with the test of your hearing.
After certifying that the player is OK, you go on to calibrate the volume-level setting of your headphones. This involves listening to a basic pair of test tones that will determine just how far down in level you should set the gain control on your amplifier. You have to be careful doing this set-up procedure, and I advise you to find a VERY quiet part of your house to do the work. Probably, this will be your regular music-listening room, with all the doors closed and the wife and kids temporarily shunted outdoors. You might be tempted to use the CD-ROM drive of your computer, but it is likely that the cooling fan in the computerís CPU will generate too much noise. It is ESSENTIAL that there be near-zero background noise while doing the tests.
Incidentally, one way to help achieve this low background noise level is to use over-the-ear headphones, which can attenuate outside noise by as much as 30 dB. The Etymotic phones mentioned above also attenuate, because they insert directly into the ear canal.
The calibration test consists of a series of test tones on track number 2 that are repeated over and over. (Track number one has an announcer giving some procedural instructions.) They are at easy-to-hear frequencies, and the idea is to set your amplifier volume control at a point where you can just barely (and I mean just barely) hear the set-up tones.
If you want, you can find someone else in your family who may have better hearing than you do to do this part of the test. Use several different people if necessary, and get that amp-gain setting as low as possible. For this part of the test it is very helpful to have a preamp, processor, or receiver that has a digital dB readout indicator. That way, it will be easy to see who obtains lowest gain setting and it will also then be easy to do the test again later on by just resetting the level to the same point. Note that if you change headphones from session to session all bets are off, since different phones will have different sensitivities. (Ditto if you decided to some of the tests again with speakers, just for kicks, although headphones are your best bet, for sure.) Once the reference gain setting is discovered, leave it be. It should not be changed during the rest of the test.
The remaining tracks, from 3 to 26, consists of pulsing tones that start out at very low levels in the left channel and get progressively louder at one-second intervals, with the level increasing by one dB each second. You listen until you can hear the tone in your left ear, and then make a note of the time-elapse number on your CD or DVD playerís digital readout. That number shows how many decibels up from the set-up reference level the signal has to increase to make it audible. You do this with each of the 24 different test frequencies between 20 Hz and 20 kHz, and then plot out the results on a graph that is supplied with the disc.
The graph contains a pre-printed curve that basically outlines just what someone with near-perfect human hearing would be able to do. You compare your results with it and see just how golden your ears happen to be. Then, you turn the headphones around and do the test all over again with the right ear.
Because there is only one graph, I suggest you go to a copy center and make quite a few duplicates to use, leaving the one that came with the disc as a master.
Now, here are some observations.
Third, as I noted before, do the test in a quiet space. When I did my series, I noticed that even the slightest bit of extraneous noise made it difficult to detect some of those threshold signals.
My wife discovered the same thing when I had her take the test. She became very aware of even very slight low-level interference noises, even with the Sony MDR-V6 phones. If your air conditioner is on that can have an effect. If you shift the phones on your head that can have an effect. If a car drives by your house during a particular sequence that can have an effect. Even your own breathing can mask some thresholds, as can any degree of tinnitus you might have, particularly at the higher test frequencies. Believe me, this test can frustrate even those who are maddeningly calm most of the time. The trick is to go at it carefully and methodically.
Fourth, make sure that your phones are working OK. Even the ones I mentioned above can have problems. In addition, the first test on the disc is at 20 Hz, and some phones (the Sony model I used, for sure) simply cannot reproduce that signal with adequate force. As a consequence, I used my phones only from 40 Hz on up and used my F1800RII subwoofer for the 20-Hz trial. (This required re-calibrating the set-up tone for my speaker systems, of course.) I also did the full test a second time with the speakers, but found that except for the bass frequencies they made task much less workable than what I got with the headphones. Stick with the phones.
Fifth, test someone in addition to yourself - particularly someone who is younger and possibly female. This can be important, because if you both do poorly at higher frequencies the problem may be with your headphones and not with your ears. Women nearly always have better hearing acuity than men at higher frequencies, so do not be afraid to drag your wife into the room and test her, too. My wife was very much not interested in doing the test, but once we got going she really got into it, and did considerably better than I did from 8 kHz on up.
