vynuhl.addict said:
I have no more to say about the largest part of your argument - have done it as well as I am able -
but I do want to agree with you about the quoted part above. No, the "scope" does not say that that improves linearity. Looking at amplifier stability says why it is used, but the penalty is often unacceptable high-order harmonic products. There are other ways of taking care of stability, often not examined enough, and they should be applied together with consideration of distortion spectra.
Dr John Ellis wrote a good article on one such practice, in EW a few years ago. If he reads here, he could give the reference.
Ah, a bit of time to myself again..... Saturday morning, family asleep still, I've just been for my jog, head is clear.
I visit DIYaudio because I love the interplay of ideas. I am very cautious what I say because I believe I have a unique product, and I won't give away the details, and I will not market it here either. Most of my buyers come straight to my website; this forum is made up of hobbyists who are not prepared to pay my price, and that means I can forget the commercial aspects, and simply debate the concepts.
Glen is outraged because I hold a diametrically opposed opinion by attacking the highly regarded, well engineered brands. Or so he thinks. The fact is I work with Ohms and Kirchoff's laws too, and I strive for linearity in all my designs. But in audio there is a psycho-acoustic element as well, and it must be addressed. He clearly doesn't suffer fools gladly, his prerogative, but we generally only learn one thing from the hostile, and that is hostility. Shame about the high sodium foccacia!! People are like avocados; soft and squishy on the outside, hard kernel of intellect inside. You have to go through the affective domain to get to the intellect, do it wrong, and no information can be exchanged.
I've just installed Comodo firewall, seem to be having trouble with it. Let's hope I can get this reply past the ports.....
Thank you Conrad, I appreciate the interest!
I understand your idea of sizzle now. I agree. When selling, and I stress I'm just having fun here discussing ideas with friends, not selling at all, you must find a differentiator from the 'why me' products out there. You need to highlight the value add aspects, point out that you don't just offer cup-holders, you have GPS navigation as well. People will pay much more for intellectual property than for silly gimmicks, but they don't always understand the subtleties of what is being offered, so marketing has to dumb it down a little. Hence terms like Optimos Class A, and Ultralinear, and single ended push pull. These are as much marketing as techical terms; they have meaning for both arenas, but are worked over linguistically to have retail appeal. Sliding bias is not so appealing, but dynamic quiescence, for all it's technical absurdity, is more appealing at the sale point.
I agree with this, but I'd caution that most people live in their heads, and this hugely influences their perception. If their physical or emotional comfort is disturbed during the two listening sessions, then it's more than likely that their perception will reflect different assessments. This corresponds roughly with our perception of the world when we are happy vis a vis sad. Very different indeed.
I'm not sure about this. In the seminal example Johan gives about lag compensation, a change from 47pF to 56pF will be heard night and day on two otherwise identical amplifiers. It's possible to measure two very slightly difference slew rates owing to the charge cycle on the miller cap, but if the amp is properly designed and slew rates are better than 10V/uS this does not explain the totally different sonic presentation. The FR will be identical within the audio band, in my experience.
I think that there are still subtleties in audio which cannot be measured using present equipment. The situation has improved hugely in the last thirty years; we now can analyse spectral distribution of distortion from signal tones, and this is a big step. However, for something approaching music, say even middle C on a 'good' piano, the complexity of the waveform really requires some sort of highly complex waterfall plot. I may be wrong, but I don't see this just yet. Graham Maynard talks of first cycle distortion, I believe this is on the right track, as that first perception of sound is important to the human ear as it's first and foremost a primordial communication mechanism with the emphasis on detecting danger.
All this rigor will cost a lot of money, and won't necessarily produce a better product. Only the creative skills of a good designer working with hunches, a lot of test equipment and a thorough knowledge of electronic theory, and particularly topologies, will achieve this at this stage. We are a long way from machine design of amplifiers.
In truth, Conrad, only a few of you feel this way, otherwise most consumer products would never be sold! I don't see too many people knocking on the doors of the carmakers, demanding why such and such a model is superior!! People just assume it's the same old tired technology recycled over and over in a repackaged body!! Of course, nothing could be further from the truth. The latest Lexus has over 100 million lines of computer code in its computers, controlling engine and transmission management, security, sound systems, air conditioning, steering, braking, etc. Any substantive proof of the effectiveness of these systems would be quite costly at the advertising end; it's adequate to simply say 'anti-collision software' and leave it at that, most people swallow it without qualm.
Many of my buyers are focussed entirely and exclusively on performance. They don't care if I use semiconductors drawn from a pteradactyl's tooth; the topology, the devices, are immaterial. They want that special sound.....
