Competing for Longest Post of the Week 
Hey monkeyman, I wasn't trying to dispute you as much as correct or clarify your position. We aren't as far apart as you seem to think. I just wanted to add my experience to yours.
And BTW, I am not confined to the world of mathematics. I am aware that there is a long-standing chasm between the position of audiophiles and engineers. I used to fight this battle as well - as an audiophile.
Like you, I was an audiophile for many years, using my ears (and most if not all of my paycheck) and my electrical experience in search of the holy grail of audio sound. There came a time when I wanted my money to go to other endevours. But at the time, I bought very expensive audio equipment and then did heavy modifications to the crossovers, grade of components, gain sections, and power supplies (very underappreciated, btw) to experiment with sound. My audio refinements included vacuum-tube equipment, since you seem to appreciate the warm sound they offer, to include an Audio Research preamp and a Conrad Johnson amp. There were three of us, in fact, that pursued this endeavor.
And we had a great test bench. One of us had a contract to do audio recordings of various live performances at the premier concert hall in Dallas (Morton Meyerson). We would sit, listen intently, and then be home three hours later listening to the same performance on our systems. We learned a great deal listening to these recordings, and produced systems that had incredible sound.
I think your characterization of how the systems sounds and the effects it has on the quality of audio is fine. What I objected to was your characterization of what THD is and your characterization of what clipping is. You speak with the understanding of what can effect the sound of the system, without the full understanding of the scientific effects that are influencing the waveform. Let's talk a little bit about THD and clipping.
THD is certainly audible. You're right that 1% is the threshold where most find it objectionable. But audiophiles can pick up effects below 1%. It may not be a harshness, but an impact on soundstaging, or phase linearity. What is pertainent here is that THD will go up as you go past rated power. Think of a sine wave turning into a sawtooth at high frequencies. The harmonic distortion goes up, and the music sounds harsh. This is not clipping, but an effect being driven by bandwidth, slew rate, and current starvation of the system. It is not uncommon for an amp to reach 10% THD well above rated power, and still not be clipping. This results in a harsh sounding system.
Amplifier manufacturers often pick the rated power based on their target customer. An amp being targeted for bass thumping rock-types might rate an amp at 200W, 20-20k, at 1% THD, because they know the customer is primarily looking at the power number. The SAME amp being targeted for classical-types might be rated at 70W, 20-20k, at 0.1% THD because they know the customer is primarily looking at the distortion.
As far as clipping goes, it is (as I think we both agree) a severe affect on the signal. Your characterization that it the low frequencies are stealing power from the mids and highs is perhaps a layman's description of what is happening to the sound, but it's not very accurate (or at least misleading) as far as what is happening to the signal. Low frequency energy causes clipping, because there is not enough energy (in music) in the mids and highs to cause clipping. But when the energy demands of the bass music exceeds the output capability of the amplifier, the signal is clipped. This causes the bass signal to be distorted, but as you surmise, the effects on the mids and highs are greater than the effects on the bass. The high and mid frequencies ride on top of the bass signals. When the bass signals are clipped, there is no room for the mids and highs. And the clipped signal itself generates a ton of harmonics. LOTS of distortion. But know this. When clipping occurs, if affects all frequencies.
As far as our listening friend is concerned. All I said was that THD could be responsible for the sound he was hearing. Never did I say that clipping could NOT be responsible. I made my post because he did not characterize it as VERY objectionable, like clipping would do. All I was stating was that a cheap amp can have very poor sound for highs at high power levels. There are certainly a LOT of systems that sound harsh at high volumes that are not clipping. That's all I'm saying. I agree that the problem could be in other areas as well - like speakers. I stated that in an earlier post.
And incidentally, nothing that you have linked or quoted from outside sources has disputed anything that I have stated. I'm not even sure why you quoted a few of the comments. I didn't see them as being relevant to the discussion. You focused on the THD discussion. You forgot to include an important phrase from your reference:
Measurements for calculating the THD are made at the output of a device under specified conditions.
Those specified conditions include the rated power of the amplifier. When you buy an amp at 200W, 20-20k, 1% THD, what they're saying is that if you take any sine waves between 20 Hz and 20kHz, and drive that at 200W output power, the aggregation of THD will not be greater than 1%. If the conditions for the measurment were at a different power level, the THD would be different. If you follow some of the links in that same wiki reference, you will see more detailed discussions that make this point clearly. The manufacturer does not specify the THD at 1W (which is probably the most important level, btw), 10W, or 300W - only at rated power. If you understand anything about electronics, it is virtually impossible for the quality and accuracy of a waveform to be precisely the same across the entire range of it's gain. The accuracy is worst at highest gain, which is why THD is highest at highest gain.
I think we probably agree that the THD spec doesn't mean a whole lot about whether it will be a great sounding amp, but it is still a valid benchmark. You do have to be careful. Some don't even qualify it at 20-20k. Some do 30-15k, or some even give the rated power and THD at only 1 kHz. Worthless!
We agree that clipping is a much more severe influence of distortion (or harsh sounding music) than is THD. If you've ever looked at a clipped signal waveform versus a slew-induced distortion waveform, you'd understand why. But clipping doesn't happen all the time. THD does. So it's presumptuous to state that clipping is occurring when the description of it's sound allows for other effects to be at cause.
As far as other effects, the acoustics of the vehicle, the speakers and their properties, skin effect, etc. I am aware of all of those and understand their impact. My original posts said that the harshness he hears could be something else other than the amp. We agree.
In summary, and let's see if we can agree on this. The listener has a system that is unable to provide a clear distortion-free sound at high power levels. If either one of us heard it, we might have a better chance at identifying it. The speakers with his system could probably be improved, but most people with this system have said that the root problem is the amplifier. I can't argue or support that. I also can't state for certain whether the problem is clipping or other forms of distortion in the system. Either can you. You think it's clipping. I think it could be problems other than clipping. Either way, he should start with a better amplifier and move on.
So should we.
by Monkeyman » Sat May 06, 2006 9:52 am 
.