Bob, please go to the AES in NY as an emissary for all of us. Please report back as to what is new and useful in audio design today. Charles and I have other limitations that preclude us from attending. My problem is monetary. I just got a $750 bill from my doctor for a treadmill test. I thought that Medicare paid for this! Oh well, that's old age for you in the USA.
Decades ago, I used to go to every AES, including Europe. You should too, if you can. Today I only attend AES in SF or LA (sometimes). I do try to get the preprints, so you be sure to get them too, and say hello to Dr. Hawksford for me, and you might think to thank him for his input that you have utilized in your circuitry. If you have dinner together, you might make a friend. I did, years ago. That is where I first met Richard Heyser, Michael Gerzon, Peter Craven, and so many others. I had Dr. Hawksford over to my office in Berkeley, decades ago. Perhaps you two should get together and compare notes. I'm serious, in fact, maybe you can be a perfect candidate for president of the AES in future. Why not? I was asked once, and your credentials should look better than mine, and you are more photogenic too! Go for it.
Decades ago, I used to go to every AES, including Europe. You should too, if you can. Today I only attend AES in SF or LA (sometimes). I do try to get the preprints, so you be sure to get them too, and say hello to Dr. Hawksford for me, and you might think to thank him for his input that you have utilized in your circuitry. If you have dinner together, you might make a friend. I did, years ago. That is where I first met Richard Heyser, Michael Gerzon, Peter Craven, and so many others. I had Dr. Hawksford over to my office in Berkeley, decades ago. Perhaps you two should get together and compare notes. I'm serious, in fact, maybe you can be a perfect candidate for president of the AES in future. Why not? I was asked once, and your credentials should look better than mine, and you are more photogenic too! Go for it.
AR-2, I am glad that someone has tried to make a comparison. I have been hoping and trying for 40 years to make IC's sound as good as discrete. I concur with Charles that global negative feedback is the main problem, but have you ever tried to design your own IC op amp? I have, and all that I can say is the individual components are trash, and it takes a genius to make anything of quality out of them. Scott, you do a great job!
AR-2m you are in SF. Why not come over to Berkeley sometime and compare notes? At least give me a call.
AR-2m you are in SF. Why not come over to Berkeley sometime and compare notes? At least give me a call.
I am perfectly serious. I am speaking in my normal voice and intent to Bob Cordell. I wish him well at AES, but it is easy for him to go, because he lives near there and he tends to go every year. I am not being sarcastic. Why should he not meet up with the other designers at the AES? I do, and did for the last 39 years.
AR2 said:
Excellent! I was getting to think I was out to lunch; actually building and listening (and I know that's not true, but there is such resistance and few step up to the plate and says this is what I've done and this is what it sounds like). I have stated my initial impressions, but am trying to keep an open mind.
I've avoided op-amps for many years but thought it was time... Partially based on the list of "good chips" these days and National's bold claim. And, while I'm still in the middle of checking it out, I can't think of any other way to do it but build it and listen. A minicule amount of board space, a few resistors, a couple bypass caps IF your layout has issues, but instant line amp, and almost as easy for a phono amp.
Mike.
mightydub said:
First of all please note that while there are obvious sonic differences that occur during the break-in period, I doubt that these will translate into measurable differences. So if you are one of those guys that thinks that anybody that hears a difference between things that measure the same is deluded, then you may as well ignore the rest of this post.
On the other hand if you are genuinely interested, I am happy to speculate. And speculate is all I can do. As far as I know, nobody "knows" the answers to the questions you have asked. (Just like nobody "knows" how gravity works.)
Careful listeners have noted a break-in phenomenon that occurs with many pieces of audio equipment. After years of experimentation, I have found that these effects are largest in dielectric materials. Hence the components that change the most during the break-in period are those that use a lot of dielectric materials:
- Capacitors
- Wires
- PCB's
As you know doubt know, the dielectric constant of an insulating solid is greater than one. This is because under the influence of an external electric field, the intrinsic charges within the solid material are redistributed. This can happen in several ways. For example, the electron "cloud" around an individual molecule can temporarily be displaced. Or the molecules themselves can re-align (rotate or translate).
The first phenomenon occurs at a exceedingly short time scale that probably has little effect upon audio. Furthermore, when the external field is relaxed, the charge around each molecule is quickly restored.
But the second phenomenon has a much longer time constant. And what's more, the original molecular structure of a solid is not necessarily restored when the external field is removed. (This is the basis for the operation of an electret.) On a macro level, this phenomenon is usually referred to as dielectric absorption (or as Tektronix called it when they discovered it in PCB materials, "hook").
I would assume that during the break-in period that there is some semi-permanent change to the molecular structure of the dielectric that leads to the audible changes observed by many. But this is only speculation, and I know of no measurements that support this.
mightydub said:
In my experience (and many others) it depends on the material. In the case of high-performance PCB materials, I find that it takes at least a week of constant signal for the bulk of the "break-in" to occur, and that smaller changes continue for up to six to eight weeks.
NB -- To anyone reading this thread that thinks people who can hear these types of "break-in" effects are deluded, please don't bother to reply. You think I am deluded, I am equally confident you are deaf, and there is no point whatsoever to hash over these types of arguments for the zillionth time.
SY said:
Sy,
I hate to be technical, but the thread topic is not an interview that Charles gave. It was a letter to the editor that Charles sent in response to some positive things that John Atkinson said about the National parts. Charles initiated it, he was not being asked by someone for his opinion in an interview.
