What objectivists and subjectivists can learn from each other
- Thread starter Keith_W
- Start date
- Status
Greg,
Lew Johnson of cj though about you when we wrote these words in his TAS 223 interview :
On more than one occasion I have thought that we had reached the zenith of design and that further improvements would only be marginal in nature. In each case this period of malaise was soon followed by truly exciting breakthrough developments resulting from new materials technology and new insights into circuit design.
OK, maybe he didn't do a test himself but he must have a link to the details of such a test? I was hoping that there was some substance to his bullish statements!!
Maybe this first quote from his web site that I have just found is appropriate? "Just because you believe something doesn't make it true."
Bob,
One of the problems seems to be that the famous four parameters are not enough to define sound quality. Specially when, as Bob Carver now openly says:
We are no longer interested in facsimile reproduction; rather we find that reproduction that sounds enchanting and real is far more interesting and serves our passions far better.
What an interesting quote. I suppose that in a nutshell is the difference between objectivists and subjectivists? One is interested in a facsimile reproduction, the other is interested in enchanting sound?
If anybody is interested in finding out more info about audio power amplifier null tests, there's a description around figure 3 and figure 4 of this page. According to the author, it was first proposed by Baxandall in 1977. A diagram of Hafler's implementation can be found in the Hafler XL-600 Manual (PDF file).
Keith,
Please be careful not to misquote Bob Carver by omission -enchanting and real .
He was not addressing "Once upon a time" ...
I have rewarded CJs efforts with my hard earned cash on more than one occiaision. Todays achievement is tomorrows standard.
Yes, Andy, the details are great but real world tests with music & real-world results are what I have asked Ethan for repeatedly.
After all, he has said 50+ years ago this was decided, surely this must be a simple request to satisfy?
Commercial designers have the problem of competing with their own products in an ongoing marketing campaign to sell "better and newer". I would imagine that without "better and newer", a lot of manufacturer's would simply just go out of business because all the competition would be for the great classic pieces of the past.
A business model incumbent on selling newer products would not ever admit that the newer products might not be better, or even the "same ole same ole". The audio press is pretty much devoted to acting as undertakers for the prior models while touting the virtues of the new stuff, mere fashion thrashing for the most part.
Last edited:
It sounds like that's between you and Ethan, definitely not something I want to be in the middle of.
My purpose in posting those links was as an example of an amplifier test that uses music as its signal source, not sine waves or some other type of signal from a piece of test equipment. Amidst all the back-and-forth, some legitimate issues pop up now and then, and the issue of using music as a test signal is one of those. Since the null test addresses that issue, it's worth looking at.
If you read through that Hafler manual, you'll see a description of how to set up and run the test. He designed the amps with trimmer capacitors which could be adjusted for a minimum in the nulled residual. An actual speaker is hooked up to the amp for the test, so that loading is taken into account. The nature of the experimental data is just the sound of the residual of the null. Given that the signal source is music, there's no hard numbers to be recorded. Nowadays, I suppose the residual could be recorded to a WAV file or similar, but the results are specific to a given system, program material played, and also the volume level chosen.
I already did that. It's demonstrated clearly in my Audio Myths video. If you're not willing to watch it, this is not my fault. I can help you though: That section starts at 53:39 into the video.
By the way, I own several pieces of test equipment, and have ready access to much more at the home of my friend Mark Weiss who lives ten minutes away.
Also, at the suggestion of a forum friend here, I have decided to ignore rude and condescending posts. People who know me realize I'm never short of words or proof to back up my opinions. So please don't take my lack of response as an admission of anything. Rather, look at the tone of your own post and see if you can understand why I won't bother to reply.
--Ethan
Ok, I'll have a look - didn't want to wade through an hour of video - I'm sure you understand
Great
oh, you asked a friend
This is a well known avoidance tactic, I'm afraid, Ethan!!