Because this advertising approach works. It gets sales, and people hand over their money. The psychology of selling is byzantine, and highly irrational. Why, in fact, do the really top class Neumann mikes still to this day use a vacuum tube? Does this improve measured performance? Interesting question......
Carlos,
You have it 100% right. We build sound systems for humans, not for snakes, giraffes, aardvarks or machines. Given the love of tubes out there, it could be that the straight wire with gain is not appropriate for this species. Furthermore, what of the perverse recording techniques used by the record companies to get that special ghetto blaster sound on popular music? How does that relate to the audiophile experience? And just what is an audiophile recording anyway? Are there published curves for this genre?
Johan,
Thank you for your courtly manners. No offense taken, I assure you!! Actually, I think THD is woefully inadequate for tube amps, particularly the SETs, which distort at manic levels, yet by repute are some of the best amplifiers in the world. Years ago I designed a Class A hybrid, using a 6SL7 t-barred front end at B+ of 340V driving a SE mosfet active load output stage. It dissipates 150W per channel (50V, 3A) for just 28W rms into 8R. FR is 25Hz-70KHz +/-1dB, distortion is around 2% at full power, yet it sounds like heaven on a stick. On the basis of specs alone this amp would be universally shunned, yet it's widely appreciated, and I have friends who say that they prefer it of all my amps. Such figures really do confound, I would suggest.
I hear your response to David's comment; a photograph of a truly great picture should render the picture with great acccuracy. Yes, there's an argument there I can see, no question. But if there were so much distortion there in the first place, and the photo added a bit of very gentle, slightly curvaceous distortion, would that not be appreciated too? (Particularly if it were of a pretty young lady and viewers had not seen the original!) These arguments could easily become circular, and I think you can appreciate that if the listener does not know the music well, or has a poor audio memory (and most of us do, in fact, have poor audio memory!) then the sound quality has no basis of comparison and it slips through our defences.....
You then say that you design for fidelity, as you have a notion that a precise copy of the original is required. This is a principled stand; you don't like distortion, so you choose to design for as little as possible. This is an idealogy, like fidelity in marriage, or communism, or the abolition of slavery. Like everything in the world, this is subject to the erosion of time, the corruption of popular usage. Since there are no amps with zero distortion, you could likely spend your life in its pursuit and consider it well spent. Ideas are very seductive, much more than even friendships, or even humanity itself. Witness Dafur, Auschwitz, Rwanda, Zimbabwe.
I have designed a special amp front end with around ten times less measured distortion than a standard diff pair. It's a sonic revelation. The accuracy and retrieval of detail is quite astonishing. However, I love a good single ended triode, too, and I built a 1.8W SET three years ago using the 6EM7 which I still love to listen to occasionally. It's particularly good on human voices. No one design has it all......
Thank you for your input,
Cheers,
Hugh
I visit DIYaudio because I love the interplay of ideas. I am very cautious what I say because I believe I have a unique product, and I won't give away the details, and I will not market it here either. Most of my buyers come straight to my website; this forum is made up of hobbyists who are not prepared to pay my price, and that means I can forget the commercial aspects, and simply debate the concepts.
Glen is outraged because I hold a diametrically opposed opinion by attacking the highly regarded, well engineered brands. Or so he thinks. The fact is I work with Ohms and Kirchoff's laws too, and I strive for linearity in all my designs. But in audio there is a psycho-acoustic element as well, and it must be addressed. He clearly doesn't suffer fools gladly, his prerogative, but we generally only learn one thing from the hostile, and that is hostility. Shame about the high sodium foccacia!! People are like avocados; soft and squishy on the outside, hard kernel of intellect inside. You have to go through the affective domain to get to the intellect, do it wrong, and no information can be exchanged.
I've just installed Comodo firewall, seem to be having trouble with it. Let's hope I can get this reply past the ports.....
Thank you Conrad, I appreciate the interest!
I understand your idea of sizzle now. I agree. When selling, and I stress I'm just having fun here discussing ideas with friends, not selling at all, you must find a differentiator from the 'why me' products out there. You need to highlight the value add aspects, point out that you don't just offer cup-holders, you have GPS navigation as well. People will pay much more for intellectual property than for silly gimmicks, but they don't always understand the subtleties of what is being offered, so marketing has to dumb it down a little. Hence terms like Optimos Class A, and Ultralinear, and single ended push pull. These are as much marketing as techical terms; they have meaning for both arenas, but are worked over linguistically to have retail appeal. Sliding bias is not so appealing, but dynamic quiescence, for all it's technical absurdity, is more appealing at the sale point.
I agree with this, but I'd caution that most people live in their heads, and this hugely influences their perception. If their physical or emotional comfort is disturbed during the two listening sessions, then it's more than likely that their perception will reflect different assessments. This corresponds roughly with our perception of the world when we are happy vis a vis sad. Very different indeed.
