Hi,
Yes....Shall I spill the beans or do you just feel like changing clothes, Jan?
Not an expert...but:
I know Otalla's work through my acquaintance with the "L'Audiophile" crew, Jean Hiraga et all.
Having done some work with them in the past when I assisted on numerous occasions when these theories were discussed.
While I won't cast votes on this, I do apply the same thinking to most of my own work.
In fact I've done so for over twenty years now and I'm absolutely convinced this is a major key to a successfull amplifier design.
IME correct presentation of spatial information embedded in stereo recordings can never be reproduced correctly when this theory is disregarded by the abuse of global feedback loops.
Just see this as a little interlude on my part if you like; I feel John and Matti Otalla are absolutely on the right track here.
No disrespect for all those willing to copntinue the debate, I'll be quiet in my little corner...
Cheers,😉
Yes....Shall I spill the beans or do you just feel like changing clothes, Jan?
Not an expert...but:
I know Otalla's work through my acquaintance with the "L'Audiophile" crew, Jean Hiraga et all.
Having done some work with them in the past when I assisted on numerous occasions when these theories were discussed.
While I won't cast votes on this, I do apply the same thinking to most of my own work.
In fact I've done so for over twenty years now and I'm absolutely convinced this is a major key to a successfull amplifier design.
IME correct presentation of spatial information embedded in stereo recordings can never be reproduced correctly when this theory is disregarded by the abuse of global feedback loops.
Just see this as a little interlude on my part if you like; I feel John and Matti Otalla are absolutely on the right track here.
No disrespect for all those willing to copntinue the debate, I'll be quiet in my little corner...
Cheers,😉
John, remember this thread, when we beat this issue of phase modulation to death? http://www.diyaudio.com/forums/showthread.php?postid=419336#post419336
We looked at the issue of taking two otherwise identical amplifiers with the same gain-bandwidth product but one with reduced low-frequency open-loop gain due to collector resistors at its VAS output. It's not difficult to see that for a fixed output voltage, the differential input voltage of the amplifier with the larger low-frequency open-loop gain will be less than or equal to that of the one with the smaller low-frequency open-loop gain at all frequencies. So the increased low-frequency open-loop gain reduces the instantaneous phase modulation, rather than increasing it. If you disagree with that, I'd like to see what your specific objection is, and the reasoning for that. In other words, I'm asking "What's your reasoning?" and not "What's your belief?".
Otala's work was important because he woke up the audio world and made them consider transient performance. However, not everything he said was correct.
We looked at the issue of taking two otherwise identical amplifiers with the same gain-bandwidth product but one with reduced low-frequency open-loop gain due to collector resistors at its VAS output. It's not difficult to see that for a fixed output voltage, the differential input voltage of the amplifier with the larger low-frequency open-loop gain will be less than or equal to that of the one with the smaller low-frequency open-loop gain at all frequencies. So the increased low-frequency open-loop gain reduces the instantaneous phase modulation, rather than increasing it. If you disagree with that, I'd like to see what your specific objection is, and the reasoning for that. In other words, I'm asking "What's your reasoning?" and not "What's your belief?".
Otala's work was important because he woke up the audio world and made them consider transient performance. However, not everything he said was correct.
Andy C, you might have beat it to death, but it doesn't change the math of Dr. Otala or Barrie Gilbert. Just today, on this website, the technical papers of several audio designers has been maligned. This includes: Matti Otala, Walt Jung, Marshal Leach, Barrie Gilbert, and me. Who do you think first developed most of the circuits that you make here on DIY audio? Do you have any idea?
I know - the mighty JC.
Where is my sigar? 😉
JC, please, which Barrie Gilbert paper did you mention?
Dimitri, the Barrie Gilbert article is called 'Are Op Amps Really Linear?' Analog Avenue www.chipcenter.com/analog/ Nov 28, 2000
By the way, I only developed some of the topologies that you folks use, but I am familiar with all of them. What is amazing to me is that my opinion is so severely questioned on this subject matter, when I have so much experience with it.
