ark@alice.UucP (Andrew Koenig) (10/30/85)
In thinking about my recent exchange with Dick Pierce (teddy!rdp), it appears to me that what I am trying to say and what he is hearing are two quite different things. What I have been trying to do is to distinguish between the form and content of an argument. To invent an example from a different domain, let's consider the following statements: "XYZ cassette decks are lousy." "XYZ cassette decks are lousy because they use non-standard recording equalization. Therefore their tapes are incompatible with other machines." "I don't like XYZ cassette decks." "I don't like XYZ cassette decks because their recordings sound distorted to me compared with the original." If you make one of the four statements above, your claim might be true or it might not. I might choose to believe it or I might not. These two states (truth and belief) do not always go hand in hand. To see why, let's look at each statement in turn. The first statement is an unsupported assertion about a (mostly) objective situation. For the sake of this discussion, let's drop the "mostly" and assume that either XYZ decks are lousy or they aren't. Now, your statement has absolutely no supporting evidence. Why should I believe or disbelieve it? Perhaps you know a lot about these matters so I should accept your authority. But I probably don't know nearly as much about the extent of your expertise as you do. So don't be surprised if I take such an unsupported assertion as really meaning: "Person X thinks XYZ decks are lousy, no reason given." Statement 2 is much better: it presents supporting information. Unfortunately, that supporting information is itself unsupported. If you told me where you heard about the equalization problem, I might be able to check it out for myself if I cared. As it is, though, if I want to verify your claim I must redo all the research myself. Unless I'm passionately interested in this issue, I'm unlikely to bother. I'll put much more trust in your statement if you include information like: "I know the EQ is different because I made this test using that test equipment and my results were..." Statement 3, if made, is almost certainly true. That is, if you say you don't like XYZ cassette decks, you probably don't. But that doesn't give me much to go on in deciding whether or not I like them. De gustibus, etc... Now if I know your tastes well, that will give me a basis for prediction. But if I don't, all bets are off. Moreover, the discussion is effectively ended unless you can answer my obvious question: "Why don't you like them?" Statement 4 is again much better. Now, at least, we have something to discuss. You say you think it sounds distorted, I can go listen to it and say "it doesn't sound that way to me," you can say "Did you try music with triangles and cymbals?" and so on. The point is that if you believe something but do not tell me why, you should not expect me to automatically believe it too. If you know something but do not tell me how you come to know it, do not expect me to have the same confidence in that knowledge as you do.
rdp@teddy.UUCP (11/01/85)
In article <4498@alice.UUCP> ark@alice.UucP (Andrew Koenig) writes: >In thinking about my recent exchange with Dick Pierce (teddy!rdp), >it appears to me that what I am trying to say and what he is hearing >are two quite different things. > >What I have been trying to do is to distinguish between the form >and content of an argument. To invent an example from a different >domain, let's consider the following statements: > > "XYZ cassette decks are lousy." > > "XYZ cassette decks are lousy because they use non-standard > recording equalization. Therefore their tapes are incompatible > with other machines." > > "I don't like XYZ cassette decks." > > "I don't like XYZ cassette decks because their recordings > sound distorted to me compared with the original." > >If you make one of the four statements above, your claim might be true >or it might not. I might choose to believe it or I might not. These >two states (truth and belief) do not always go hand in hand. To see why, >let's look at each statement in turn. > OK, Andy, this seems to have gotten completely out of hand, so, if you desire, then I will respond in more detail. In my original posting I made some assertions about the Bose 901's. How 'bout I address the ones I remember, and provide more backround. I described an experiment where a recording of a person is compared to that same person standing in the same room, poking his head in various resonant containers. The experiment I and many others in the audio business have performed many times demonstrates that the Bose 901 suffers from gross, narrow band frequency response anomolies. Other than suggesting that people try the experiment for themselves, how shall I demonstrate the results? I stated (I believe) that the assertion that Bose made that having multiple drivers tends to cancel out individual anomilies is incorrect. This can be demonstrated both by