Well said. My handle here, perfknee, is short for "Performance knee". That is the place on the price vs performance graph where the slope increases and it starts becoming very costly to get additional performance. That's the place where I like to buy: More expensive than mass market junk, but way less expensive than "high end".
As my contribution to this debate, I point you to the writeup on a guy who will pay $10k for anyone who can completely reliably tell the difference between ANY two amplifiers by just listening. He has still not found anyone to give his money to.
Richard Clark Amplifier Challenge FAQ
This doesn't mean that there aren't audible amplifer differences, but it does mean that they are not easy to hear.
From my perspective, and to simplify the argument, I would say that audio, as with most things, tends to have a type of relationship where the quality of the reproduction increases logarithmically, but the cost increases exponentially.
The above is in regard to actual equipment here, i.e. source, amp, speakers. I start to get a little dubious when we get into cables, but since there is electrical interaction between the cable and the rest of the electronics, I suppose that if a system has high enough resolution, then you might hear changes. That doesn't to me justify spending multiple thousands on cables in my opinion.
In regards to things like cable stands, shakti stones, tice clocks, etc., I'm just not going to go there.
As usual, YMMV.
Peace,
Dave Gerecke
The above is in regard to actual equipment here, i.e. source, amp, speakers. I start to get a little dubious when we get into cables, but since there is electrical interaction between the cable and the rest of the electronics, I suppose that if a system has high enough resolution, then you might hear changes. That doesn't to me justify spending multiple thousands on cables in my opinion.
In regards to things like cable stands, shakti stones, tice clocks, etc., I'm just not going to go there.
As usual, YMMV.
Peace,
Dave Gerecke
As to THD, my understanding is that 1% is normally near the "performance knee". The harmonic distortion is usually inaudible below 1%, but sometimes audible above 1%. In that sense it is a useful measurement; I don't buy any equipment with a THD above about 1%, and try to limit it to 0.1% where cost effective. This way I buy components where I don't worry about hearing harmonic distortion, without needing to pore over THD graphs (and stress over obtaining them). That is a useful shortcut.
It's trivial today to hook a component up to a spectrum analyzer and get THD vs freq graphs of each of the different harmonics. But that doesn't mean that they are free or comparable. Decent spectrum analyzers cost a few thousand dollars (including the cost of a PC for PC-based analyzers), and good ones can easily exceed $10k. They all have minor quirks that make the graphs obtained by different spectrum analyzers difficult to compare. The single THD number still has a place in cutting through that cost and comparison difficulty.
Something else to realize, and it took me a long time to understand this, is that THD does not capture all distortions. Delay a signal, invert it, change its amplitude, and it will still show up as the same THD graphs and number. Lots of important factors don't show up in THD: drifts in performance over time, thermal stability of an amplifer, input/output impedance, and perhaps most importantly: clipping behavior.
It's trivial today to hook a component up to a spectrum analyzer and get THD vs freq graphs of each of the different harmonics. But that doesn't mean that they are free or comparable. Decent spectrum analyzers cost a few thousand dollars (including the cost of a PC for PC-based analyzers), and good ones can easily exceed $10k. They all have minor quirks that make the graphs obtained by different spectrum analyzers difficult to compare. The single THD number still has a place in cutting through that cost and comparison difficulty.
Something else to realize, and it took me a long time to understand this, is that THD does not capture all distortions. Delay a signal, invert it, change its amplitude, and it will still show up as the same THD graphs and number. Lots of important factors don't show up in THD: drifts in performance over time, thermal stability of an amplifer, input/output impedance, and perhaps most importantly: clipping behavior.
logarithmic = exponential
Did you mean the quality of reproduction increases linearly (versus what, effort to improve?)
Ok, let's go back to where this started.
You said "Hearing is evidence. Believers in the objectivist religion are among those who deny it."
Hearing is evidence of what exactly?
Generally, they don't say they "hear" anything. They usually just relate their subjective experience.
Because it's not entirely analogous. By that I mean that subjective perceptions of differences can come about by ways other than expectation bias. It can be things as simple as listening differently or a difference in focus of attention, etc.
But at the end of the day, I'm ok with "placebo effect" when used informally.
I wouldn't as the perceived difference had nothing to do with one's ears, but only one's mind.
