Exactly! A sad state of affairs alright.
DIY audio is still just as much fun as a hobby though. What a shame there's so many here that seem to think this is an informal scientific peer review forum and have to keep telling us we're wrong to have opinions on what sounds good
Best wishes,
Owdeo
I found the Cheever thesis a good read, and am digging up his other writings. Of course to me he is preaching to the choir in my case, but I find his way of delivering not so preachy, it's nice. It is more in line with the subjectivist camp but that's what home audio is all about, who would really want to sit at home with audio that doesn't engage them?. IMHO the psychoacoustic aspects need a lot more questioning as we already have an abundant supply of the perfect maths in no small part thanks to all of the engineers 🙂.
Colin
Last edited:
except that more Real Psychoacoustics is readily available for reading, has been developed to a much higher degree in recent decades with the man hours invested in perceptual codec development
all of which indicates Cheever's reasoning is pretty useless, largely ignorant - even of the knowledge available at the time
and he's wrong about feedback too
so what exactly did you think you agreed with?
Ditto that, thanks John Curl and Colin.
BTW I think SY meant that sarcastically Colin.
Jcx,
First I must say I do not ever take anything at face value or follow an audio religion per se, I will willingly look into every aspect of audio and draw my own conclusions, some even based on intuition that has served me very well over the years, without ego or prejudice, confidence?, maybe. I agree with Dan Cheever on the audibility of certain distortion and do agree with the idea that higher distortion can actually sound more accurate than it's dramatically lower distortion counterpart. As otala motivated others to look deeper, even if if it theoretically meant disproving his ideas, Cheever is similar in my eyes. All the tech babble is great, it is needed, but this audio game is entirely about music reproduction and not computer science. There will never be one flavour, human taste simply will not allow that, just as much as no two ears hear the same.
Colin
He clearly polarized, his writings also helped me on the journey, for me it came to two things, either zero loop feedback, or increase it from zero until it sounds best again, no further.
Colin
Amps made in accordance with Cheever's thesis should be compared with amps having very high levels of NFB and very low THD preceded by a little electronic circuit delivering a fair amount of components of nice harmonic distortion as depicted by Cheever.
That's the 6 of one a half dozen of the other, inject it into a very low the thd amp or build it into the amp?, who's right, who's wrong?, the customers and enjoyers have the final say right?. I have my own thesis as many others do, and mine starts with the lowest added thd at the source whil the power end is acceptable to have 0.1% and still be called ultra low 🙂.
Colin
Colin
You can have as many opinions as you like on what sounds good to you. You can't have your own private version of maths and physics about how circuits work and what accurately reproduces a signal.owdeo said:
Many do, in fact. Of course, it doesn't lead to quality in design engineering, but that's not what "high end" audio is about, it's about entertainment, and private versions of math and physics are all part of storytelling.
Hear hear DF96. You cannot reinvent physics. Speaking of psychoacoustics, I read recently that a sample of listeners could not detect second harmonic distortion until the level was ten percent !!
I must admit I am curious to make a SET amp to try it on a range of program material.
So many audiophiles apparently still like them in spite of the plangent criticism from "blameless" Doug Self.
I must admit I am curious to make a SET amp to try it on a range of program material.
So many audiophiles apparently still like them in spite of the plangent criticism from "blameless" Doug Self.
Physics can and has been reinvented by the likes of Newton, Einstein, Bohr, Feynman and Hawking...but it's quite unlikely that a Principia Audica will emerge from the likes of the golden ear crowd.
Member
Joined 2009
Paid Member
I wouldn't say physics has been reinvented but rather it has been refined. Newton still holds true for most of what we experience, Einstein holds true for a little bit more and we are still searching for more refinement to explain a few more things. Physics is an extension of the way our minds work at times, trying to establish a model of the world inside our heads. Audio seems to be like this at times, we establish a model of the world of amplifiers inside our heads and this is why we end up with expectation bias etc. Have you noticed that that when the objectivist model-building self in you listens to music it is really just comparing things to the internal model, but when you move away from it and enjoy the music without thinking, the subjectivist is freed. As a result, the sound of an amplifier with or without gnf produces a different result depending on our state of mind. Some days the ultra low distortion amp pleases most, some days the SET amp pleases most. Isn't there room for both points of view ? (no math needed)
niggling details...
what's your standard of "proof" then?
Otala's TIM/PIM work proceeds from a wrong assumption to a wrong prescription, adds for free wrong claims about measurement,
the above shown by analysis, built hardware including Cordell's MOSFET Amp and Bob's custom phase selective IMD measurement hardware
even initial "converts" Walt Jung and Marshall Leach both later admitting in print that Otala's "flat loop gain" prescription was not a necessary condition to reduce TIM/PIM
still no evidence 40 years later for audibility difference of "FM" vs "AM" IMD
the "parsing" comment sounds to me a very odd statement - do you mean it as a criticism?
it seems to me that "parsing" is fundamental to analyzing theories, arguments, "stories"
careful parsing is needed to rescue anything from Otala's TIM/PIM work
what's your standard of "proof" then?