By the way, I only developed some of the topologies that you folks use, but I am familiar with all of them. What is amazing to me is that my opinion is so severely questioned on this subject matter, when I have so much experience with it.
Mikeks, Jan, Andy_C,
Why attack John? For all his comments, good and bad, he has created more products of undoubted sonic quality than most of us and if he chooses not to indulge multiple lines of heavy math to prove his point, who can blame him? He says - and I KNOW this is true - that good audio design needs only simple math, a few basic equations, and a lot of iteration and experience. Why can you not accept that as the truth, and stop indulging pointless intergalactic urinary competitions (IUCs)?
I have the highest regard for Mikek and Andy_C's math ability. Stupendous stuff. But more than erudite to me, and from my standpoint explains very little, least of all about why certain types of distortion are tolerated by the human organism. What about building the damn things and listening to them, and forming a few conclusions based on sonics alone?
Jan, your comment about this being a stand and deliver club is absurd to the point of being childish, and not a little pompous. People are here because they want calm, anecdotal dialogue. They have no right to demand rigorous, almost legalistic proof, no right to demand formal educative process, no right to demand adversarial response, and no right to expect closely written scientific text. This is a forum, for crissakes, and people are here of their own volition. Jeopardise that volition by adversarial behaviour and they will leave. Who needs the brain damage? What mature individual enjoys flame wars? And why is the fighting so intense when in truth almost no-one knows anyone else, except by reputation, and no one has any loyalties, except to their own stubbornly held views?
Where has the civilisation gone amongst all you clever people?
This is my POV, I know it will incense, but it needs to be said and I will not continue this further. I am spending less and less time here these days because of these IUCs, and I'm fed up with them. It's OK if someone is stupid, or ignorant, or doesn't want to know, or holds to erroneous beliefs. It's NOT OK if someone demands proof at all costs, of a rigorous, maths-based type, and shows disrespect to some of the greatest names in audio, first Walt Jung, and now John Curl. Something similar happened recently to Graham Maynard, who I think is just brilliantly clever, and it shows no sign of abating.
Wise up, guys. Try to be more circumspect.
Deeply felt,
Hugh
Why attack John? For all his comments, good and bad, he has created more products of undoubted sonic quality than most of us and if he chooses not to indulge multiple lines of heavy math to prove his point, who can blame him? He says - and I KNOW this is true - that good audio design needs only simple math, a few basic equations, and a lot of iteration and experience. Why can you not accept that as the truth, and stop indulging pointless intergalactic urinary competitions (IUCs)?
I have the highest regard for Mikek and Andy_C's math ability. Stupendous stuff. But more than erudite to me, and from my standpoint explains very little, least of all about why certain types of distortion are tolerated by the human organism. What about building the damn things and listening to them, and forming a few conclusions based on sonics alone?
Jan, your comment about this being a stand and deliver club is absurd to the point of being childish, and not a little pompous. People are here because they want calm, anecdotal dialogue. They have no right to demand rigorous, almost legalistic proof, no right to demand formal educative process, no right to demand adversarial response, and no right to expect closely written scientific text. This is a forum, for crissakes, and people are here of their own volition. Jeopardise that volition by adversarial behaviour and they will leave. Who needs the brain damage? What mature individual enjoys flame wars? And why is the fighting so intense when in truth almost no-one knows anyone else, except by reputation, and no one has any loyalties, except to their own stubbornly held views?
Where has the civilisation gone amongst all you clever people?
This is my POV, I know it will incense, but it needs to be said and I will not continue this further. I am spending less and less time here these days because of these IUCs, and I'm fed up with them. It's OK if someone is stupid, or ignorant, or doesn't want to know, or holds to erroneous beliefs. It's NOT OK if someone demands proof at all costs, of a rigorous, maths-based type, and shows disrespect to some of the greatest names in audio, first Walt Jung, and now John Curl. Something similar happened recently to Graham Maynard, who I think is just brilliantly clever, and it shows no sign of abating.
