Not NIST maybe LIGO. I figure anyone here can see through 0.000000000225% THD claims to three digits no less. Since this is more than 64 bit float precision the claim of any simulated numbers are specious. I can see where this is going, no actual circuits will be revealed due to "IP" considerations, the same old merry go round.
Last edited:
The numbers aren't from simulations. They are from measured distortions with some applied math. It really is simple and believable if you know the theory, but I'm keeping my mouth shut 
Practical distortions will obviously be higher due to a number of factors, but with what I know about this type of circuit the practical distortion should not be far off from the theory disregarding noise.
You're right though I won't reveal the circuits, you don't know how badly I want to make it an open source thing, but it would be foolish since I can barely support myself at the moment and I got all my eggs in this basket. You can go ahead and reverse engineer it if I ever get it released. Just don't screw me over pls
Practical distortions will obviously be higher due to a number of factors, but with what I know about this type of circuit the practical distortion should not be far off from the theory disregarding noise.
You're right though I won't reveal the circuits, you don't know how badly I want to make it an open source thing, but it would be foolish since I can barely support myself at the moment and I got all my eggs in this basket. You can go ahead and reverse engineer it if I ever get it released. Just don't screw me over pls
Last edited:
By looking at the wealth of info on this forum you might get to figure out that " what you hear isn't what others might hear.....or like ". Everyone has preferences and they might not match with yours. Marketing a design isn't easy at all. You can see several good designers ( are you better ?) on this forum who haven't made a killing with their good designs....over the years! Low distortion isn't the holy grail !
Suggest you compare your systems with known good commercial products and sit with others who are into audio from your area and see what THEY think of your design. Who knows, if its so great then someone from the industry might make you an offer for work or designs ! Try talking to some of them. Doing it on your own......well as Voltwide just said....you might be 50 years late !
In the 80's the Japanese went down the VERY LOW distortion rabbit hole. Everyone followed them....for a few years. What happened ? It appeared that it wasn't the answer to good sound ! In fact some very well liked amps have distortion in the region of 2 to 3% like SE tube amps. They aren't producing accurate sound BUT people LIKE it and PAY for it !
You are bogged down in distortion figures I think. You've missed out on a lot of other factors I think. Keep reading posts on this forum there is a lot more to learn. All the best.
Last edited:
Sounding distortionless is a pretty low bar to clear.
I'm pretty sure the objections to your claims are because the numbers are incredibly hard to arrive at in reality on a PCB, not in SPICE. To get data you can trust from simulation is not easy; the results are only as good as the models and the setup. Even if you did as you claimed and applied some technique to estimate the distortion (like the op-amp noise gain trick for example), I think there is still going to be a practical limit. There may be solder joints with higher THD than your 0.000000000225%
. Surely 0.000000000225% THD is so far below the noise floor it's beyond academic anyway?It's hard enough for most people to not mess up the PCB in some way to achieve datasheet level performance for very high performance ICs and regulators.
The Jung regulator you cited, for example, has incredibly low output impedance, noise, and PSRR over the bandwidths I have seen measured. At some point the layout is going to dominate your realized performance. It's highly unlikely you can improve on this by anywhere near the factor you described except in SPICE.
Last edited:
That raises an interesting question - who is making the most profit from audio at the moment -- the market has changed a lot over the years, there's not many "hi fi" shops with listening rooms left. So, where is the money in audio, today? What does the team think?
@hellokitty123 - so far you have not convinced anyone here that you have anything special to offer, perhaps you will have better luck with kickstarter? You may raise enough funds to get some prototypes/products made, and let your backers test them out, that way, at least you can back up your claims. Good luck with your endeavor!
I'm not really sure why people are correlating theory with marketing. I am not going down the "low distortion" path. That's just a coincidence of design. My deign goal is to replicate reality while providing euphony.
I don't think we know how to measure everything we hear, I've experienced this countless times myself. Thus I design the most technically perfect amplifiers from all angles that I can in hopes I hit the mark.
I refuse to use GNFB or anything that "might" interfere with the sound quality, I take no chances.
I'm more than aware that THD means less than nothing to how something sounds, it's the basis behind all of my R&D. And obviously real life specs will not match the theory due to real life limitations, doesn't mean I won't implement them to the best of my ability.
If a potato sounded better I would use a potato.
As it stands the sound is spooky realistic with a completely realistic timbre and the entire spectrum is the pure opposite of fatiguing like it's covered in silk and oil. The treble is mindbogglingly smooth and non fatiguing.
