500+ Member: Building amps and assuring water resistance.
The Scots Guide to Analog electronics
Adding to the info contained in this thread I add the Scots Guide to Electronics in particular the Audio analog section. This site contains easy to understand analog design concepts in a semi technical easy to understand language. http://www.st-andrews.ac.uk/~www_pa/...io/Analog.html
the animated gif depicting the Class of amplifiers allows instant visulization of how this actualy works as a system http://www.bcae1.com/ampclass.htm
Well I personally don't dismiss subjective tests for the simple reason that we do not know that we are objectively measuring everything that is important to human hearing. Like say the way in which a signal is distorted. Maybe one amp distorts a signal differently than another, despite both having the same measured level of distortion. So it is possible we hear things that just don't show up on the tests we conduct.
However, I have studied a lot of Psychology and I do know the power of suggestion and the mind's ability to fool itself. My favourite experiment along these lines, which I can't find a link to despite my best efforts, was one allegedly conducte don speaker wire, really on the power of suggestion and peer pressure. They had a room with expert listeners in it and put a nice audio setup. They then introduce the three wire candidates, two high end cables, and normal braided copper wire, for reference. They would then play a peice of music on each, with assistants switching the wire in between each replay.
Well opinons disagreed about which of the two high end cables was best, but the experts were united that both sounded far better than the copper wire. The catch? The wires were never chainged, it was normal copper wire the whole time. All percieved changes were just in the mind of the listener.
Now this isn't proof that different wire doesn't matter (that wasn't tested), but it IS proof (or rather one study in a line of proof) of the power of the mind.
What I'd like to see more of is stuff you deal with in psychology all the time: proper blind experiements. However I also want to see them done properly. The problem is, you can't really ask someone reliably to tell if they like something in a quick A/B test. They need time to listen on a wide variety of material, espically if the change is supposed to be subtile.
It's something I'd like to try with something like a PPA sometime, and maybe will. Build two amps with just razor fine tolerances on everything, and have different opamps in them. Then, put them both in identicle boxes, rigged to be tamper resistant. But a label A on one box, and B on the other, randomize which box is labeled which. Give them to test subjects for a good period of time, a couple days at least. Let them go back and forth all they like. When their time is up, take back the boxes and ask which was better, or no difference and WHY they liked it better. See what kind of result you get.
If the subjects have no idea what the variable is, and there is no way for them to find out, then you have a pretty good chance of getting a real, unbiased opinon. Then you can see if there's any trends in what people like and, more importantly, why they like it. Just because the like results are 50/50 doesn't mean no difference was heard, it might just be different preferences. However if there is no corrilation between why somethign was liked, probably people are just making things up (since most will figure they are supposed to hear a difference, maybe as a control give some subjecst two idnenticle amps). However if you see real correlation between a given opamp and a given reason for liking it, you probably have a real difference, and can then maybe try and measure it.
Thing is, empirical tests of subjective data are TOUGH to properly design and execute, and there seems to be little intrest in teh audio industry to find out what they might reveal. If they show it's all in your head, all the audiophile firms are screwed. If they show there is osmething going on, all the pro firms are screwed. Either way you have something many people don't want to hear.
Personally, I hope to maybe do graduate research on the topic someday, but I'm not sure that will be feasable. It would take a fair bit of money to organise something like that, and I imagine grants for that sort of thing are hard to come by.
Dude, the argument here is exactly the one between those who accept scientific blind tests as the main practical window to The Truth(TM) and the fools that do not, so they can justify their hi-fi equipment fetishism. I use silver wire and Auricaps and Caddock MK132 resistors, but at least I don't pretend that I could actually hear a difference (nor would most of the self-proclaimed golden ears; if you come across them in real life, it's always fun to challenge them to a blind test and embarass them ).
But here's the thing: Currently, the empiricists (of which I am one, have my copy of The Logic of Scientific Discovery and everything) rely almost exclusively on scopes, FFTs and so on. Objective measurements, in other words. The reason is because we know that people lie, and even when they don't their perception isn't absolute. Measurements are, my scope never lies to me (provided it's calibrated).
Ok, great, however what we empiricists do NOT know is if we have discovered everything that is important to human hearing. We know frequency response is, so we measure that. We know noise is, so we measure that, etc. But, what if there is a property we DON'T measure, beacuse we aren't aware of or believe it is unimportant, that is important to perception? Well that would mean, despite our claims that two things sound the same because they are identicle on the scope, we are wrong.
One possible example would be the nature of distorion. All I ever see measured is the amount, never the kind. I don't know enough about audio distortion to say for sure, but I would infer that distortion does not distort all harmonics of a note evenly. Thus the distortion itself will have a harmonic shape, just as the note does. The difference between a trumpet at A440 and a claranet at A440 isn't the pitch, the fundimental is 440Hz in both cases. Nor is the difference in the frequency of the harmonics, they are all integer multiples of 440Hz. The difference is in the relitive amplitudes of the harmonics, which are loud, which are soft. THAT is what makes sound to a human, and how we identify speech and so on.