Wise up, guys. Try to be more circumspect.
Deeply felt,
Hugh
Easy Hugh, it's just a website. 
The trouble with all these references to papers and articles is that it takes effort and time to go through them. Better if the person quoting such things is able to provide a precis for those with limited time.
I'd propose this as a general axiom for constructive arguments.
The trouble with all these references to papers and articles is that it takes effort and time to go through them. Better if the person quoting such things is able to provide a precis for those with limited time.
John, thanks for answering in a way to something i posted.
As you thought you may be adressed, the question was also adressed to you.
For the above: I tried. I went to evening school, went to university, and after 2 years now, i know: You don´t learn anything there besides talking the same lies, blabla, and nonsense the professors do. This may be different at other places. I learned more form this board, especially from Nelson Pass, than from any of my university classes.
I was running my own business for 10 years, before that university excursion, and will do again. This educational system does not help in any way - i need to read and learn by my own. Classes there are only waste of time.
I´m sure in case you would get less involved in those "foodfights", you would find the patience to explain the mechanism mentioned by you, also for uneducated persons like me.
Thanks Andy C, I may have learned somthing useful today. Now I know why violins and brass don't sound on my system quite way they do in live performance. The sound as the sticky horse hairs slide over the cat gut and the ring as the sound is difracted by the brass bell just aren't presented quite properly. It's due to higher order nonlinearities.
That thread where you "beat the issue of phase modulation to death" had some interesting stuff on it including Cheever's thesis.
http://w3.mit.edu/cheever/www/cheever_thesis.pdf
I don't give a fig about his presentation on feedback problems, the part that was interesting to me was the section on the "aural envelope." pp 42 - 50
He then develops fairly persuasive case that nonlinear harmonics falling outside this envelope will under certain conditons be heard as distorting. This part is interesting and seems to stand up independently of any particular amplifier topology.
I'm not yet competent to judge his arguments about feedback, amplifier topologies and nonlinear products but the psychoacoustic part is definitely worth investigating by an amplifier designer, I should think.
There is also a recent article in JAES V52 No 7/8 Aug 2004 by Tan, Moore et al that also develops distortion metric based on hearing. It's quite elaborate, more rigorously done than Cheevert's work but I don't think its as useful for me, a consumer, to use to think about this problem as his.
That thread where you "beat the issue of phase modulation to death" had some interesting stuff on it including Cheever's thesis.
http://w3.mit.edu/cheever/www/cheever_thesis.pdf
I don't give a fig about his presentation on feedback problems, the part that was interesting to me was the section on the "aural envelope." pp 42 - 50
p42
He then develops fairly persuasive case that nonlinear harmonics falling outside this envelope will under certain conditons be heard as distorting. This part is interesting and seems to stand up independently of any particular amplifier topology.
I'm not yet competent to judge his arguments about feedback, amplifier topologies and nonlinear products but the psychoacoustic part is definitely worth investigating by an amplifier designer, I should think.
There is also a recent article in JAES V52 No 7/8 Aug 2004 by Tan, Moore et al that also develops distortion metric based on hearing. It's quite elaborate, more rigorously done than Cheevert's work but I don't think its as useful for me, a consumer, to use to think about this problem as his.
Hi all,
Just a brief note as I'm at work. It looks like John and others have interpreted some of my statements as some kind of personal attack. I just wanted to set the record straight and say that this was never my intention. I had only intended to address issues of a technical nature, but in the process may have come off as somewhat abrasive. If I have offended anyone, I am sorry.
P.S. The slang "Beat XYZ to death" has meant in my experience in engineering "covered XYZ in excruciating detail", not "exposed XYZ as being in error". Maybe that's what caused it?
