Enough already with Cheever's thesis. It's beyond me why anybody would choose to waste 5 years of their life arguing about one single paper, that they don't even consider to be well written.
Think of all the time that COULD have been spent reading other papers. GOOD papers. That would also make for more interesting discussion here. Can we move on to something else now, pleeeease?
For one reason because JC said that it should be standard reading for anyone who designs amplifiers and since JC is widely regardeded and acknowledged as an excellent designer himself, his opinion carries weight. Yet the paper made no sense to me.
But far more importantly, because its author received an MEE from an accredited school in the US. This is evidence of the corruption of the accreditation process especially in a field of study which is important to me. It means that intellectual standards have deteriorated to where there is no longer any value or credibility to them. JC's more recent statement about respecting effort is typical of the degeneration of social standards and values that started during the 1960s which has it that if you've made an honest effort, failure is acceptable, that the effort itself is sufficient. That philosophy depricates the value of real achievement and success. It's a view that accepts mediocre as sufficient which is one I reject.
BTW, I have read many other papers in the last five years and some of them were outstanding, even breakthroughs. Unfortunately none of them were in the field of audio. I've also read a lot of inane stuff that got published too.
It is the 2nd harmonic of 1kHz basis frequency and its amplitude is one part in ten thousand, not in ten million!! I doubt your qualification in this discussion.
Then it should be childs play for you to find the flaws in my conclusions rather than pick around the edges such as whether it's one part in 10,000 or one part in 10,000,000 and which camp Threshold is in. How about a frontal assault on my assertion that the harmonic distribution of THD components doesn't matter beans once THD is below a given threshold of inaudibility no matter what the distribution?
Yeah, what's a factor of a thousand, anyway? Meaningless. 
If you can't do basic electrical calculations, then it's probably a good idea to learn how before participating in engineering discussions. There are plenty of internet forums which encourage people who aren't familiar with basic electronics to speculate, criticize, and theorize, and perhaps you might be more comfortable on one of them?
If you can't do basic electrical calculations, then it's probably a good idea to learn how before participating in engineering discussions. There are plenty of internet forums which encourage people who aren't familiar with basic electronics to speculate, criticize, and theorize, and perhaps you might be more comfortable on one of them?
I can understand the frustration of those who are threatened by the possibility that the Cheever thesis fails. It's not just because of its direct consequences but there are far wider implications. What that means is if there is a threshold beyond which improvement is not just one of diminishing returns but of no returns, then you will reach a point where you can't go on reinventing the same thing over and over again becuase there is no point to it. IMO insofar as audio amplifiers, preamplifiers, and CD players are concerned, we reached that point a long time ago. Then the only innovations are to cut cost, improve reliability, improve efficiency. Therefore if the $30 CD player is within the same inaudibly different from ideal performance band as the $3000 player, there is no point, no market for the more expensive player. Same for amplifiers and preamplifiers. The wider implication is that if you don't get the results you want from your best effort, the best that can be done with the paradigm it has been developed under and assumed correct, then you have to go back and re-examine your most basic assumptions. You have to find out where and why they are wrong. You have to rethink the entire problem from scratch. And you have to come up with radically different answers. This is real work. This entails real research and real thinking. This could result in real breakthroughs. But it is much harder, much more time consuming, and more expensive than just tinkering around with the same old ideas over and over again. So the validity of invalidity of Cheever does not stand alone by itself but represents an intellectual battleground with far wider implications. Failing to successfully defend Cheever means you fail to defend the notion that there are an infinite number of incremental improvements you can make that will matter, that will be of real value. In short that the issue is the value of the present concept of high end audio itself.
An intelligent argument can be made against the Cheever thesis, and it already has, many times, and by people who grasp engineering basics. It's all old news. Do you have anything new to add?
If you drop me a PM with your email address, I can suggest some basic books on physics and electronics so you can contribute something useful. I'd even be willing to donate a few to you if you promise to read them and do the problems.
If you drop me a PM with your email address, I can suggest some basic books on physics and electronics so you can contribute something useful. I'd even be willing to donate a few to you if you promise to read them and do the problems.
THD is now measured as amplitude ratio, and the SW calculate it from voltage. -80dB = 1 : 10 000.
http://www.sengpielaudio.com/calculator-thd.htm
http://www.sengpielaudio.com/calculator-thd.htm
Last edited:
Thanks guys it was "really" starting out as a bad day. But you "really" gave me a belly laugh. "Really!"
I would be interested in your book list!
So as I poorly recall Sabin and logarithms should be added to the undiscovered country!
Some folks here have a mind like a steel trap... rusted shut.
Folks are lazy, for a pure resistance it does not matter 10X the V is 100X the power. The conveinience of using dBV or dBu in audio is that we rarely are doing match terminated systems and this removes the load dependence. With optics it is easy to get confused because "amps per watt" crosses the two domains.
Yes, I forgot for a moment that there is dbv and dbp. As P=Vsquared/Z one is -10 log and the other is -20 log. Therefore db for power would be the square root and it would be one in 10 exp-4 instead of 10exp-8. It's been awhile.
BTW Sy I'll bet my shelf of books on physics and electrical engineering has more books on it than yours does. Since I don't design amplifiers or smash atoms for a living I haven't read them in awhile. About 40+ years.
- Status
- Not open for further replies.