I feel frustrated that equipment like the AP5xx is out there, and all it's being used for, it seems, is the same old measurements, just with finer resolution. The actual methodology used in the testing is advancing at the rate of a Hawaiian lava flow - so nothing is going to get "burnt", plenty of time for dicky behaviours to move slightly sideways, and be completely missed ...
No our ears are pretty crap and any good microphone is better. Our brains do amazing things to create sense from our rubbish transducers, at which point its no longer an engineering problem, but one for Psychoacoustics. Expectation bias then adds to the confusion.
Heck, whisky tastes better out a cut glass tumbler. I know its the same whiskey and my mind is playing tricks with me, but I accept it. IF you feel the same about your flooby dust tweaks and admit it, no one here has an issue with that. If you claim with no evidence bar ' I heard it', then you are open to be challenged.
I also like listening to vinyl despite it being a compromised format, which I accept. And FWIW I would be very happy to have JC designed amplifiers in my house.
This is MY delusion. And I am comfortable with it.
The minds of rationalists are open... to evidence. And extraordinary claims demand extraordinary evidence.
Simple one, I've mentioned variations of this a number of times: record the distortion spectrum of an amp with low level signals while placing its power supply under greater and greater stress, easily done with a normal stereo amp - one channel driven at varying loads, stimulii, and the other has the same input throughout, and is the one measured for distortion.
Fancy, very high resolution audio test equipment is not really much better IN REALITY from earlier test equipment that would measure down to .005% or so. The tests are boring to set up and run with the more modern equipment, and you usually don't find much that you can work with in any realistic way.
Yes, of course you are.
And since no-one did that here, you might indeed be hearing that from a ghost.
Unless you're prepared to be that fussy why would anyone pay you? Plenty of other people out there prepared to be that fussy for fun. The world doesn't owe you a living.
Couldn't hack it at a professional level.
As for consumer audio design, the game's not worth the candle, because of the erosion of standards at that level. The answer is a professional approach, but unfortunately you reject that, instead contributing to that erosion.
Just like I used an entry-level world-class professional film camera all those years ago, now I use entry level professional sound equipment, follow best practises, and I don't even think about whether it's capable of doing good enough. I know that if I'm not getting world class results it's because of some failing on my part, no excuses. Just like I have a Fender Strat and a Marshall amp.
That's a bit of an exaggeration, because some of the equipment has stood idle a while, but I do have a Strat and a few amps including the Marshall.
No offence, 42, but we're trying to raise the next generation to be more technically competent than the previous, this flight in the face of reason stuff is not helping. It's very romantic and all that, but... if you're not part of the solution you're part of the problem. Historically the effect of the improvement in competencies has been to improve mankind's lot. The contrary view is called the Luddite fallacy.
Exactly. Feedback from simulations demonstrates the interactions between real world PS, and conventional audio circuits - therefore the actual bench results should be even 'worse' ...
OK. You got diffraction and maybe some resonance from the glass. Could make a difference.
In the present system incarnation where I put the speaker cables makes no difference to the sound.
Environment is touchy, no question. Years ago I had a living room in old house, lathe and plaster walls, with large single glazed windows, loose in worn sashes, 2 adjoining sides. The room was practically transparent to LF but the HF reflection from the glass had to be dealt with. Once that was fixed it was a fabulous listening room.
Tomchr provided that sort of evidence on a thread you were subscribed to. . http://www.diyaudio.com/forums/vend...er-achieving-0-0004-thd-n-22.html#post4091617. You then moved the goalposts. to asking for 19-20KHz IMD, which you got. But no attempt by you to make any meaningful analysis. But you have your data and can build the amplifier if you want.
Disclaimer: I liked the measurements so much I am building the amp, 4 of them in fact. I don't believe I can hear the improvements, but I will know that the power amp is not the limiting factor in my system by many dB. I am also aware of the huge bias that will give me and make everything sound better. And all for less than a silly cable
Well, you've totally got the wrong end of the stick, with me - I'm intensely interested in finding out intelligent solutions for improving audio - but a lot of what I see, from "objectivists", is a lack of intelligence: they start with denying key aspects of the whole scenario of audio reproduction and the listener - and if they are challenged they fall on the floor of the supermarket, kicking and screaming, "It ain't so, it ain't so!!" - mainly because a Great Man hasn't said that it could possibly be so,
...I'm interested in what actually happens, not what the literature says - reality trumps learned writings, I'm afraid. Feedback is a marvellous thing - if you keep going in the wrong direction eventually even the thickest skulled, most ego-driven individual usually gets it in the end - that one is wasting one's times thinking in a particular way. With my experiments in audio over the years I've slowly built up an understanding of what really matters for achieving "convincing audio" - and that good ol' feedback has always done its job along this journey, keeping me on track,
.
No, that is not at all what I just mentioned - the test Tom carried out was a totally conventional IMD, which does not measure what I'm talking about. It is very likely that his amplifier would do well in precisely what I'm mentioning now, but there are no results to analyse, in that respect.
I might mention again I did a LM38xx, which was not composite, but which achieved excellent subjective results, by devoting a major chunk of the resources to having a bulletproof PS - more than one way to skin a ...
Last edited:
No, that is not all what I just mentioned - the test Tom carried out was a totally conventional IMD, which does not measure what I'm talking about. It is very likely that his amplifier would do well in precisely what I'm mentioning now, but there are no results to analyse, in that respect.
I might mention again I did a LM38xx, which was not composite, but which achieved excellent subjective results, by devoting a major chunk of the resources to having a bulletproof PS - more than one way to skin a ...
He showed full power with an excellent and a merely adequate power supply. How is that not what you want to see? It proves that intelligent people look at the whole problem rather than just lobbing more capacitors in a box with the belief that magic will occur. Good engineering and good measurements. Works for me.
What disturbs people is that I claim to get good results, without doing it the conventional way - that makes sense, and I can live with it,
You can claim anything you want. Doesn't make it true and doesn't mean anyone else won't think your system sounds as bad as the colours on your TV look.
I might mention again I did a LM38xx, which was not composite, but which achieved excellent subjective results, by devoting a major chunk of the resources to having a bulletproof PS - more than one way to skin a ...
Yes, the most interesting result for me from Tom's thread is that the typical (non-composite) LM3886 is indeed limited, even from the pov of traditional measurements, by its power supply.
Then the interesting question at least for me becomes, in terms of power supply artifacts - is cure as good as prevention? Does a composite amp give the same subjective result as having a cap-array PSU?
- Status
- Not open for further replies.