What is wrong with op-amps?

Status
Not open for further replies.
English has a remarkably rich vocabulary, though there's a German word I'd like to see brought in to it. In any case... as you and I both pointed out, in order to achieve audibility, the distortions in the IC op-amp conference paper had to be magnified by 40dB. There was apparently no comparative test of a discrete circuit to see if any differences specific to an IC could be aurally detected. I'd hope that a referee would pick that up if this were ever submitted as an actual JAES paper and the authors would correct that.
 
bear said:
The point is that Geddes seems to have found that human preferences do not correlate to merely low distortion figures.
True. Some actually prefer higher distortion figures, and of course there are other things which matter too such as frequency response. Geddes, like most of us, is not a fan of THD. If I understand him correctly, he does believe that distortion matters which is why he invented his own distortion measure. If you were simply using "THD" as shorthand for 'any distortion measure' then it would have been better if you said so.

Body language?
I believe that is the one. We all send and receive such signals; the fact that we are often not conscious of this actually makes these signals more important as we can't consciously ignore what we don't realise we are receiving.

jan.didden said:
I read that paper. Not sure what the value is of this.

They fudged the circuits such that the opamps produced 100x more distortion as in normal operation, to make it audible / measureable.....

Then they say:

"The assumption is made that these distortions still have some effect on listener perception of character and quality even when they are below the detection threshold"
So John Curl once again posts a link to a paper (actually just a conference proceeding?) which does not actually support his views, even though the title strongly implies that it does. Wait a minute, I think I get it now: JC posts links to papers written by people who think like him, even though the actual data in their papers does not support that view.
 
I read that paper. Not sure what the value is of this.

They fudged the circuits such that the opamps produced 100x more distortion as in normal operation, to make it audible / measureable.....

Then they say:

"The assumption is made that these distortions still have some effect on listener perception of character and quality even when they are below the detection threshold"

Well, if this is the assumption, why go through the whole testing stuff? They could have just put the assumption under the title line and be done.

Jan

Hey guys. You're advised to immediately stop using your amplifiers, as Jan has identified a conceptual issue with this thing "amplification".

:joker:
 
English has a remarkably rich vocabulary, though there's a German word I'd like to see brought in to it. In any case... as you and I both pointed out, in order to achieve audibility, the distortions in the IC op-amp conference paper had to be magnified by 40dB. There was apparently no comparative test of a discrete circuit to see if any differences specific to an IC could be aurally detected. I'd hope that a referee would pick that up if this were ever submitted as an actual JAES paper and the authors would correct that.

If they correct it, it all falls apart, so will not happen.

I always considered Massenburg a a serious guy but this is so flawed, I would be ashamed if I did something like this.
Don't now the other guys.

Jan
 
Last edited:
So John Curl once again posts a link to a paper (actually just a conference proceeding?) which does not actually support his views, even though the title strongly implies that it does. Wait a minute, I think I get it now: JC posts links to papers written by people who think like him, even though the actual data in their papers does not support that view.

Don't tell me you only noticed it now??

Jan
 
Last edited:
If they correct it, it all falls apart, so will not happen.

Maybe, maybe not. I look at it this way: if they show that with a 40dB magnification of distortion, there's a systematic audible difference between IC opamps and discrete circuits, that's interesting (if not particularly useful, since ordinary circuits do not provide 40dB boosts to distortion). If they show no systematic difference, that's equally interesting and also not particularly useful, except as an academic exercise.

But of course, many journals papers are only of academic interest.😀
 
"You guys" are a piece of work.

Really.

The paper Curl referenced was for the purpose of pointing out a new(?) look at a possible mechanism for "differences" that escaped "old" measurement methods.

A lot of buzzing around the hive, once the bees get disturbed (from their nice placid world).

So, yes the discussion is not about headphones. However a participant, Kirchoff, ventured out of the hive into daylight and told us what he listens to - the rest of you are unwilling for "personal reasons" (that only you know - but I can guess!). And we find that (despite protestations to the contrary) that there are real and significant flaws in independently measured data (for his phones). Some participants, the same ones who demand "data/measurements/DBTs" wish to sweep this aside. Why?

Then too, the owner of said phones goes on to tell us that he picked them, HOW??
On the basis of measurements? DBTs? Nope.
Personal listening.

Odd, isn't that??

Now, I'm going to say again, that if you've never experienced reproduction that meets the criteria that I put forth earlier, then you just don't know, and the basis for your opinions and beliefs is highly limited and problematic for people who take such absolutist views.

Is it possible for you measurement freaks to say that there is a reasonable possibility that just perhaps what I am saying is true??

And the question on the table is "what is wrong with opamps".

When a poster put up about independent power supplies for each stage, what did you engineering and measurement freaks do with that? Dismiss it out of hand. Hmmm... really? Babble on about things that are merely bickering and arguments leading nowhere, based on nothing more than strongly held opinions, backed with little substance. Peh.

Oh wait, the "engineer's" view is that it is "good enough" now, so there is nothing more to do. Is that it??

Please, give me a break.
 
Last edited:
Most of us have come across equipment that measures well but sounds dull, measurement alone is not enough, we all still use out ears in final judgement. But without the use of measurements, how do you design a circuit? That's why they are so important. I look for decent measurements AND good sound quality, but then i am a fussy git 😛
 
Op-amps in home audio compromise the signal which has previously passed through hundreds of the things in a studio, in the same way that a normal power lead compromises the electrons which have previously passed through many miles of cheap alloy transmission lines and transformers.

It's B/S folks, it's not so much what you use, it's how you use it.

Of course i'm forgetting about those who just come here for a free argument 😉
 
Gentlemen,

As much as you Subjectivists may dislike this: the burden of proof is still on you.
The ability to perceive differences can only be established only in DBT. As SY has often said: peeking is an absolute no no. Even digital tracks appear to be analyzed first before listening is done, so that technique does not count. Cheating is simply not permitted.
Even then there should be 100% identification in a DBT in a fully controlled listening environment: And, if no 100% identification has been scored, its has been nothing but rolling the dice.


Why is such a test so hard, Subjectivists? An why do you always chicken out with the usual excuses about the fallacies of DBT?
After all, you all are convinced differences between Opamp circuits and properly designed discretes are clearly audible. So it should be a piece of cake, wouldn’t it?
You were the ones with the claims, you are the ones that have to substantiate. Everything elso is purely anecdotal. So far, the anecdotes and handwaving occupy about half of the 1778 posts.

Good Luck,

Eelco
 
Status
Not open for further replies.