KeithEmo
Member of the Trade: Emotiva
- Joined
- Aug 13, 2014
- Posts
- 1,698
- Likes
- 868
I don't disagree with BigShot here... but I do disagree with the vehemence with which he defends his point of view.
I've done lots of "engineering work" - designing and testing various equipment and components - which often requires that things be measured.
In almost every case, there is a reasonably well established level of accuracy that is required for a successful result.
In some circuits, you need resistors matched to within 1%; in others a 10% part will be perfectly adequate.
However, when I measure resistors, I still use a meter that is certified to be accurate within 1/10 of 1%.
(The difference in price for the more accurate meter is inconsequential.)
The reason I do this essentially boils down to: "It never hurts to be more accurate than you need to be".
In engineering, we often discuss a more specific benefit, commonly known as "safety margin".
("If you start out with something that's a lot better than what you need, then it will still be OK, even if it drifts a little bit, or if you were a little optimistic.")
And, not surprisingly, I tend to apply a similar standard or guideline to the audio equipment I own.
Whenever possible, I don't buy equipment that is "just good enough that its flaws are inaudible to me"...
I prefer to have equipment that is "a lot better than the bare minimum that I need"...
(So, if I was quite certain that THD was only audible above 0.5%, I would still prefer to have an amplifier with a THD of 0.05% instead of 0.5%, "because it has a better safety margin".)
And, yes, I do also agree that there is "taking things too far"......
That's why I haven't spent $10k on a meter that's accurate to 0.001%.....
(However, if someone were to offer me one for only 20% more than my current meter cost, I might consider it...)
And why I'm willing to settle for audio equipment that's "probably only 10x better than what I really need".....
But I absolutely don't agree that "there is absolutely no reason to pursue better performance past what's demonstrably audible" - I'll take my 10x safety margin; even if it costs a little extra; as long as it isn't too much extra.
I've done lots of "engineering work" - designing and testing various equipment and components - which often requires that things be measured.
In almost every case, there is a reasonably well established level of accuracy that is required for a successful result.
In some circuits, you need resistors matched to within 1%; in others a 10% part will be perfectly adequate.
However, when I measure resistors, I still use a meter that is certified to be accurate within 1/10 of 1%.
(The difference in price for the more accurate meter is inconsequential.)
The reason I do this essentially boils down to: "It never hurts to be more accurate than you need to be".
In engineering, we often discuss a more specific benefit, commonly known as "safety margin".
("If you start out with something that's a lot better than what you need, then it will still be OK, even if it drifts a little bit, or if you were a little optimistic.")
And, not surprisingly, I tend to apply a similar standard or guideline to the audio equipment I own.
Whenever possible, I don't buy equipment that is "just good enough that its flaws are inaudible to me"...
I prefer to have equipment that is "a lot better than the bare minimum that I need"...
(So, if I was quite certain that THD was only audible above 0.5%, I would still prefer to have an amplifier with a THD of 0.05% instead of 0.5%, "because it has a better safety margin".)
And, yes, I do also agree that there is "taking things too far"......
That's why I haven't spent $10k on a meter that's accurate to 0.001%.....
(However, if someone were to offer me one for only 20% more than my current meter cost, I might consider it...)
And why I'm willing to settle for audio equipment that's "probably only 10x better than what I really need".....
But I absolutely don't agree that "there is absolutely no reason to pursue better performance past what's demonstrably audible" - I'll take my 10x safety margin; even if it costs a little extra; as long as it isn't too much extra.
The important thing to remember about listening tests is that they are not intended to detect all differences, just AUDIBLE ones. There can be differences that are measurable, but not audible. That may be important from a theoretical point of view, but not from a practical one. We are looking for differences that will impact our systems when we are listening to music in our living rooms. The average controlled listening test with tones detects differences that are an order of magnitude smaller than anything that would make a difference when you're listening to Beethoven on the couch. There is such a thing as good enough. Too many people chase down rabbit holes of absolutism. Most differences you read about in audiophile forums are completely irrelevant to real world music listening. You have to make an effort to learn what the numbers actually sound like to really understand. There are a couple of good AES demonstrations in my sig file if you are interested.