Yes, all precepts are only in our minds.
Yes.
I disagree as we can determine "actual audible difference" outside of the precepts, i.e. by using blind tests and statistical analysis.
I'm not advocating that the terms "hearing" or "heard" be done away with in all contexts. I'm not intending to be quite that anal.
But if in response to say, "No one has demonstrated actual audible differences between cables" someone says "I hear differences and if you don't, you're either deaf, have a crappy system, or both," well that's a whole other matter as they're clearly making an objective claim in that instance.
Yes, it is.
Only that I wish some people wouldn't try and pass it off as more than that. Or if they do, they accept the burden of proof and substantiate their claims.
No claim that doesn't necessarily mean that others will necessarily hear what you hear. There are plenty of things which we know to be audible which not everyone can hear for themselves.
I just searched your blog and you never even use the words "hear" or "heard" in the context we're talking about here. You're simply describing your subjective experience. Which is just fine with me. I only have problems when people try and pass their subjective experiences off as more than what it is and effectively go on to make objective claims.
It's rather a moot point. As I said above, in your blog, you never use the words "hear" or "heard" in the context we're discussing here. And at the time I wrote what I did in my previous reply, I hadn't read your blog and wasn't speaking of anything you said there specifically but was only speaking in general terms.
I didn't say you'd made an objective claim, only that when an objective claim is made, it comes with a burden of proof. Do you still disagree with this?
se
Yes, perceiving is always subjective, but we can use tools such as blind testing and statistical analysis to try and determine whether a given perception has a foundation in physical reality, i.e. an actual audible difference.
se
Hello Perfknee, here's a couple of graphs that should explain what I mean. Basically, what I am saying is that your monetary investment (possibly time also) increases exponentially (whether DIY or purchased) and the performance grows logarithmically, reference the two graphs.
The first graph is an exponential growth, the second logarithmic
Peace,
Dave
The first graph is an exponential growth, the second logarithmic
Peace,
Dave
Attachments
Dunno. Depends what's being heard. Let's move it out of the modality of hearing and into seeing. Seeing 'yellow' is not evidence of anything but the perception of yellow. Would you say that hearing is evidence of anything?
So for you, placebo effect must be caused by expectation bias? If there's none, there can't be any 'placebo effect' ?
So then ears are able to hear things independently of mind? What exactly are you saying in order to make this distinction between 'ears' and 'mind'?
Read again, you're suffering from expectation bias.
I did not say 'precepts'.Only if everyone agrees on the statistical significance level and then only after the experiment. And so it should be called 'experimentally determined, statistically ascertained lack of difference at the 95% confidence level'.
So then only on diyaudio? What's the extent of the anality going to be?
Yes, there must be other possibilities for why you don't hear a difference, including that they're subject to the placebo effect. But that's a red herring in this context, I've suggested nothing of the kind here yet here we are, having a debate.
I have no issue with what you wish for. I don't even mind that you decided to take the bait of what I was saying and interpret it rather differently to what my intended meaning was.
The phrase 'pass it off' has a meaning of some deception being employed. I'm taking the phrase as its used in the legal sense - did you mean it in this sense - that there's dishonesty involved?
But that's exactly what you're doing when you deny that others hear what they hear. Your basis for denying what they hear without experiment is that you've objectively redefined the word 'hear' and then apply your own humpty dumpty meaning to what those guys are saying. Then you wish they wouldn't use the word 'hear' in the sense that you've redefined it in. Well I've got news for you - you're projecting
You were taking issues with specific things I'd said. Now it seems you've withdrawn some of those things. What's left?
I don't see how its possible to make objective claims in the realm of hearing. I also don't see them being made when someone says 'I hear a difference' or when they just describe what they hear. So it seems you're tilting at windmills.
They sure are pretty though
Thanks for your counsel but I think I'll take it with a pinch of salt, after some of the other ideas you've brought forth, I'm not convinced of your expertise.
Same here..
There is consistency in your assumptions , the others that share the same assumptions are in the same boat. Obvious some here don't even listen to music, while others don't have a system capable of evaluating anything , so i understand and empathize with your position ...
Shhhh.....
Thanks Dave for trying to clarify. I don't mean to beat this to death, but I still don't get what the second graph is about... what are the horizontal and vertical axes of the two graphs?