Otala's TIM/PIM work proceeds from a wrong assumption to a wrong prescription, adds for free wrong claims about measurement,
the above shown by analysis, built hardware including Cordell's MOSFET Amp and Bob's custom phase selective IMD measurement hardware
even initial "converts" Walt Jung and Marshall Leach both later admitting in print that Otala's "flat loop gain" prescription was not a necessary condition to reduce TIM/PIM
still no evidence 40 years later for audibility difference of "FM" vs "AM" IMD
the "parsing" comment sounds to me a very odd statement - do you mean it as a criticism?
it seems to me that "parsing" is fundamental to analyzing theories, arguments, "stories"
careful parsing is needed to rescue anything from Otala's TIM/PIM work
Last edited:
Otala was fundamentally right, even from the beginning. He just did not have 'all' the answers originally, and he made the mistake of not admitting some early 'errors', like extremely high open loop bandwidth was NECESSARY for high slew rate limiting, when it was only a useful practice for increasing slew rate. However, over time, he found that PIM or Phase Intermodulation distortion, was an ADDED contributor to why higher open loop bandwidth was necessary for ultimate audio reproduction. This has been shown to be measurable by Ron Quan in several AES papers, but it is not the only problem with low open loop feedback, because IF you make an gain block linear enough and fast enough even without negative feedback, then PIM can still be minimal even with a LOW open loop bandwidth.
However, I feel that there is STILL something else wrong with low open loop bandwidth, and for my BEST EFFORTS, I will use high open loop bandwidth in all my discrete designs. I will even attempt to use high open loop bandwidth IC op amps when I can. These are usually 'video' op amps, rather than 'audio' op amps, for some reason.
However, I feel that there is STILL something else wrong with low open loop bandwidth, and for my BEST EFFORTS, I will use high open loop bandwidth in all my discrete designs. I will even attempt to use high open loop bandwidth IC op amps when I can. These are usually 'video' op amps, rather than 'audio' op amps, for some reason.
Member
Joined 2009
Paid Member
John, your designs are likely now so good that further improvement is inaudible - why would you design any more amps ?
Matti Otala i think has been somewhat instrumentalized by people who had second thoughts, moreover when we look at the timeline.
Up to about 1975 any amateur could design, or use an existing design, and build an amplfier that was as good, if not better, as a commercial design.
Then in a matter of a few years, as insightfully pointed by Illimzn, rapid pogress in semiconductors that were avalaible only to big japanese firms definitly relegated the amateurs and a lot of little firms designs as outdated peformance wise with no mean to equal dirt cheap commercial amplifiers.
Matti Otala papers were used to "prove" that those superior amplifiers where actualy no good, that is, their better distorsion and BW products were dissmissed as useless and even counterproductive for the sound "quality", hence wth some kind of sleight of hand the advantages of thoses amplifiers were nullified.
So behind a pseudo technical debate the real aim was differentiation as a mean to keep existing in this industry.
Up to about 1975 any amateur could design, or use an existing design, and build an amplfier that was as good, if not better, as a commercial design.
Then in a matter of a few years, as insightfully pointed by Illimzn, rapid pogress in semiconductors that were avalaible only to big japanese firms definitly relegated the amateurs and a lot of little firms designs as outdated peformance wise with no mean to equal dirt cheap commercial amplifiers.
Matti Otala papers were used to "prove" that those superior amplifiers where actualy no good, that is, their better distorsion and BW products were dissmissed as useless and even counterproductive for the sound "quality", hence wth some kind of sleight of hand the advantages of thoses amplifiers were nullified.
So behind a pseudo technical debate the real aim was differentiation as a mean to keep existing in this industry.
Last edited:
I had a friend who was "into" SET tube equipment with no global feedback - using these with some large horn speakers using Lowther drivers. I wanted to see how these sounded in comparison to a negative feedback Class A (single ended) transistor amplifier I had built. Both had the ability to involve me in the music.
My friend thought the SET had a better bass response whereas I thought this was somewhat loose. I thought the transistor amplifier was more revealing at high frequencies.
I think we all perceive sound in different ways. On another occasion I took two other amplifiers I had built to the home of another friend whose interests were in classical music. Being regular in attending concerts he pointed out some sibilance in one of my amplifiers. To that point it did not matter to me at all.
I remember reading about a 1950's demonstration in the U.K. of a musician playing an orchestral instrument on stage in a hall behind a curtain and this being played back to the audience through equipment also behind the curtain. I think the results were used as an advertising slogan for the brand of equipment used in the test.
It would be interesting to repeat the tests with the audience with a skilled listening panel interspersed among general members of the public. Issue blindfolds to all give them an electronic device to enable audience voting. Set up systems using the same speakers and have a switching arrangement to engage a Global Negative Feedback subject amplifier or one with without.
Ask them to pick the differences between the two systems. Tally the results for the reviewing panel and the general public separately and consider the statistics. See what is the consensus of the combined groups and the separate divisions therein.
Last edited: