The problem we have is that the accepted electronics theories are based on sound knowledge accumulated over a couple of centuries by clever people. The underlying theory cannot be discarded on the whim of people who may not even understand it; that would require rewriting the whole of modern science and engineering. Contrary to the opinion of some, real scientists are working quite hard to find problems in current theory and so improve it. It is astonishingly unlikely that this, when it happens, will affect circuit theory and hence audio electronics.Joshua_G said:
We need to be careful to distinguish two separate issues:
1. what technical performance is needed to achieve audio realism?
2. what engineering techniques can provide this performance?
Some aspects of 1 are still up for debate and further test data. Most aspects of 2 are simply matters of being right or wrong.
A Turkish origin for Indo-European languages : Nature News & Comment
We are all speaking Turkish. So hit the hookah, have a trukish delight, sit back and enjoy.
We are all speaking Turkish. So hit the hookah, have a trukish delight, sit back and enjoy.
AFAIK, there is nothing wrong in the existing electronics theories.
What may be missing is implementing a fuller picture, derived from existing electronics theories, into the measurements and sound quality evaluations, done by many engineers.
For instance:
1. Possibly amps behave somewhat differently under constant signal (single frequency, single amplitude, or dual frequencies, constant amplitudes) and under complex signals, like those representing music.
2. Possibly the presence of RFI in analogue circuits modulates the audio signal - something which regularly, or traditionally, isn't tested for.
3. Possibly, the spectrum spread of harmonic distortion is more meaningful than THD.
4. Possibly, truly meaningful measurements ought to be performed (also) on an entire reproduction setup - as opposed to each piece of gear in itself. To my view, this has special significance in measuring various cables.
So, there is still hope. Until then, I can rely only upon my subjective impressions.
I concur.
It seems that the debate is related and correlated to the acceptance, or discarding, of subjective reports. Or, the debate is related and correlated to the degree certain engineers take the prevailing, accepted, measurements as giving the full picture concerning the degree of realism of audio reproduction. It seems to me that those engineers who are convinced that THD and SNR give the whole picture, concerning audio reproduction realism - will not have a second look at the prevailing, widely accepted techniques. Those who are open minded, will go on searching and experimenting.
Not necessarily.
Should there be agreed ways and methods of objectively evaluating sound quality, ways and methods that will coincide with the majority of subjective reports - it will be self-evident (and possibly measurable) which engineering techniques are more effective than others.
1. Partly agree. Single frequencies do not excite some possible problems because of the constant envelope. Music is not that complex, though. Response of amps to varying signal envelopes would be interesting.Joshua_G said:
2. Agree.
3. A number of THD weightings or other distortion measures have been proposed. Advertisers still use THD, which sometimes leads to an accusation that engineers believe in THD; most do not. Ignoring THD is as silly as giving it too much prominence.
4. Disagree. Well-designed units should not unduly interact with each other or cables.
Unlikely to happen when many subjective reports remain merely anecdotal and seem to disappear when investigated. As others have said, it is interesting that effects which circuit theory and known psychacoustics say should happen seem to survive proper testing, while those which science says should not occur are the ones which disappear. Why is 'test stress' so selective in its operation?
You are both correct.
In situ is absolutely how the systems should be tested.
and
Well designed units will be independent of the test enviro, be it the bench or in situ.
The large majority of the consumer audio equipment is not well designed with respect to EMC. As such, they are sensitive to their in situ environment.
Ground loops, common safety bonds are the main culprit, followed by inadequate internal component design to eliminate magfield interactions within chassis and to the external connections.
jn
Exactly. I would probably use less diplomatic words than "not well designed" and I would rather say their design is flawed with respect to EMC.
"Somewhat"? This stuff reads like some audiophile claims.
Root race - Wikipedia, the free encyclopedia
Don't forget they are going to migrate to Mercury.
So please be sure you don't confuse the actual historic Aryans with the reused named version.
I believe an even more diplomatic phrase would be:
As the discipline of ElectroMagnetic Compatibility begins to mature, previously unsuspected equipment interactions caused by inter and intra chassis currents are being discovered. This interaction becomes more significant when the equipment signal power gain rises to the levels of 10e+5 to 10e+7, and this interaction can affect the reference ground potentials.
Present equipment design criteria will take time to absorb this effect, and methods to combat it need to be established.
jn
Better?
THD weightings isn't widely accepted and practiced, it isn't an 'industry standard'.
Which leave the consumer not knowing how each piece of gear was tested. Thus, as a consumer, I cannot rely on published measurement results.
I didn't propose ignoring THD (for designers and manufacturers). For me, as consumer, THD is meaningless, unless exceptionally high, like way over 1% for a power amp.
It's impossible to cancel the interaction between the source output impedance, the target input impedance and the connecting cable impedance - over the entire frequency range of interest - and the receptivity or immunity of cables to picking up external EMI, including in the RF range - when attached to certain gear and terminated by it.
Also, it's my experience that different power cords have different sound signature (when feeding audio gear). Moreover, a power cable of certain make and model, when terminated power plugs of certain make but different models (different materials, or coatings) have different sound signature. I have no clue how to test power cords, for conventional, accepted tests will show no difference whatsoever. Possibly measuring audio gear with different power cords will show differences, I don't know, I never tried it.
Which is one of the main reasons why conventional, accepted tests mean nothing to me (as a consumer). I do know what I hear. When a test changes what I hear, or the way I hear - that test is definitely flawed.
It amuses me to note that I saw no one recommending DBT testing first the validity of the proposed test. Relying on a test, without first proving its' validity - is a very unscientific approach.
The experience of listening to music, enjoying it and appreciating it, is highly subjective. This is true to both listening to live music and reproduced music.
Most tests that try to neutralize the subjective aspect remove the major element in music appreciation (both live and reproduced), hence, undermining the ability of appreciating the degree of realism in reproduced music.
Which bring us back to what in my view is the major issue and obstacle: the test methods and practices. This apply both to electronics measurements and listening tests.
A test whereby the subject(s) don't know what gear is playing at any given time are perfectly legitimate and recommended. However, everything else should be checked. It includes listening to music the subject(s) is/are familiar with, on subject's own audio setup, in the subject's home - and prolonged listening of few long sessions, on different days.
Thank you for the link.
More new evidence that I belong to a semi-human stock.
Not a word about golden ears and development of super audiophilia
Nevermind, I’ll buy land in California.
George
Some interesting thoughts in your posts, Joshua ... personally, the THD fixation has demonstrated itself totally meaningless; having recently been able to 'force' ludicrously "mediocre", midfi equipment to give excellent accounts of themselves - and on other occasions hideously expensive and finely spec'd equipment was almost unlistenable to -- there needs to be 'another way' ...
The clue, as I see it, is how the low level distortion manifests - my ears have no trouble identifying this aspect, filtering it out; but how one should set up a measuring regime to provide numbers about such, is in the 'To Do' column ...
- Status
- Not open for further replies.