Just a brief note as I'm at work. It looks like John and others have interpreted some of my statements as some kind of personal attack. I just wanted to set the record straight and say that this was never my intention. I had only intended to address issues of a technical nature, but in the process may have come off as somewhat abrasive. If I have offended anyone, I am sorry.
P.S. The slang "Beat XYZ to death" has meant in my experience in engineering "covered XYZ in excruciating detail", not "exposed XYZ as being in error". Maybe that's what caused it?
I have to agree with Hugh......
Although I don't find much to argue with Andy_C about.
He is right though: you don't need to write pages of LDEs to build a good sounding product. You may need to if it is important to understand why it does or doesn't, as some of our more erudite members feel.
Some of you may be put off by John's attitude. But who can blame him? He is here at his discretion, passing on his experience as he sees fit.
And some of his experience has been getting ****ed on by the AES, and a handful of their minions, for no apparent reason other than personality clashes. He is right: the AES turned into a "Get Marshall Leach Society" in the 80s, and many of us left.
Now this place is turning into the same crap. A handful of erudite guys who only respect tons of rigourous math, and ignore practical experience.
As P. J. O'Rourke would say:
"Age and guile beats youth and a bad hair cut."
Some of you guys need a hair cut...............
Jocko
Although I don't find much to argue with Andy_C about.
He is right though: you don't need to write pages of LDEs to build a good sounding product. You may need to if it is important to understand why it does or doesn't, as some of our more erudite members feel.
Some of you may be put off by John's attitude. But who can blame him? He is here at his discretion, passing on his experience as he sees fit.
And some of his experience has been getting ****ed on by the AES, and a handful of their minions, for no apparent reason other than personality clashes. He is right: the AES turned into a "Get Marshall Leach Society" in the 80s, and many of us left.
Now this place is turning into the same crap. A handful of erudite guys who only respect tons of rigourous math, and ignore practical experience.
As P. J. O'Rourke would say:
"Age and guile beats youth and a bad hair cut."
Some of you guys need a hair cut...............
Jocko
sorry, I don´t want to fight with anyone here, and i don´t expect math. I consider math as a modell for something you cant tell in plain words. In case you can tell it in normal language instead of formulas you may have understood it much better.
I only want to learn soemthing about the ideas poted here by John and others.
Re: I have to agree with Hugh......
Ouch. I wasn't aware of that. It pains me to think that someone felt my post was a slam of him. If it weren't for him, I wouldn't be an engineer today. (I read his article in Audio years ago, sent him a letter, and he sent back tons of info about the curriculum). I have never been an AES member, so I'm not familiar with its politics.
Jocko Homo said:
Ouch. I wasn't aware of that. It pains me to think that someone felt my post was a slam of him. If it weren't for him, I wouldn't be an engineer today. (I read his article in Audio years ago, sent him a letter, and he sent back tons of info about the curriculum). I have never been an AES member, so I'm not familiar with its politics.
Till, I am sorry for your rather 'sophomoric' position with regards to university. Perhaps you should change colleges. However, I agree with you in part. What I learned in college was the math that I would never learn on my own. While it has been over 40 years for me since I took 'Differential Equations' or 'Advanced Calculus', two required courses in my junior year in physics, the basic concepts remain, especially when I read research material.
For example, almost 30 years ago, I 'INSTANTLY' understood what Herr Manger was doing with his exotic loudspeaker, once he gave me a reprint of the 1925 article of 'Rice and Kellog' on the design of the loudspeaker. The differential equations with dot notation form were readable on inspection.
Without my 'required' math background, I could still be scratching my head over what is really unique about his speaker design.
This is the power of a 'strong' technical background. You have to learn the tools that people, over 100's of years, have developed.
Now, did I find that getting through college actually 'educated' me? Not when I actually worked in electronic design. At first, I felt completely helpless, but with the help of a few experienced engineers who I worked for, and literally 100's of lunch hours in the company library, I was able to begin to grasp electronic design, and do independent work.