Sound separation is perfect and everything about the sound is so clean and pinpoint precise.
I can convincingly believe the music even on music I previously thought was too artificial.
I think I have achieved what I am looking for as far as HI-FI sound goes. It's difficult for me to imagine the sound being any more technically perfect.
Adding distortion for euphonic sound is what I aim for next, some tubes or solid state parts forced into triode curves will probably do the trick here, at least that's why my experience tells me.
I don't think we know how to measure everything we hear, I've experienced this countless times myself. Thus I design the most technically perfect amplifiers from all angles that I can in hopes I hit the mark.
I refuse to use GNFB or anything that "might" interfere with the sound quality, I take no chances.
I'm more than aware that THD means less than nothing to how something sounds, it's the basis behind all of my R&D. And obviously real life specs will not match the theory due to real life limitations, doesn't mean I won't implement them to the best of my ability.
If a potato sounded better I would use a potato.
I disagree, I've heard a lot of expensive stuff and they all disappoint me (to put it nicely). When I say distortionless I mean no flaws in any aspect of the sound when compared to reality, that's the goal to me.
As it stands the sound is spooky realistic with a completely realistic timbre and the entire spectrum is the pure opposite of fatiguing like it's covered in silk and oil. The treble is mindbogglingly smooth and non fatiguing.
Sound separation is perfect and everything about the sound is so clean and pinpoint precise.
I can convincingly believe the music even on music I previously thought was too artificial.
I think I have achieved what I am looking for as far as HI-FI sound goes. It's difficult for me to imagine the sound being any more technically perfect.
Adding distortion for euphonic sound is what I aim for next, some tubes or solid state parts forced into triode curves will probably do the trick here, at least that's why my experience tells me.
Last edited:
This is the wrong place to convince people that a single component will do this i.e. irrespective of speaker choice. Kickstarter is a great idea, look at these guys in all probability they will never deliver anything but the the perfect showpiece for the growing number of folks that buy LP's and never play them.
MAG-LEV Audio | The First Levitating Turntable by MAG-LEV Audio —Kickstarter
I'm not trying to convince anyone of anything. I don't gain anything from that at this point.
Although I do admit the result of this sound is in the dac output stage I designed rather than the power buffer. The power buffer as far as I can tell is soundless which it better be, seeing as it has effectively no distortion.
I've heard it sound like utter garbage and the best thing I've ever heard so clearly it is not a bottleneck here.
As far as sound tweaking goes my focus will be on the dac stages and distortive stages mostly at this point.
I've thought about kickstarter, my fear is that it will be too effective. I have no manufacturing or distribution set up and the "product" is not near in presentable state.
I need to find someone capable of measuring my system before I go touting official specs too, which will no doubt be the driving force behind the interest of the average audio consumer as most people still think specs = sound.
I want to finish R&D covering all the angles before I release it.
Then there's the problem of, whats stopping people from instantly copycatting my designs?
Although I do admit the result of this sound is in the dac output stage I designed rather than the power buffer. The power buffer as far as I can tell is soundless which it better be, seeing as it has effectively no distortion.
I've heard it sound like utter garbage and the best thing I've ever heard so clearly it is not a bottleneck here.
As far as sound tweaking goes my focus will be on the dac stages and distortive stages mostly at this point.
I've thought about kickstarter, my fear is that it will be too effective. I have no manufacturing or distribution set up and the "product" is not near in presentable state.
I need to find someone capable of measuring my system before I go touting official specs too, which will no doubt be the driving force behind the interest of the average audio consumer as most people still think specs = sound.
I want to finish R&D covering all the angles before I release it.
Then there's the problem of, whats stopping people from instantly copycatting my designs?
Last edited:
Well, you will have to finish designing your product to begin with (so far you are about half way there, since you need somebody to design the other half), and then build some prototypes, in the flesh, not just some protoboard jungle and even less plain simulations.
That requires money and getting money requires marketing.
Either market the product as something worth manufacturing or at least market yourself as a cool Designer worth helping.
"Money makes the World go round" and you need some of it, hence: Marketing is involved.
Well, "low distortion" is a measurable goal, easy to check whether you achieve it or not; while "replicate reality while providing euphony" is Poetry, Fairydust, Mojo, 23000 people will have 23000 different opinions about it and nobody will convince anybody else.
We too.
You mean you "try" to design
It is for others to verify you actually achieved that.
I´d LOVE to see which parts you use that are SO linear that their combined distortion adds up to 0.000000000225% THD without NFB.