I'm not sure about this. In the seminal example Johan gives about lag compensation, a change from 47pF to 56pF will be heard night and day on two otherwise identical amplifiers. It's possible to measure two very slightly difference slew rates owing to the charge cycle on the miller cap, but if the amp is properly designed and slew rates are better than 10V/uS this does not explain the totally different sonic presentation. The FR will be identical within the audio band, in my experience.
I think that there are still subtleties in audio which cannot be measured using present equipment. The situation has improved hugely in the last thirty years; we now can analyse spectral distribution of distortion from signal tones, and this is a big step. However, for something approaching music, say even middle C on a 'good' piano, the complexity of the waveform really requires some sort of highly complex waterfall plot. I may be wrong, but I don't see this just yet. Graham Maynard talks of first cycle distortion, I believe this is on the right track, as that first perception of sound is important to the human ear as it's first and foremost a primordial communication mechanism with the emphasis on detecting danger.
All this rigor will cost a lot of money, and won't necessarily produce a better product. Only the creative skills of a good designer working with hunches, a lot of test equipment and a thorough knowledge of electronic theory, and particularly topologies, will achieve this at this stage. We are a long way from machine design of amplifiers.
In truth, Conrad, only a few of you feel this way, otherwise most consumer products would never be sold! I don't see too many people knocking on the doors of the carmakers, demanding why such and such a model is superior!! People just assume it's the same old tired technology recycled over and over in a repackaged body!! Of course, nothing could be further from the truth. The latest Lexus has over 100 million lines of computer code in its computers, controlling engine and transmission management, security, sound systems, air conditioning, steering, braking, etc. Any substantive proof of the effectiveness of these systems would be quite costly at the advertising end; it's adequate to simply say 'anti-collision software' and leave it at that, most people swallow it without qualm.
Many of my buyers are focussed entirely and exclusively on performance. They don't care if I use semiconductors drawn from a pteradactyl's tooth; the topology, the devices, are immaterial. They want that special sound.....
Because this advertising approach works. It gets sales, and people hand over their money. The psychology of selling is byzantine, and highly irrational. Why, in fact, do the really top class Neumann mikes still to this day use a vacuum tube? Does this improve measured performance? Interesting question......
Carlos,
You have it 100% right. We build sound systems for humans, not for snakes, giraffes, aardvarks or machines. Given the love of tubes out there, it could be that the straight wire with gain is not appropriate for this species. Furthermore, what of the perverse recording techniques used by the record companies to get that special ghetto blaster sound on popular music? How does that relate to the audiophile experience? And just what is an audiophile recording anyway? Are there published curves for this genre?
Johan,
Thank you for your courtly manners. No offense taken, I assure you!! Actually, I think THD is woefully inadequate for tube amps, particularly the SETs, which distort at manic levels, yet by repute are some of the best amplifiers in the world. Years ago I designed a Class A hybrid, using a 6SL7 t-barred front end at B+ of 340V driving a SE mosfet active load output stage. It dissipates 150W per channel (50V, 3A) for just 28W rms into 8R. FR is 25Hz-70KHz +/-1dB, distortion is around 2% at full power, yet it sounds like heaven on a stick. On the basis of specs alone this amp would be universally shunned, yet it's widely appreciated, and I have friends who say that they prefer it of all my amps. Such figures really do confound, I would suggest.
I hear your response to David's comment; a photograph of a truly great picture should render the picture with great acccuracy. Yes, there's an argument there I can see, no question. But if there were so much distortion there in the first place, and the photo added a bit of very gentle, slightly curvaceous distortion, would that not be appreciated too? (Particularly if it were of a pretty young lady and viewers had not seen the original!) These arguments could easily become circular, and I think you can appreciate that if the listener does not know the music well, or has a poor audio memory (and most of us do, in fact, have poor audio memory!) then the sound quality has no basis of comparison and it slips through our defences.....
You then say that you design for fidelity, as you have a notion that a precise copy of the original is required. This is a principled stand; you don't like distortion, so you choose to design for as little as possible. This is an idealogy, like fidelity in marriage, or communism, or the abolition of slavery. Like everything in the world, this is subject to the erosion of time, the corruption of popular usage. Since there are no amps with zero distortion, you could likely spend your life in its pursuit and consider it well spent. Ideas are very seductive, much more than even friendships, or even humanity itself. Witness Dafur, Auschwitz, Rwanda, Zimbabwe.
I have designed a special amp front end with around ten times less measured distortion than a standard diff pair. It's a sonic revelation. The accuracy and retrieval of detail is quite astonishing. However, I love a good single ended triode, too, and I built a 1.8W SET three years ago using the 6EM7 which I still love to listen to occasionally. It's particularly good on human voices. No one design has it all......