For example, almost 30 years ago, I 'INSTANTLY' understood what Herr Manger was doing with his exotic loudspeaker, once he gave me a reprint of the 1925 article of 'Rice and Kellog' on the design of the loudspeaker. The differential equations with dot notation form were readable on inspection.
Without my 'required' math background, I could still be scratching my head over what is really unique about his speaker design.
This is the power of a 'strong' technical background. You have to learn the tools that people, over 100's of years, have developed.
Now, did I find that getting through college actually 'educated' me? Not when I actually worked in electronic design. At first, I felt completely helpless, but with the help of a few experienced engineers who I worked for, and literally 100's of lunch hours in the company library, I was able to begin to grasp electronic design, and do independent work.
Wasn't looking your direction, Andy.......
Not surprised that you were not aware of the AES proceedings in class. No reason for it.
But if you do get hold of some old JAES, you will see Cherry, Grade, Cordell, et al, tear into him......or anyone who agrees with him. Yes, they can produce tons of eqautions, the mainstay of that useless debating society, to justify their postion.
But make no mistake, their main position was personality clashes.
Marshall has problably learned much since thenm, and if he writes that his postion has moved some, it is because his experience, coupled with his theoretical background, has given him further insight.
Till:
Are yo sure that you aren't going the sam university that I did?[joke]
Jocko
Not surprised that you were not aware of the AES proceedings in class. No reason for it.
But if you do get hold of some old JAES, you will see Cherry, Grade, Cordell, et al, tear into him......or anyone who agrees with him. Yes, they can produce tons of eqautions, the mainstay of that useless debating society, to justify their postion.
But make no mistake, their main position was personality clashes.
Marshall has problably learned much since thenm, and if he writes that his postion has moved some, it is because his experience, coupled with his theoretical background, has given him further insight.
Till:
Are yo sure that you aren't going the sam university that I did?[joke]
Jocko
I HAVE to chime in here:
John, I could not agree more!
I even skipped classes in college to spend time in the library reading books & journals and "looking at" 😀 those in languages I couldn't read.
I was almost in tears (note: I said almost - I am not a girly man!) when my company closed the library some years ago. Still haven't forgiven them. 😡 And I don't think they've realized yet how much of a setback that has created yet ...
Don't tell my wife, she thinks I'm nerdy enough as it is ... "What do you mean I've got too many books?" 🙄
mlloyd1
john curl said:
John, I could not agree more!
I even skipped classes in college to spend time in the library reading books & journals and "looking at" 😀 those in languages I couldn't read.
I was almost in tears (note: I said almost - I am not a girly man!) when my company closed the library some years ago. Still haven't forgiven them. 😡 And I don't think they've realized yet how much of a setback that has created yet ...
Don't tell my wife, she thinks I'm nerdy enough as it is ... "What do you mean I've got too many books?" 🙄
mlloyd1
A guy can NEVER have too many books, cds, lps, cars, rods, guns, dogs.....
If we don't have enough toys we get in trouble.
I have to say though, the internet has cut down on my magazine buying.
What you say about libraries is true. I have a university library card for just in case but every time I go there I throw a swivet. It simply isn't as good as it was 20 years ago. They're spending too much money on "systems" and not enough on acquisitions. Same malaise in businesses, I expect.
Staying sort of with the thread's topic, wanted to get a IEEE article the other day and they wanted $35!!!
They could easily get me for 10 articles for $50. So the internet isn't a panacea.
A good avenue of getting exotic technical info is the university engineering library. While, I am not a student at UC Berkeley, and I haven't attended in any way or more than 30 years, I can still go to the engineering library, go to the periodicals, and copy virtually anything that I want. Almost everything in electronic engineering is there, including 'Electronics World'. There must be many situations like this around the world. Check out your own university technical library.
john curl said:
Heh. I lived a Berkeley a few years ago, and I agree the Cal engineering library is a wonder. Only problem is, there are *so* *many* magazines to choose from that one could blow an entire day just following his nose.