When Charles calls other designers mere packagers because they choose to use ICs in a blanket condemnation, he is crossing the line of defending his no-NFB and discrete design views, and deserves to have his feet held to the fire.
When a competent IC company like National takes the needs of the audiophile seriously and tries to do something about it, people should be cheering them on, not deriding them like Charles has done. It is probably true that the National parts cannot reach the highest rung of high-end audio performance, but they are a valuable addition to our arsenal.
It is unfortunate that some people feel threatened by the emergence of new technology.
Bob
john curl said:
I'll be happy to report back what I find. I also have not gone as much in recent years, and only go when it is on the East Coast. I do miss the good old days. I don't even bother to buy the preprints anymore. Things seem not quite the same.
We are all indebted to Dr. Hawksford for coming up with HEC, and I have never failed to give him credit for it. I merely applied it for the first time in a practical way to a MOSFET power amplifier, and showed how very good his technique was. Bruce Candy, on the other hand...
.Cheers,
Bob
Bob Cordell said:
Today. Everyone knows that today. Back then the engineering community and Crown thought it a fine advance until the product line switched to the 'truly excellent' 741. The arguments were as fierce and the details lost in age (mine), but I recall one AES or IEEE demo in which a dozen 741 in series drove a pair of headphones and members challenged to hear when the chain was switched out of circuit. Not picking sides (glass guy), just nits.
Bob Cordell said:
I hate to be technical, but you are wrong.
I wrote John a private e-mail expressing my disappointment at his promulgation of the myth that "low measured THD+N on a test bench with steady-state signals and a resistive load automatically guarantees good sound quality with real music in a real stereo system". I've lived through that one before, 35 years ago, and it's no more true today than it was then. 35 years ago the "objectivists" thought that the distortion of a Crown IC-150 was low enough to guarantee perfect sound quality. Now the bar has been raised and JA implied that 0.00003% is low enough to guarantee perfect sound quality.
John thought that my perspective deserved a wider audience, so he asked if he could print my letter in the magazine. If you don't agree, that's fine, I don't care.
Write your own letter to Stereophile. If John thinks it's interesting, I'm sure he'll print it too. He's a pretty open minded kind of guy.
Bob Cordell said:
Bob, you have designed quite a few original circuits in your time. As you know, it isn't always easy and requires some degree of effort on your part. Compare that to the Sony SCD-1 (and SCD-777) SACD players, where the entire analog output stage was copied, op-amp for op-amp, resistor value for resistor value, from a Burr-Brown app note from the early '90s. Would you call that effort an act of engineering or an act of packaging?
Bob Cordell said:
And just what, exactly, would that new "technology" be? Another 10 dB of feedback compared to their previous "ultra-low distortion" op-amp?
john curl said:
LOL -- "ICK-150"!
In a way, you didn't miss much. On the other hand, listening to that preamp was the turning point for me. Read the story here:
http://www.audioasylum.com/cgi/m.pl?forum=critics&n=34397
And speaking of other websites, I just ran across something interesting. A lot of DIY'ers have reported good sonic results from substituting the new National parts for other op-amps in their gear. But here are a half-dozen people that had bad sonic results from those new parts:
http://www.audioasylum.com/audio/tweaks/messages/14/149480.html
Who knows why they performed well in some circuits but not others? But clearly the bottom line is that you have to listen to know how it sounds...
Charles Hansen said:
So I guess removing those op-amps would be an act of unpackaging?
I left only I/V op-amp in SCD-777 and take the output right from there. And I do use LM4562 and don't find anything wrong with it. I tried briefly LM6172 which initaialy seemed to be better but in a long run I preferred the other IC.
Now, there are some well regarded circuits that feature op-amps only, a Walker phonostage comes to mind, pictured below. It disappears from Audiogon as fast as Vendetta and is priced usually higher.
Attachments
Charles Hansen said:
Charles,
You and I are sitting at a table. The wall is behind you. I have a plate of spaghetti. I'm going to toss a single noodle at the wall behind you. Since you're closer to the wall than I am, turn around and check to see if it sticks.
Much is made of short signal paths. Now, let's set aside discussion about how many parts--passive or active--are in the signal path, parts quality, and whether the traces on the PCB are oxygen free or not. Here's the noodle: Is it possible that some of the observations about short signal path boil down to the fact that a smaller circuit is exposed to less PCB?
Just to simplify things, let's skip the obvious things like stray capacitance, quality of the copper used to coat the board, and so forth.
Incidentally, it's clear that some of these folks are not even in the same restaurant, much less at the same table. There's nothing wrong with obsessing with specifications...until one person tries to ram it down another's throat. Sorta like the old line about "you're free to swing your fist as wildly as you like--until it contacts my nose." It's interesting to note that they contribute nothing but vitriol to the thread.
Grey
P.S.: Do you know of a source for smallish quantities of raw better-than-FR4 PCBs? I do my own boards and wouldn't mind investigating other materials.
P.P.S.: I don't think that anyone, least of all Nelson, would claim that the BOSOZ is a SOTA contender. It's a simple project piece, meant to entertain and instruct. It's quite possible that a top of the line opamp might sound better than a BOSOZ. To attempt to shoot down a discrete vs. chip comparison based on the BOSOZ is pretty lame. Compare a chip to Nelson's production pieces, not his projects, fer cryin' out loud.
- Status
- Not open for further replies.