Thank you for your input,
Cheers,
Hugh
Latest News.
After careful consideration I have come to terms with the fact that the added VAS overcurrent bjt is detrimental to the sound. I have no technical explanation for this, sorry, but it destroys sonics on higher frequencies and lower bass, takes away from some of the effortless mid frequencies in which I love so dearly about this circuit as is. I think ill take the risks involved for better sonics.
Colin
After careful consideration I have come to terms with the fact that the added VAS overcurrent bjt is detrimental to the sound. I have no technical explanation for this, sorry, but it destroys sonics on higher frequencies and lower bass, takes away from some of the effortless mid frequencies in which I love so dearly about this circuit as is. I think ill take the risks involved for better sonics.
Colin
Hugh, very enjoyable chat, and I don't have much more to add save this one maybe OT story. My mom has a studio grand piano, a Knabe that was actually built a few towns away from where I live. She's had it since she was a teenager, about 60 years ago. I know its beautiful mellow sound very well. A number of years ago, a local musician ran a high-end audio salon here, where he had a new and well regarded (and probably very expensive) Yamaha piano. When I heard him play it, I was horrified. To me, it was strident and caused tremendous fatigue after just a few minutes. To this experienced and excellent musician, it was a top notch piano. So, beauty is in the ear and brain of the listener.
vynuhl.addict, I usually leave out current protection. My philosophy is, "I built it, if I blow it up I can fix it- several times if necessary." Surprisingly, I only seem to blow up commercial products, never my own much more vulnerable stuff. Go figure.
Regards,
CH
vynuhl.addict, I usually leave out current protection. My philosophy is, "I built it, if I blow it up I can fix it- several times if necessary." Surprisingly, I only seem to blow up commercial products, never my own much more vulnerable stuff. Go figure.
Regards,
CH
Hi, Vynuhl.addict,
There are many other examples, like non-degenerated bipolar input differential, non-degenerated VAS emitor, bootstrap VAS load, etc. These are "bad" things from good electronic design POV, but people likes the sound that those produces, so what can we say ?😀 You can experiment with bootstrapped VAS load. Make it ordinary bootstrapp, then modify it to non-clipping type (something like putting D3 referenced to a LED drop from supply like fig.8 of citation.pdf by NP, to make the output/driver cannot clip, the base cannot be more than the rail voltage). The modification will remove the sparkling and "speed" sensation from the sound. Old bootstrapp sounds better, inspite it is actually makes the output/driver clipping.
But this principle cannot hold for every amp. Like NP said, there is no such thing as best audio power amplifier. Every design has it's own place. For power lower than 50W maybe people will prefer these "coloured" designs, saying they sound better. But for higher power amp, up to 1000's watt amplification (up to area of Crown, LabGruppen, MC2 products), you will need the least coloured amps, the blameless ones. In this application, we can only use the least distorted amp, no other way. Not any "voicing" will work in this area. The cleanest, the most blameless will sounds the best here. Here those superb spec power amps shows their inner strength they are capable of. I can imagine if I use the principles of making <50W amp for making pro-amps, the sound will be terrible 😀.
Why is this? I don't know. Our ear is very precious and delicate device. But they also have their weakness. For example, you cannot say an amp sound good or bad IF you don't have another amp to compare to. Our ear simply don't have it's own standard, it can only compare. Another example, anything you hear more than 15 minutes will be regarded as "good" for our ear, because the ears are adjusting itself to these sound and sense anything as "good". So tuning an amp, loudspeaker, etc, needs a rest intervals, if not you will say anything sounds good 😀 And the next day when you hear the result, you may say to yourself "what am I doing all day yesterday?" 😀
Another thing that we usually fooled by is the source material. CD's, vinyl's, tapes, all have the "taste" of the recording enginner while working on the material. I learn this from a friend of mine.
Many people don't realize this (they take all recordings are perfect), so if their sound systems sounds not good, they will blame their gears, seldom thinks about the material used. They will spend alot of money trying to fix the sound by changing the cables, the amps, the speakers, but do nothing to change the source material. My friend shows me this by a demo, it really opens my eyes, how a certain CD has been edited, they sound completely different, much better. I began to know the difference between hearing the "soundsystem" and listening to music. If you are hearing the soundsystem, you will be playing the same album, the same song, at the same track time over and over, just to notice how long is the glass "tingggggg...." sound, or Don William's "wooooo..." but you never actually enjoys the whole song, you just listens to part of it. Another symptom of those who hears the soundsystem (not enjoying the music), is that they never listen to the whole CD/LP until finish, maybe they never listen to 1 track to finish either 😀 They change the song alot (from the same bunch of songs), only hearing a fraction of a song, and sadly they never play the song/album that they personally really like, just because those albums sounds bad on their soundsystem. Here I learn about the need of a good gears and not skeptic about tone control and equalizer. Through good gears and good understanding about recording material, you will be enjoying the songs/albums you really like 😀. I think this is what our hobby should pay off, our enjoyment. Not buying/changing gears, cables,etc every month in a "never ending story" route.