So, maybe the shape of a device's distortion (if indeed distortion has a shape as I postulate) is something that effects our perception. We hear two things with the same THD, but one has a shape that sounds better than the other.
I'm not saying this is for sure that it exists or if it does that it matters, I am simply using it as an example of something that we do not attempt to measure and figure out what is good and bad.
Well, rather than just chase my tail around, the research I think would be interesting to do would be to take two things that are widely regarded to have good, but different sound that ought not, like the op amps, and do a proper, multi-level test on them. If the study shows that there probably IS a difference in sound, I'd then want to do more research to try and find out what causes that difference.
Or maybe (more likely I believe) I'll find that there IS no difference. In that case, we have some real, valid, empirical evidence to argue that. Claiming "it's all in your head" when you don't have evidence of that is kind of going out on a limb. It's probably all in our heads, but we need to do more testing to validate that hypothesis.
I've been mulling this one over and I might see if it's something that some funding and facilities could be gotten for. Need to hash out the design better first though. It's an interesting topic that is pretty much absent from the psychological journals.
Ok, great, however what we empiricists do NOT know is if we have discovered everything that is important to human hearing.
Of course measurements do not reflect everything that may affect perception. After all, distortion figures are but summary statistics, and don't tell you the detailed profile of the distortion, which also matters to perception besides the magnitude of the distortion. Again, this is not at issue. A properly conducted blind test covers all known and unknown matters of perception. Denying this is denying rationality and science as a way to understand the world -- it turns into a religion.
I've said this several times at the various audio forums: by far the greatest difference between live sound (and I listen to live classical and jazz concerts extensively) and electronically reproduced one is in creating a proper soundstage, which is to some extent a geometric problem, and includes the recording arrangement's ability to record the proper directional distribution of sound, and the reproduction at the other end of such directionality in a way that parasitics are eliminated. By parasitics I mean things such as room reflections and more importantly, sound intended for one ear reaching the other ear, which is not a problem with headphones but it is a problem for speakers. Generally, stereo separation is crudely increased in the recording in order to accomodate for this crosstalk, but the results are not life-like (and worst for headphones). The proper way to deal with this is signal processing and speaker arrangements which can eliminate most of the crosstalk (there are numerous examples of related research that can be found on the web). This way binaural recordings can be listened to properly with speakers. But those have limitations too, such as different HRTFs (head related transfer functions) on the recording head or dummy head, and the listeners head/ears, which makes especially up/down and front/back positioning less effective than if the same shape head/ears were used (again, perhaps signal processing can accomodate for this if the listener's HRTF is known; imagine everyone getting it measured and feeding it into the signal processor as a custom profile ); etc. etc.
Personally, I think Doug Self represents a radical extreme in the objectivist camp, and he's putting his evidence through a contradictory-fact-removing-filter worthy of Fox News. I don't have the listening experience to judge any of his conclusions, but I can certainly critique his logic.
Overall, his book seems to have one underlying theme: "I think all amplifiers sound the same, but some people don't, so here's how you can swindle more customers by advertising a really low THD measurement that you and I both know doesn't make a bit of difference." So should it surprise us if his arguements don't hold much water?
The most obvious rebuttal to objectivism is just that amplifiers are bulit for listening, and all that matters in the end is the subjective perception of the listener. Perhaps in Futurama they could have a THD-analyzing robot that loves nothing more than to listen to beautiful undistorted sine waves, but that's not me.
Which brings us to another clear mistake: his near-total reliance on THD measurements as the final word on amplifier quality. In Part 1 of this article:
you can see a great example of where THD measurements specatcularly fail to predict the behavior of the system. What Self really needs to remember is that distortion residuals will fully characterize a memoryless nonlinear system, and the various linear ODE tools (state-space, Laplace, Fourier) will fully characterize an LTI system. But to analyze a real amplifier which is neither memoryless nor linear, none of those tools are theoretically appropriate. In fact, there IS no general theory that allows one to analytically predict the behavior of such a system. Even proving stability requires you to "guess" a Lyapunov function.
Another fatal flaw is his lack of respect for human hearing. It's true, we can only percieve a 10% step change in amplitude or a 0.2% step change in frequency, but those measurements have little to do with how we actually use our hearing. We can nevertheless perform a myriad of feats that not even the best microphones and DSP algorithms can match...like picking a voice we recognize out of a room full of people. Also relevant is the fact (which I unfortuneately know only from anecdotes) that a difference in volume that's not directly perceptible will nonetheless make a louder amplifier sound better.
So to conclude, I don't disparage mathematical analysis as a way to better sound quality...it's just essential to have the right tools. Doug Self does not have those tools, and furthermore he seems to have given up looking.
The most obvious rebuttal to objectivism is just that amplifiers are bulit for listening, and all that matters in the end is the subjective perception of the listener.