Sixth, be aware of the effects of the temporary threshold shift that can occur if the participant has been exposed to any loud sounds for several hours prior to doing the test. This can temporarily lower oneís hearing sensitivity. In other words, do not test anyone who has just hopped off a motorcycle, just run a leaf blower, or just finished vacuuming the carpet.
Seventh-and this is important-be aware that there is the potential to damage your hearing if you let the levels run up too far at very high frequencies in a vain attempt to hear something that you cannot hear. If you have trouble hearing any tones at a frequency just above one that you only had slight trouble with, do not attempt to go higher still and then let the levels rise to a maximum. If you cannot hear a given high frequency, going still higher up at still higher levels is not going to you any good at all. I would imagine that there are desperate people out there who would persist in running up the gain with the amplifier volume control to the point where their dog is running out of the room. Donít be one of those people.
Also, be aware that you may be in a position to fry a tweeter or headphone element just as much as you might fry your ears if you persist in trying to hear signals that you are no longer able to detect. Do not persist in doing something that might deliver high average levels to your driver elements.
The disc costs $24.95. For more information, go to the website I listed near the beginning of this article. The site has buckets of information about hearing loss issues, as well as other interesting products. The company even has an on-line hearing test you can take, although it will not be as convenient to do as the CD version or as defining if your computerís cooling fan is loud enough to mask some of the threshold signals.
Oh, yes, I am sure that some of you are dying to know how I did with the test. Well, we have to remember that evaluating hearing acuity above 8 kHz is an often daunting proposition even for trained audiologists. There is a problem with repeatability in frequency response with any headphones, as they do not always couple to an individualís head quite the same way each time they are put on. I suppose this makes a strong case for the Etymotic phones I mentioned previously. However, all I have at the time are my Sony MDR-V6 units.
Anyway, with that disclaimer in place, I will note that I am pushing 60 years of age, and the test indicates that I have the hearing acuity of a 60-year-old man. That is, I have a loss in hearing acuity above 8 kHz that showed up during several test sessions. I can still hear out to 11 kHz, but my ability to hear a threshold signal at that frequency is limited. Fortunately, I appear to have no traumatic noise-related problems at all. Such problems often cause a substantial dip in acuity starting at about 4 kHz, and nothing of that kind showed up in my evaluation. I am flat throughout the midrange.
Of course, musical signals will often be well above threshold levels during playback, and so one can certainly hear things with decent ability at those levels. However, the ability to hear threshold signals is very important when one is listening for proper musical harmonics, harmonic-distortion artifacts, electrical attenuation in the top octave, and performance differences between items like tweeters. With serious high-fidelity product evaluating the devil is in the details, and although "years of experience" counts for something, the details often involve threshold hearing acuity.
So, what does this mean for a product tester like me? Well, I can easily match the midrange and low end of the reference line on the graph right down to 20 Hz. That means that yours truly should be able to test subwoofers as well as any whippersnapper. And I think that most of what good surround processors offer can also be determined by hearing as good as mine. And of course, I can remain a reference standard for those of you out there who are in the same age bracket that I am. Whatís more, I think I can still get a reasonably good handle on what speakers should sound like, at least when using my hearing in combination with my room-curve measuring technique.
However, I now intend to make more use of my wife than before. She is 10 years younger than I, and is in much better auditory shape. Her hearing acuity at 12 kHz matches mine at 8 kHz, and she can hear right out to 16 kHz. I am done at 11.5 kHz.
She will be hauled into the main listening room during some of my comparison sessions, and will assist me with determining if Speaker A sounds better than Speaker B, or if a given surround processor is notably effective in synthesizing a proper and realistic sense of hall space.
I believe it would be a good idea
One final note. I believe it would be a good idea if EVERY audio product tester gets a copy of this disc, uses it, and reports his findings to his readers. I can think of no better way to objectively separate the real golden eared wire, amp, and CD player testers from those who are kidding everybody - including themselves. It would also be a good idea if every high-end audio salesman also took the test and made a copy of his chart available to his customers.
Relax, sales guys, I am just kidding.