For average audio systems, it is OK to have a reproduction that sounds "more beautiful" than the live performance, our ear just like that beter. Usually the perfect reproduction like live performances will not sound good at low level/living room level listenings
There are some things that I observed among the audio community. Ordinary people (customer who don't understand electronics) have their own opinions about sound. Sometimes what they hear as "good" is actually a bad thing from good electronic design POV. Your case above maybe one of them.
There are many other examples, like non-degenerated bipolar input differential, non-degenerated VAS emitor, bootstrap VAS load, etc. These are "bad" things from good electronic design POV, but people likes the sound that those produces, so what can we say ?😀 You can experiment with bootstrapped VAS load. Make it ordinary bootstrapp, then modify it to non-clipping type (something like putting D3 referenced to a LED drop from supply like fig.8 of citation.pdf by NP, to make the output/driver cannot clip, the base cannot be more than the rail voltage). The modification will remove the sparkling and "speed" sensation from the sound. Old bootstrapp sounds better, inspite it is actually makes the output/driver clipping.
But this principle cannot hold for every amp. Like NP said, there is no such thing as best audio power amplifier. Every design has it's own place. For power lower than 50W maybe people will prefer these "coloured" designs, saying they sound better. But for higher power amp, up to 1000's watt amplification (up to area of Crown, LabGruppen, MC2 products), you will need the least coloured amps, the blameless ones. In this application, we can only use the least distorted amp, no other way. Not any "voicing" will work in this area. The cleanest, the most blameless will sounds the best here. Here those superb spec power amps shows their inner strength they are capable of. I can imagine if I use the principles of making <50W amp for making pro-amps, the sound will be terrible 😀.
Why is this? I don't know. Our ear is very precious and delicate device. But they also have their weakness. For example, you cannot say an amp sound good or bad IF you don't have another amp to compare to. Our ear simply don't have it's own standard, it can only compare. Another example, anything you hear more than 15 minutes will be regarded as "good" for our ear, because the ears are adjusting itself to these sound and sense anything as "good". So tuning an amp, loudspeaker, etc, needs a rest intervals, if not you will say anything sounds good 😀 And the next day when you hear the result, you may say to yourself "what am I doing all day yesterday?" 😀
Another thing that we usually fooled by is the source material. CD's, vinyl's, tapes, all have the "taste" of the recording enginner while working on the material. I learn this from a friend of mine.
Many people don't realize this (they take all recordings are perfect), so if their sound systems sounds not good, they will blame their gears, seldom thinks about the material used. They will spend alot of money trying to fix the sound by changing the cables, the amps, the speakers, but do nothing to change the source material. My friend shows me this by a demo, it really opens my eyes, how a certain CD has been edited, they sound completely different, much better. I began to know the difference between hearing the "soundsystem" and listening to music. If you are hearing the soundsystem, you will be playing the same album, the same song, at the same track time over and over, just to notice how long is the glass "tingggggg...." sound, or Don William's "wooooo..." but you never actually enjoys the whole song, you just listens to part of it. Another symptom of those who hears the soundsystem (not enjoying the music), is that they never listen to the whole CD/LP until finish, maybe they never listen to 1 track to finish either 😀 They change the song alot (from the same bunch of songs), only hearing a fraction of a song, and sadly they never play the song/album that they personally really like, just because those albums sounds bad on their soundsystem. Here I learn about the need of a good gears and not skeptic about tone control and equalizer. Through good gears and good understanding about recording material, you will be enjoying the songs/albums you really like 😀. I think this is what our hobby should pay off, our enjoyment. Not buying/changing gears, cables,etc every month in a "never ending story" route.
For average audio systems, it is OK to have a reproduction that sounds "more beautiful" than the live performance, our ear just like that beter. Usually the perfect reproduction like live performances will not sound good at low level/living room level listenings
Hi lumanauw,
I find myself very much trying to strike a balance, I want to be able to really enjoy listening to music and am much closer now than i ever have been. Coming from the perspective of being a musician I find myself weighing final sound grading on a few issues, can i understand the workings clearly in complicated pieces? do vocals(key)shout at me, or do they sing to me?(on well recorded pieces that is, are they engaging or do i find yself thinking about work the next day 🙂. I have chosen to base my recent project on the blameless design, this by its very core impressed me in the fact that each stage was looked at individually to give a good starting point. I have been listening to alot of music tonight and I am consistently so impressed by what I hear that I am excited to be sending Carlos a parts furnished board to let him hear with the same parts i use in mine and he can tweak at his own choice if he feels the need. No black gates or fancy caps except maybe a humble solen mkp and a panasonic FC for the voltage divider. There may be no best but there can be the ones that make you want to be a part of the performance and get that blood moving and become lost in the harmonic richness, to me it has to captivate rather than make everything except direct to disk recordings sound bad..