So, I've answered your questions even though your tone was arrogant & smug - you might be quoting one of your own favourite quotes "Often we fault in others what we see in ourselves."
Last edited:
Although I see your point, I think that audio quality had a significant positive derivative for the last 20 years, at less for the brands I have experience. Although as you say, sometimes the audio press unjustifiably trashes the older model, most of the time the new model is really better than the old one.
There were a few cases where this did not happen - every manufacturer had punctual cases of failed models in his history - but IMHO they were exceptions, quickly corrected by market pressure.
A few times I shared your thoughts in the past, and even re-assembled old systems with great expectations (and a reasonable reduction in costs), just to discover some time after that the new models were really better.
So, the most interesting part of your video piece on Null testing was your statement @56:28
So let's just look at this statement more closely. You are saying that nulling two files proves that these two files will sound identical, right? Then in the above statement you flip this on it's head & state that two "identical sounding" files can show a difference in the nulling. Hehe, it's a douzie. I'm sure you see the logical flaw in what you say?
How can you now say that the nulling test, even though it doesn't null in the cases above, it's immaterial as the two files sound identical - by what test are you saying that they sound identical?
I will listen to more of your video as it is fun but this is a good place to start, no?
For those who need the video link http://www.youtube.com/watch?v=BYTlN6wjcvQ
So let's just look at this statement more closely. You are saying that nulling two files proves that these two files will sound identical, right? Then in the above statement you flip this on it's head & state that two "identical sounding" files can show a difference in the nulling. Hehe, it's a douzie. I'm sure you see the logical flaw in what you say?
How can you now say that the nulling test, even though it doesn't null in the cases above, it's immaterial as the two files sound identical - by what test are you saying that they sound identical
?I will listen to more of your video as it is fun but this is a good place to start, no?
For those who need the video link http://www.youtube.com/watch?v=BYTlN6wjcvQ
Last edited:
So the next interesting quote from your video @56:48
Hehe, you do it again - a logical tautology - how can you have a test that proves two files sound identical except when the test shows that they are different (but you know the files sound identical)
Why cite the test at all as when it shows a difference you ignore this difference & state that the two files sound identical anyway. Why not just skip the Null test & just say that you know the files "sound identical"? A subjective evaluation not pretending to be science or pseudo-science in this case?
Hehe, you do it again - a logical tautology - how can you have a test that proves two files sound identical except when the test shows that they are different (but you know the files sound identical
)Why cite the test at all as when it shows a difference you ignore this difference & state that the two files sound identical anyway. Why not just skip the Null test & just say that you know the files "sound identical"? A subjective evaluation not pretending to be science or pseudo-science in this case?
Last edited:
Bingo! ...All 'bout emotional strings. ...Keys (piano), notes (musical) ...
Last edited:
The bigger catch is that the consumerist objectivists as opposed to purely objectivist will throw out better measurements between two devices as these improvements are supposed to be inaudible anyway. That begs the question, inaudible by who? Not once have I ever seen any of these audibility tests that have an acceptable alpha if it is to be applied to the general human population simply because the sample sizes are two small and the selection process of test subjects is loose. That is fine for personal education and application, even commercial application, but to use these tests and apply the results to over 6 Billion people, worse yet to piggyback these results on academic tests that make no such claims even to make assertions of fact is simply arrogant. Especially if this body of questionable assertions are used to prop one's self as an authority. Yet we hear these statements about fact and truth. We must remember that while there are many accepted theories in science very few are considered laws. Statement of facts elevates these statements as laws not theories.
This type of antics I've seen work in other forums. Here, not so much huh? Here in WBF, we not only have a set of trained listeners we have bonafide Musicians, EEs, ECEs, Recording Engineers, Mastering Engineers, Certified Acousticians and a whole bunch of Scientists from other related and unrelated fields that DO know the utmost importance of proper protocol for testing. These aren't limited to cost no object types either and include a great cross section of the CE market. It's a dangerous place to be for any and all posers.
- Status
Similar threads
