Feedback affects Soundstage, Imaging, Transients ?

Status
Not open for further replies.
SY, I understand you. But please, try it yourself. Compare the NJM4565D and the OPA2134.

Yes, an amp should be neutral. But for some it can be ok when it "lifts"
for example pop-music when you are not interested in the real record.

I don't know how to go on and convince the critics to try it.
I want to put the worse NJM4565D back in the CD-Player!
 
A lot of revelations involves memory, which is not that great by a long shot. How many of us have listened to our set-up and after being on holiday for a few weeks come back and listen again asking yourself, I am sure it sounded better before. And yet nothing has changed.


Me not - if at all, rather just the opposite.
Different aliens around my place ???

🙂
Michael
 
SY, I understand you. But please, try it yourself. Compare the NJM4565D and the OPA2134.

Yes, an amp should be neutral. But for some it can be ok when it "lifts"
for example pop-music when you are not interested in the real record.

I don't know how to go on and convince the critics to try it.
I want to put the worse NJM4565D back in the CD-Player!

The way I see it is that an amplifier should be completely transparent.

I propose a Turing test for opamps at unity gain = maximum feedback. Simple ingredients: opamps and a switch. Now, feed a signal to the switch so you can have direct feed (A), or feed through the opamp (B), to an end amplification stage and a fair pair of speakers.

Switch and listen, ABX. The only criterium is whether there is a difference between A and B, not if one of the two sounds better. If there is no difference, there is no difference, which is exactly what I want from an amplifier that is completely transparent. If there is a difference, the modified signal might be preferable to some, or better, but it would still be wrong, since this is not what a perfect amplifier should do.

I did this test some time ago with a couple of opamps I had laying around, in a very unscientific setting, with me both behind the switch and listening at the same time. Very disappointing in that I could not hear any differences. None, nichts, nada, zip, niente, rien, helemaal noppes.

So, I came to the conclusion that you will need to daisy chain a number of opamps in order to reliably hear differences. And this is what the Turing test for opamps is all about. How many opamps of a certain design do you need to put in series, before a human being starts to notice a difference from the original signal? To that end, let me define the VAC number for a given opamp VACi, where i is the number of opamps needed to be put in series before there is a statistical significance that >50% of a control group will hear a difference from the original signal.

Since I could not hear a difference from one opamp, for me that settles the question if feedback destroys transients. It doesn't for me. The question rather is, how many of such feedback loops over a nice modern opamp can you daisy chain, before audible artifacts become significant. My guess is that you might need quite a number of the best ones.

vac
 
Vacu, you must have used flawed components like speakers or headphones to come to that absolute conclusion. It is an invalid test.

Tannoy K3838, QuadII, MC2105, old stuff like that, old ears too.

It is not an conclusion I hold to be true for everybody, just my own experience. I must add, an experience I would like to objectify as much as possible, because the results of such testing would be relevant to me and many others. At all stages of the signal chain, opamps are used, with feedback, whether you like it or not from a conceptual point of view. It would just be very practical to know which design has the highest VAC number.

Please, do the test yourself on your own speakers, it only takes an hour or so to scavenge the components from your work bench and bang them together.
 
Looking at only piece of the equipment at a time does not cut the mustard i think.It is always the complete chain. Dark Side Of The MOON was recorded on a NEVe board as far as i know with bazzillons of NE5534 in the chain although biased into Class A and all kind of tricks. Still this record and especially the remastered limited edition of the CD can fry my tweeters in a minute. The only other record i know that can do that was recorded on a Levinson modified Studer. It is called "A Life", a drum recording.
 
Rupert Neve was working on it, even at that time. BC109 and 2N2484 is of cause even worse. Maybe it was "Mystery And Imagination" that used that board. Not a bad recording ether. Anyway, it is the artists uncompromising will that creates the art and not the equipment. Even in the stone age. That is the systematic problem i see in this discussion.
 
BC109 and 2N2484 is of cause even worse.

Yet, they are good enough if there s minimal care in the designs.


Anyway, it is the artists uncompromising will that creates the art and not the equipment.

Their DSOTM record was created thanks to huge amounts of money.
No other musical combo had as much costly equipement ,
surely more than a million $ in their personnal studio...

Creating something original just with one instrument , like Coltrane,
is what can really be called art, and indeed, that s in line with Joseph
Beuys s concept of social art , contrary to the mega businesses like PF.
 
Der Meister zeigt sich in der Beschränkung.
Sorry, i can not translate that into english. I know i violate a forum rule.
Yes, i agree with my heart. I did not meet Coltrane but i was a pupil of Beuys. The nothingness is the fullness. The concept of transparency is illusion. The artist needs a canvas to paint on. Possibilities, a palette at least.
 
Their DSOTM record was created thanks to huge amounts of money.
No other musical combo had as much costly equipement ,
surely more than a million $ in their personnal studio...

Creating something original just with one instrument , like Coltrane,
is what can really be called art, and indeed, that s in line with Joseph
Beuys s concept of social art , contrary to the mega businesses like PF.
I like both. I suspect PF only got the really big budgets after the success of DSOTM. That was laboriously recorded using analogue synthesizers.

How many opamps are in the chain AFTER the tracks are mixed is a good question. To some extent the filtering and effects on the individual source tracks do not really count as much
 
Status
Not open for further replies.