Colin
I find myself very much trying to strike a balance, I want to be able to really enjoy listening to music and am much closer now than i ever have been. Coming from the perspective of being a musician I find myself weighing final sound grading on a few issues, can i understand the workings clearly in complicated pieces? do vocals(key)shout at me, or do they sing to me?(on well recorded pieces that is, are they engaging or do i find yself thinking about work the next day 🙂. I have chosen to base my recent project on the blameless design, this by its very core impressed me in the fact that each stage was looked at individually to give a good starting point. I have been listening to alot of music tonight and I am consistently so impressed by what I hear that I am excited to be sending Carlos a parts furnished board to let him hear with the same parts i use in mine and he can tweak at his own choice if he feels the need. No black gates or fancy caps except maybe a humble solen mkp and a panasonic FC for the voltage divider. There may be no best but there can be the ones that make you want to be a part of the performance and get that blood moving and become lost in the harmonic richness, to me it has to captivate rather than make everything except direct to disk recordings sound bad..
Colin
A few further thoughts:
Obviously there is no such thing as a perfect anything. Fortunately the practical situation regarding hearing makes matters easier. The previously mentioned controlled tests, involving a total of thousands of listeners, satisfied the requirements of acceptable statistics (low deviation, dispersion etc.) Tests also included blind comparisons with live material.
Thus one can say today, to summarise, that distortion levels of below about 0,1% and discordant harmonics presence (this includes about all of high order odd harmonics) of below the threshold of hearing, constitutes an inaudible effect. This is where Self preferred to use the term "blameless" rather than scientifically perfect. I cannot enter Hugh's concept of there being "so much distortion to begin with ..." into the equation, if I understood him correctly. If he referred to the original, that is part of the input, whether beautiful, ugly, un-curvatious or whatever. To go back to the photographic analogy, the camera does not need to be scientifically prefect, it only needs to provide better definition than is detectable by the eye to provide a "perfect" reproduction.
That is what I call the high fidelity approach. It does not begrudge colouring amplifiers a place in the sun. If folks want to tread the streets until they chance upon something that suits their particular taste of musicality (whether close to the original or not) - it is their time and money.
Two last points: How often do you go to live concerts, and if you want to comment on equipment, when last did you have your ears tested? (Behold the experience of one audio critic, who sometimes found a system's h.f. response wanting. An audiogram revealed that he could not hear a scat above 9 KHz. No wonder.)
Regards.
Obviously there is no such thing as a perfect anything. Fortunately the practical situation regarding hearing makes matters easier. The previously mentioned controlled tests, involving a total of thousands of listeners, satisfied the requirements of acceptable statistics (low deviation, dispersion etc.) Tests also included blind comparisons with live material.
Thus one can say today, to summarise, that distortion levels of below about 0,1% and discordant harmonics presence (this includes about all of high order odd harmonics) of below the threshold of hearing, constitutes an inaudible effect. This is where Self preferred to use the term "blameless" rather than scientifically perfect. I cannot enter Hugh's concept of there being "so much distortion to begin with ..." into the equation, if I understood him correctly. If he referred to the original, that is part of the input, whether beautiful, ugly, un-curvatious or whatever. To go back to the photographic analogy, the camera does not need to be scientifically prefect, it only needs to provide better definition than is detectable by the eye to provide a "perfect" reproduction.
That is what I call the high fidelity approach. It does not begrudge colouring amplifiers a place in the sun. If folks want to tread the streets until they chance upon something that suits their particular taste of musicality (whether close to the original or not) - it is their time and money.
Two last points: How often do you go to live concerts, and if you want to comment on equipment, when last did you have your ears tested? (Behold the experience of one audio critic, who sometimes found a system's h.f. response wanting. An audiogram revealed that he could not hear a scat above 9 KHz. No wonder.)
Regards.