Yes, and it is what a blind test measures (there are even used in psychology, after all). This is one of the things that bothers me most, why do publicized blind tests show a spectacular failure of distinguishing between amplifiers? Likewise for the subtraction tests -- if you can't hear a residue, then how could there be audible distortion? There simply is no rational argument against blind testing. Some argue that a nil result doesn't prove there is no difference, which is true, but statistically over a number of trials the likelihood that there is a difference can become vanishingly small.
Yes, I've seen this before, but not heard much discussion about it.
Quote:
We can nevertheless perform a myriad of feats that not even the best microphones and DSP algorithms can match...like picking a voice we recognize out of a room full of people.
Part of that trick has to do with directionality. It's much much harder to do with a mono signal. The rest of the process is indeed DSP, although implemented in wetware instead of silicon.
Quote:
Also relevant is the fact (which I unfortuneately know only from anecdotes) that a difference in volume that's not directly perceptible will nonetheless make a louder amplifier sound better.
Yes, it is very important in blind tests that levels are properly matched. But if just a slight difference in volume can mask other issues, doesn't that make them insignificant?
One thing that Self left out is the issue of imaging. The ear can distinguish clicks as separate even if they are at a rate above 20 kHz.
The only rational objection I came up with regarding this article was his discussion of phase. Apparently he is unaware that phase is an important component of directional perception; though, on the other hand, with most speaker setups it's already messed up enough so that the amplifier is the least offender in that respect.
Also, on the Crossover Design page ( http://www.silcom.com/~aludwig/Sysde...ove_Design.htm ), it is mentioned that ears can still hear clicks occuring at a rate faster than 20 kHz as distinct (in the Transient Response section of the page, though the server is down at the time of writing, but you can use the Google cache). Which leaves me wondering whether amplifier blind tests would have more positive results in a setup where phase distortion from the speakers and room reflections are eliminated (that still leaves crosstalk to mess up the imaging, however, so using headphones with a binaural recording would be better, even though headphones do not present a realistic load to a power amp).
Ok, maybe I should get more specific. To start with, here's one of the claims of subjectivism Self claims is false:
Degradation effects exist in amplifiers that are unknown to engineering science, and are not revealed by the usual measurements.
In fact, this statement is obviously true. It's essentially an axiom of the scientific method. It's hubris to assume that we've uncovered EVERY distortion that nature put into the amplifier, so of course, there will be some we haven't discovered yet. The burden of proof is therefore on him to show that these unmeasurable distortions don't change the subjective perception of quality. Instead, he assumes that they simply don't exist.
So, if Doug Self has assumed from the beginning that there is no such thing as a distortion mechanism that doesn't show up on his Audio Precision System 2 or whatever, why should we treat it as significant that he doesn't find any?
For instance, here's what he says in his rebuttal to the claim that amplifiers can't be completely characterized with just sine waves:
You must remember that an amplifier has no perspective on the signal arriving at its input, but literally takes it as it comes.
That's quite false. Since amplifiers are not memoryless, their internal state DOES depend on the past history of the music, and since they're nonlinear, other parameters like the gain and distortion spectrum do depend on the previous signal. Sine waves are an easy test of an amplifier because they're periodic, and thus admit a steady-state solution in which all transient behavior has died out.
Slew rate limiting is an obvious example: if you crank in a step that causes slew rate limiting to kick in, other signals will be silenced entirely until the ramp is done, since the input stage is saturated.
This interaction between linear and nonlinear effects is also the crux of what peufeu is talking about, and his research (along with Lavardin's) clearly shows a potential mechanism for audible distortion that doesn't show up on a THD spectrum. Listening tests (maybe not blind ones) show a very obvious difference from this new effect.
Now, I've never heard a "blameless" amplifier so it may be that its distortion is too low to begin with for any of this to matter, but it does not change the fact that Self's reasoning is wrong.
Sycraft's comments here are right on the money, in my opinion. He's basically saying the exact same thing I am:
Quote:
Ok, great, however what we empiricists do NOT know is if we have discovered everything that is important to human hearing. We know frequency response is, so we measure that. We know noise is, so we measure that, etc. But, what if there is a property we DON'T measure, beacuse we aren't aware of or believe it is unimportant, that is important to perception? Well that would mean, despite our claims that two things sound the same because they are identicle on the scope, we are wrong.
Also, I think he's got a great point on blind tests:
Quote:
Build two amps with just razor fine tolerances on everything, and have different opamps in them. Then, put them both in identicle boxes, rigged to be tamper resistant. But a label A on one box, and B on the other, randomize which box is labeled which. Give them to test subjects for a good period of time, a couple days at least. Let them go back and forth all they like. When their time is up, take back the boxes and ask which was better, or no difference and WHY they liked it better.
So, when you get some blind tests that were done that way, then it will really show something. It's really common for some little flaw to creep up after a few days of listening that just makes it fatiguing and unenjoyable, making most types of blind tests regrettably impractical.
The Scots Guide to Analog electronics
).
Linear Mode