some theory and some actual measurement. One anomoly which is most apparent is in the region around 2000 Hz. This is due to the primary rim resonance of the driver. That rim resonace (it's frequency) is determined by two things: The size of the driver (about 3 1/2" diameter) and the composition of the driver (which determines the propogaction velocity of sound in the cone). The composition of the drivers did not substantially change from unit to unit. To do this would require to many production runs for a single model of driver. When Bose was buying drivers from the likes of Becker, they were buying single production runs. How do I know this? The sales manager of Becker told me so. One of the incoming inspectors at Bose also told me so. Taking a Bose 901 apart and looking at the drivers, noting that production numbers on all of them were the same told me so. Measuring the bloody damned drivers told me so. So given that the driver composition was relatively constant AND the driver diameter was relatively constant might lead on to the conclusion that problems such as the driver rim resonances are also constant from driver to driver. I, and many others, have measured Bose drivers and many other drivers, and the information handed us by the equipment is a little hard to ignore. If you don't accept the evidence, fine. I stated that Bose has, in the past, used grimly cheap parts. Well, when I was in the business, I had access to the manufacturers OEM price lists. People like Becker were supllying the drivers to Bose. The drivers were exactly the same as those found in 4 inch "wide band" automotive speakers. The exact cost, I do not remember, but it was apparent from these price lists that the component costs for the 901 (including equalizer) was significantly less than that of speakers which retailed for less than 2/3 the price (Infinity and the like being examples). I woudl invite anyone to survey the entire history of production 901's and examine what I and others have examined. If their conslusion are substantially different than my own, I stand corrected. But the evidence is before us, in every pair. I alluded to gross anmolies in frequency response. How do I prove the existance of these anomolies over the net? Is not stating that, using a General Radio Acoustic Measuring Set consisting of a 1521 Graphic Level Recorder, combined with a 1309 Audio Frequency Generator and a 1564 1/3 and 1/10 octave Audio Analyzer, and a (number not remembered) narrow band Audio Spectrum Analyzer, and a Crown IMA intermodulation analyzer, and a Sound TEchnology 1750 distortion analyzer, and.. and..., I found errors in frequency response as great as 30 db under conditions where other speakers had errors with 6 or 8 db, or that harmonic distortion at what might be considered reasonable (85-90db) where sometime orders of magnitude greater than that of other speakers under similar operating conditions (25% versus .5%). Did I relate the story of a high level of customer dissatisfaction at Discount Records (a local record chain here) because gross record defects (grove damage, mistracking, etc.) that caused customers to bring records back for exchange simply were inaudible over the stores Bose 901s? How do I prove this? I don't think I can, the store (I believe) is long out of business. But if I can dig up David Satz, Dave Ranada, Steve Owades, Peter Engel, Bob Morrison, Steve Bauman, and a host of others, they will confirm this and many of the other points. How do I prove that magazines such as Stereo Review HAD the advertising policies, that are now gone because of a great deal of pressure from a variety of sources (I like to think I was partially responsible for this editorial policy change, but that might be wishful thinking)? One quick comment here, I think the current policy of not printing bad reviews is not much better. The buying public has a right to all information, bad or good, and what right has a publisher from restricting the flow of such information? The policy existed, it was known to damn near every manufacturer. It was not illegal, or even strictly dishonest (although I disagree). Frankly, who gives a shit? This discussion has gone too far. The speaker, in my opinion, does not deserve my time or energies. This discussion does not even belong here. If, Andrew, you want to continue, I suggest email. I frankly don't think the 901's, nor even the mode of hating 901's are worth the time. I don't think criticisms of anecdotal stories (anecdotal does not mean fictional here) are worth the time. If there is a significant number of people who want the discussion to continue, fine, but I suspect there is not. The discussion has been not uninteresting, for sure, but I grow weary of it. (it just occured to me that this discussion is not substantially different than that engaged by Bose proponents back in my hi-fi days, the actual discussion was not about the pros and cons of the product, but rathjer on the semantics of the arguments and criticisms). Dick Pierce