Very interesting.... very good point of views...this thread is beeing very
important to me.
thanks all,
Carlos
important to me.
thanks all,
Carlos
Hi Johan,
Most live concerts I know I have been to are way too loud so for me and sound good to bad depending on where you are luckily enough to be seated or standing ,I am carefull not to draw too much comparison..Secondly this cannot apply to multitracked studio recordings. Last time I had my hearing checked my hi-frequency response was to18khz, and had some loss in the very low frequencies in the left ear this was assessed by audiolab hearing for work. I always use hearing protection at live concerts and when we play shows as most club sonics are no less than aggravating for loud levels. I dont consider myself Golden eared whatsoever and number one do strive for low distortion sounds, but sometimes achieving those can alter high order harmonics. I have to agree with Hugh in the fact if I had a choice between the fundamental and more second haronic I would take this over increased high order harmonics which will do nothing but become a source of aggravation over second harmonics which will add a pleasing body to the tone. I guess in the world of fidelity it may come down to a "pick your poison". I honestly believe in striving for the truest input to output signal with the lowest added distortion in the higher harmonics, but i guess to really enjoy a perfect amp one would have to live in a anechoic chamber..For myself knowing the sound of these instrumennts in person I find myself looking for timbral accuracy for the most part, on recording that arent eqd to death which are most of them.
Most live concerts I know I have been to are way too loud so for me and sound good to bad depending on where you are luckily enough to be seated or standing ,I am carefull not to draw too much comparison..Secondly this cannot apply to multitracked studio recordings. Last time I had my hearing checked my hi-frequency response was to18khz, and had some loss in the very low frequencies in the left ear this was assessed by audiolab hearing for work. I always use hearing protection at live concerts and when we play shows as most club sonics are no less than aggravating for loud levels. I dont consider myself Golden eared whatsoever and number one do strive for low distortion sounds, but sometimes achieving those can alter high order harmonics. I have to agree with Hugh in the fact if I had a choice between the fundamental and more second haronic I would take this over increased high order harmonics which will do nothing but become a source of aggravation over second harmonics which will add a pleasing body to the tone. I guess in the world of fidelity it may come down to a "pick your poison". I honestly believe in striving for the truest input to output signal with the lowest added distortion in the higher harmonics, but i guess to really enjoy a perfect amp one would have to live in a anechoic chamber..For myself knowing the sound of these instrumennts in person I find myself looking for timbral accuracy for the most part, on recording that arent eqd to death which are most of them.
Vynuhl.addict,
I must agree with all your points, and I think it is an important contribution!
Regarding the loudness of some performances, you are so correct!! How often did I get a headache there, etc. I should have mentioned that my taste is more for symphony concerts, which you will immediately grasp is a different matter.
I would also agree with Hugh and yourself on the quoted point. Only those are not the only options - one does not need to have audible 2nd harmonic products in order to get low high-order harmonics. One can get all below audible level. But (if not said earlier) the non-generation of high-order products is most important and the downfall of many sometimes classy amplifiers.
If I may expand a little, one early source (I believe Peter Walker may have been involved) said that some high-order harmonics responsible for listener fatigue were so low (level) that they might have been below the conscious threshold of audibility. This would be very difficult to test, but opens up an interesting field. It is known that the brain can react to auditory sensations that we cannot consciously perceive. A simple example is that high levels of super audio frequency signals can cause a head-ache and discomfort. Etc, just for the record.
All this tech-talk might give the impression that one is constantly gazing at some screen or other. Just to confirm that I am still normal (well, hopefully moderately so!) - I measure and watch screens when I design and construct, but I also definitely switch all those off when I relax and listen to symphonies, Vangelis, Lloyd-Webber, Domingo, Rock Classics et al. That is why I got bitten by the bug in the first place.
Regards.
I must agree with all your points, and I think it is an important contribution!
Regarding the loudness of some performances, you are so correct!! How often did I get a headache there, etc. I should have mentioned that my taste is more for symphony concerts, which you will immediately grasp is a different matter.
I would also agree with Hugh and yourself on the quoted point. Only those are not the only options - one does not need to have audible 2nd harmonic products in order to get low high-order harmonics. One can get all below audible level. But (if not said earlier) the non-generation of high-order products is most important and the downfall of many sometimes classy amplifiers.
If I may expand a little, one early source (I believe Peter Walker may have been involved) said that some high-order harmonics responsible for listener fatigue were so low (level) that they might have been below the conscious threshold of audibility. This would be very difficult to test, but opens up an interesting field. It is known that the brain can react to auditory sensations that we cannot consciously perceive. A simple example is that high levels of super audio frequency signals can cause a head-ache and discomfort. Etc, just for the record.
All this tech-talk might give the impression that one is constantly gazing at some screen or other. Just to confirm that I am still normal (well, hopefully moderately so!) - I measure and watch screens when I design and construct, but I also definitely switch all those off when I relax and listen to symphonies, Vangelis, Lloyd-Webber, Domingo, Rock Classics et al. That is why I got bitten by the bug in the first place.
Regards.