Wait a minute: when you compare your amp to others, they can not be listened at on their own, you need some live sound source (say, a violin or piano or orchestra), which in due time went through a lot of elements, from microphone to preamp to mixer to recorder to eventully some Audio file which is then decoded and played through your Amplifier and then through Speakers (or Headphones).
And the reproduction you hear is compared to the Live Audio Source? (the violin, piano, etc.) and you say they perfectly match each other?
Forgive me but all those elements, even if the most advanced State of the Art, each and every one will produce at least a little distortion .
Suppose all audio chain elements add up to , say, 0.3% distortion (an impossible goal but let´s accept it for a minute, on the argument´s sake).
Do you think **anybody** can distinguish between 0.3% and 0.300000000225% ?
Because that´s exactly what you are claiming.
You can NEVER hear 0.000000000225% directly, it will ALWAYS be buried in the rest of the chain´s distortion, how do you separate one from the other for analysis?
Since the rest is 1000000000 times larger/stronger.
Let alone the noise floor.
A bit of advice, prosaic circuit patents tend to be of little use there is virtually always prior art somewhere buried in time after all transistors have only three legs and there are only so many ways of connecting them.
As for computed and not simulated results, since just about any serious amplifier has 100 or so nodes including internal device nodes, hand calculations would involve inverting 100 X 100 matrices by hand. So it remains a mystery as to what you mean by calculating answers using 16 digits by hand.
Agreed which is why I started by making the amplifier as technically perfect as possible. I think we can all agree on what sounds "real".
Euphonics are a flavor that can be added after.
I think you are trying to read way too much into my words.
I never said anything about not using NFB. I said no GNFB.
Perfect is a strong word. I would say very convincing. As you say I have no live orchestra to directly compare but in my room I can believe it is real and that is what I aim for.
Well obviously
No, I think THD as a whole is almost useless as a spec. But I can't exactly say my amplifier has "98% euphony" can I?
I never claimed anything of the sort. Once again you are reading way too much into my words.
As I've said, I don't think we can measure all we hear. Since we can't measure what we hear what parameters do we design under?
The only thing we can do is eliminate all of the known flaws in the system to the best of our ability and hope we get it right, or add distortions until it sounds good.
I only required third grade math in this case. You're trying to understand the answer without knowing the problem. Unless you know the design there is no point in making presumptions about how it is calculated.
Last edited:
OK so no complex variables and no performance vs frequency. You don't have to disclose anything by demonstrating the response of an RC low pass with nothing but third grade math.
That might be interesting, but I was interested in the third grade math IIRC percentages is a fifth grade module in the standard curriculum.
"euphony" is a useless term anyway, as it is by definition subjective...
And there you have your next problem - that is subjective, and also very dependent on the source material, listening environment, etc...
Here's a clip I took a while back of 60v p-p into 8 ohms using my design.
The distortion goes down as I increase the amount of output devices in parallel, this is in line with the fact that the theory will not match the practice due to practical limitations and noise floor.
However using headphone loads it approaches -300db in sim.
Personally I enjoy high distortion tube amps, I think euphony and hi-fi are two sides of the same coin but complete opposites of each other. To me euphony is when you brain releases dopamine from the sound and creates a "high" in your brain. I've only experienced this from high distortion tube set-ups, except for one example.
On the other end of the spectrum we have hi-fi where everything just sounds impressively real but not necessarily dopamine inducing.
Headphones don't have room conditions as a variable and they are difficult to convincingly make you believe you are "there". A good challenge.
The distortion goes down as I increase the amount of output devices in parallel, this is in line with the fact that the theory will not match the practice due to practical limitations and noise floor.
However using headphone loads it approaches -300db in sim.
I understand where you're coming from but I have to agree to disagree. Since we all live in real life I think it is more of an objective thing. I'm not saying everyone will enjoy such a thing.
Personally I enjoy high distortion tube amps, I think euphony and hi-fi are two sides of the same coin but complete opposites of each other. To me euphony is when you brain releases dopamine from the sound and creates a "high" in your brain. I've only experienced this from high distortion tube set-ups, except for one example.
On the other end of the spectrum we have hi-fi where everything just sounds impressively real but not necessarily dopamine inducing.
Agreed, this is why I use HD800 headphones for sound tweaking. IMO if you cannot get it right into a 300 ohm load that has more detail most speakers can hope for then you cannot get it right into an 8 ohm load with room conditions as a factor.
Headphones don't have room conditions as a variable and they are difficult to convincingly make you believe you are "there". A good challenge.
Last edited:
- Status
- Not open for further replies.