The first graph is exactly the "Performance Knee" that I was talking about, with performance (vertical axis) versus price (horizontal axis). The "Knee" where I like to buy is right in the horizontal middle of that graph.
THD is a next to useless specification.
From your criteria, then you must not own any speakers?
dave
Actually I make my speakers, don't buy them
My understanding is that speakers typically have THD from about 10% for good ones to 25% for "average", and higher yet for poor quality. I don't know of any speakers that get anywhere near 1%.
Speakers having such a high level of distortion is a big part why I'm confident that 0.1% THD in active components is not audible (dwarfed by the speaker distortion).
AVE...
It can be simplified a little bit, if we recognize, which claims about equipment and sound perception are true, and which are just pure bulls**t. When someone claims that changing the USB cable that connects DAC to PC makes the setup sound better, he is either lying or is an idiot. I read the blogs about CES2011 and everything that was said there about any kind of cables was loads of bulls**t. People who are buying that stuff don't deserve to have money, therefore they should give it to the charity and then kill themselves - average IQ of human population will be increased noticeably...Well we never said it was easy!
Good PC, good software and proper interface and you can do FFT of the output and you can also perform automatically FFT for all frequencies using sweep function...
Probably yes, but again: can you hear them? Or, to be precise, can you perceive them with your mind without looking at the brands and price tags? If not, then there is no point to buy the equipment that costs more than new car...
Buy cheap - in most cases it's crap. But there are sometime exceptions to that rule. But overpricing is more common practice...
I just believe that above certain price level all the differences are created by audiophile's mind, but audiophile won't admit that he/she was fooled. The best we can do is to make sure that the others won't be fooled by marketing, reviews and by other audiophiles...
Besides the cost of simple repair of old car is relatively low when compared to the cost of repairing the new car. Also some cars nowadays are made in this particular way, that ensures there is huge market for spare parts and authorized service providers. Have you ever wondered, why new equipment nowadays malfunctions just after the service warranty ends? My washing machine malfunctioned one week before the end of warranty - insides looked like someone had thrown a grenade inside. I don't leave grenades in my dirty socks. Other example: my father's Audi 100 from 1982 after repairing some minor problems was used as a cab (my father was a taxi driver). We did 200 thousand kilometers before we had any serious problems with it. Spare parts we bought for less than $300, we did the repairs in garage. In the same time my uncle bought new Audi 80. In three years he spent about $3000 in authorized service provider, for example for replacing the engine that was broken apart by cold (it happens if you buy diesel fuel smuggled by Ukrainians)...
Willow won't conduct electricity...
We achieved this goal long time ago with the creation of CD. We can record and playback any recording without loosing any qualities of the sound, that would be noticeable by our hearing. The last goal is to recreate the perception of sound position, and this was done first with stereo systems, later with Surround systems, and now will be done again with new kinds of software to create more realistic binaural recordings. Check out virtual barbershop for example...
But when we look at hifi and Hi-End equipment and accessories, it's all consisted of cool-looking outsides, pseudo-scientific bulls**t, marketing tricks, miss-information and fake reviews...
If we focus on what really we are discussing ,it is just advertisement :
just pick up any kind of item from any magazine ,and you'll end up to say they're lying . But that's just the job of the advertiser ,the creative , who often doesn't really matter what's the promotion for ,but just giving a nice cut to the picture and to the text . At a design school where I went ,there were few hours in the course where we studied the advertisement pages in magazines .
just pick up any kind of item from any magazine ,and you'll end up to say they're lying . But that's just the job of the advertiser ,the creative , who often doesn't really matter what's the promotion for ,but just giving a nice cut to the picture and to the text . At a design school where I went ,there were few hours in the course where we studied the advertisement pages in magazines .
But almost all the harmonic distortion products in a speaker are 2nd or 3rd order. Electronics almost always have higher orders. It is those higher orders that you want to be fearful of.
It is also possible for an amp1 (with highish THD)/speaker to have combined THD lower than amp2 (with low THD)/speaker.
THD is far too simple to have any meaning.
To stretch an analogy, "i'm only going to buy a high performace car if the dealer is on the south side of the street"
dave
Why? If the cable does not pass the analog USB signal down the cable with true enuff fidelity, one can get timing errors at the other end.
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.