Greg Erskine said:
🙄
Hey? I just posted that avatar because I thought she had nice eyes. I never even noticed what you are talking about.
Cheers,
Glen
Greg Erskine said:
Largos ? Where ?
I get seasick from looking at eyes.
Admitted, i'm a coco-"Nut", solid rock on the exterior and 90% lactation inside.
I've talked audio on ocassion with someone who did power electronics at Philips-U.
Whenever we discussed amp dissipation he started rolling his eye balls, peculiar view if you're adressing a strawberry jelly jar type with freckles.
Last time i heard he was developing high efficiency electrical gear in African starving areas.
Hi Everyone,
I thought that seeing that this thread has sort of drawn to a semi close and become a coconut thread 🙂, Id just like to thank everyone for their insights and contributions and technical knowledge, I originally got into this DIY Audio out of an unhappiness with the sonics of the offerings within my price range and I always feel fortunate to be a part of this group that has so much to offer and willing to take the time for constructive thoughts and offerings.
Thanks
Colin
I thought that seeing that this thread has sort of drawn to a semi close and become a coconut thread 🙂, Id just like to thank everyone for their insights and contributions and technical knowledge, I originally got into this DIY Audio out of an unhappiness with the sonics of the offerings within my price range and I always feel fortunate to be a part of this group that has so much to offer and willing to take the time for constructive thoughts and offerings.
Thanks
Colin
Re: I know you did not direct the answer to me Conrad..but i have something to say
This why we should be careful not make general statements like:
This amplifier sounds good with this and that design detail
Anything subjective is personal. Not necessarily true for anybody except for you.
With measurement we get some objective data that is not subjective, but can be used by everybody.
Some issues remaining to solve and to question are:
- How useful are these measured data?
- How do different data effect human perception, in general?
- What data measured are relevant, and what is irrelevant, regarding sound qualiy?
- Are there some better ways to get verified measured values to describe sound quality?
Earl Geddes have done some research with people at listening tests:
Distortion Perception Results
Sound Quality Distortion Perception
http://www.gedlee.com/
I want more such. Good investigations 🙂
Geddes has made me taking low THD figures with more than one grain of salt.
destroyer X said:
This why we should be careful not make general statements like:
This amplifier sounds good with this and that design detail
Anything subjective is personal. Not necessarily true for anybody except for you.
With measurement we get some objective data that is not subjective, but can be used by everybody.
Some issues remaining to solve and to question are:
- How useful are these measured data?
- How do different data effect human perception, in general?
- What data measured are relevant, and what is irrelevant, regarding sound qualiy?
- Are there some better ways to get verified measured values to describe sound quality?
Earl Geddes have done some research with people at listening tests:
Distortion Perception Results
Sound Quality Distortion Perception
http://www.gedlee.com/
I want more such. Good investigations 🙂
Geddes has made me taking low THD figures with more than one grain of salt.
I think this has been "beat to death" ,but solutions seem to have been found to satisfy everyone, (and they really work).
as discussed on the "CCS vs. Bootstrap shootout"
Bootstrap the CCS...😱
This seems to be the best of both words and has been
done on my "blameless" with success.
Current mirror or no current mirror..maybe add a cascode??
these changes to the small signal stages definitely have
the biggest impact on detail, soundstage which also can
influence how you perceive the bass response (psychoacoustically).
Just degenerating Re in the LPT (OLG) ,can audibly change the
amps charactor for better or worse.
Also, on a standard "blameless" the choice of devices
makes a world of difference (I learned this the hard way)
how much current you run them at (ideal Ic) or just basic
passives (silver mica vs. ceramic)can be more of a factor
than topology itself.
Ahh , but what it does at 100k+ has much to do with stability
and what it does/how it sounds in the audible range.
[example very good 100k square response means effortless at
20k, perfect at 1-10k etc.,right?]
OS
as discussed on the "CCS vs. Bootstrap shootout"
Bootstrap the CCS...😱
This seems to be the best of both words and has been
done on my "blameless" with success.
Current mirror or no current mirror..maybe add a cascode??
these changes to the small signal stages definitely have
the biggest impact on detail, soundstage which also can
influence how you perceive the bass response (psychoacoustically).
Just degenerating Re in the LPT (OLG) ,can audibly change the
amps charactor for better or worse.
Also, on a standard "blameless" the choice of devices
makes a world of difference (I learned this the hard way)
how much current you run them at (ideal Ic) or just basic
passives (silver mica vs. ceramic)can be more of a factor
than topology itself.
Ahh , but what it does at 100k+ has much to do with stability
and what it does/how it sounds in the audible range.
[example very good 100k square response means effortless at
20k, perfect at 1-10k etc.,right?]
OS
- Status
- Not open for further replies.