Bay Area people might also like Stanford's libraries - the presence of CCRMA ensures audio related stuff is quite present. Haven't hit the UCSC engg library yet; I understand it's quite decent also.
Cheers,
Francois.
Math or no math (never needed more than algebra to date--my calculus and such continues to gather dust), I have always pursued wide bandwidth circuits because they sound better. One--not the only, of course--reason to use feedback is to increase bandwidth. The classic graph being the Bode plot where you've got something like 1kHz bandwidth open loop, then apply feedback and...hey, presto!...you've got 100kHz. Uh, guys, it doesn't take much in the way of experimentation to discover that a 100kHz circuit that achieves its bandwidth 'honestly' sounds better than one that gets there via feedback.
All of which is an indirect way of saying that lower amounts of NFB produce better sounding equipment. (Incidentally, I've been playing with a tube circuit that shows something like .06% distortion open loop--not too shabby for hollow state.) Generally, the bandwidth I'm seeking can be achieved with very simple circuits. Adding more and more stages cuts the bandwidth while increasing the gain so that more negative feedback can be applied to re-increase the bandwidth that you lost by adding the...huh? I missed something there. If your goal is to go from Paris to London, why go right on through London, across the North Pole, down though Russia, China, Australia, the South Pole, and so on...just to get to somewhere you already were? All that extra travel racked up time, expense and a lot of unnecessary wear and tear on you and your luggage.
Now, granted, I generally aim for something in the vicinity of about 250-400kHz (the tube circuit mentioned above comes in at about 375kHz--just about perfect in my book), but that's because I can get away with it here in the hinterlands. If I lived in NYC or Tokyo I wouldn't be able to do that--at least not without heavy shielding, etc. Yes, it makes people nervous who, like John or Nelson, have to earn their daily bread by sending out equipment into hostile environments. That said, and assuming that your local environment will allow it, it's an illuminating experiment.
The Calvinist attitude that if-it-sounds-good-it-must-be-bad is clearly a popular one in certain quarters. I have no problem if you get your jollies by sleeping on a bed of nails. Just don't go ranting at people who don't share your perversion.
Jocko,
In the past you've said things that could be interpreted as critical of my position regarding, shall we say...arrogant members. Several posts you've made recently sound as though you and I are on the same side of the fence, after all.
Just an observation.
Grey
All of which is an indirect way of saying that lower amounts of NFB produce better sounding equipment. (Incidentally, I've been playing with a tube circuit that shows something like .06% distortion open loop--not too shabby for hollow state.) Generally, the bandwidth I'm seeking can be achieved with very simple circuits. Adding more and more stages cuts the bandwidth while increasing the gain so that more negative feedback can be applied to re-increase the bandwidth that you lost by adding the...huh? I missed something there. If your goal is to go from Paris to London, why go right on through London, across the North Pole, down though Russia, China, Australia, the South Pole, and so on...just to get to somewhere you already were? All that extra travel racked up time, expense and a lot of unnecessary wear and tear on you and your luggage.
Now, granted, I generally aim for something in the vicinity of about 250-400kHz (the tube circuit mentioned above comes in at about 375kHz--just about perfect in my book), but that's because I can get away with it here in the hinterlands. If I lived in NYC or Tokyo I wouldn't be able to do that--at least not without heavy shielding, etc. Yes, it makes people nervous who, like John or Nelson, have to earn their daily bread by sending out equipment into hostile environments. That said, and assuming that your local environment will allow it, it's an illuminating experiment.
The Calvinist attitude that if-it-sounds-good-it-must-be-bad is clearly a popular one in certain quarters. I have no problem if you get your jollies by sleeping on a bed of nails. Just don't go ranting at people who don't share your perversion.
Jocko,
In the past you've said things that could be interpreted as critical of my position regarding, shall we say...arrogant members. Several posts you've made recently sound as though you and I are on the same side of the fence, after all.
Just an observation.
Grey
- Status
- Not open for further replies.