pg. 208 Stereophile mag Oct 2007 Industry Update

Status
Not open for further replies.
stoolpigeon said:


I'm a bassist...and, yes, I can play bass in that manner. I also play guitar, but not the way Satriani does--that would take several years of dedicated work on guitar and that's a luxury I cannot afford.
I take it from your snide tone that your answer is no. Until you can play at that level, kindly refrain from unkind comparisons.
There's an interesting thing that happens when I put down my bass and take up guitar for a couple of weeks. When I play bass, I don't think all that much of Eric Clapton's playing--not bad, mind you, just not necessarily worthy of all the praise he's garnered over the years. After I've played guitar for a while, I begin noticing different things in his playing and my respect for him grows. Then I put down my guitar and go back to bass and, sure enough, my opinion of Clapton slides back down a few notches. I've done the cycle any number of times, but my perceptions remain the same regardless. Curiously, he's the only guitarist I feel that way about...why, I do not know.

Grey
 
Satriani?

I was going to be more diplomatic but I have to agree with Bogdan.

The thing is Grey that when you set yourself up as as a (self proclaimed) golden ear and talk about listening to sound bounce off walls (who the hell does that?) and and then you compare Satriani (of whom I have nothing against) to giants of the jazz and classical world it is not unreasonable for people to wonder if you have any idea what you are talking about.

Sorry about the long sentence.
 
It's a fascinating reflection of human nature to watch people seize on the most trivial of things so as to ignore the larger point.
As a bit of narrative I mentioned that I was listening to Joe Satriani while I was etching a circuit board in the kitchen. The fact that I'd been listening to Saint-Saens earlier didn't enter into the narrative so I didn't view it as relevant. Now a number of posts have been wasted on Satriani (and by extension, not-so-veiled criticism of what I choose to listen to while I etch circuit boards), but not one post has dealt with the basic issue, which is that it all boils down to attention to detail when you're listening. Not background music to etch circuit boards by, but really listening. Given that each successive post has strayed farther and farther afield, it's becoming more and more like trying to hold the attention of ill-behaved first-graders...some are staring out the window...some are passing notes to each other...some are doodling in their notebooks...some are throwing wads of paper...but they're not paying attention.
To the extent that this reflects the personality of the posters involved, it's no wonder that they cannot hear differences in equipment...they're too distracted by the ladybug on the windowpane.
Perhaps by the time you get to high school you'll have learned enough to settle down and focus.
Or maybe not.

Grey
 
jackinnj said:
I wonder what Charles thinks of this evil opamp -- optimized for audio, 190MHz BW, 1,900V/uS

Here's what I say -- why bother with feedback? I built zero feedback video amplifiers for our DVD player that had 400 MHz bandwidth.

There are two points to keep in mind:

a) Every time I have compared a circuit with feedback to a circuit without feedback, the zero feedback circuit sounds (or looks, in the case of video) better.

b) It's not that hard to build a circuit to just about any required performance level without using feedback.

So the real question is "why use feedback?"

Yes, I suppose I could have used feedback to get 800 MHz bandwidth instead of 400 MHz, but why bother? Or I could reduce the distortion of our new line stage from 0.002% to 0.001% by using feedback, but do you really think it would sound better?
 
rdf said:
Hi Charles. Because of what can be the contentious nature of definitions around here, did you mean the performance described is achievable without local degeneration or without inter-stage feedback? (And congrats on the Stereophile award.)

OK, it's time for another argument about semantics. Here's the deal:

a) There are more different ways to apply feedback to a circuit than there are terms to describe them. So invariably there will be confusion when talking about feedback.

b) People who make their living in electronics usually describe things one way when talking about a single stage versus a complete circuit. For example if you read a textbook describing (say) a single common-source FET with an unbypassed source resistor, they will often say that it uses "local feedback" or "degeneration".

But already we have stepped in the doggy-do.

Many people talk about a complementary feedback pair (CFP) as having "local feedback" because the loop is short (only around two devices). So is "local feedback" "local" because it is applied to only one stage or because it is only applied around two stages? And what about if you send feedback from the drain to the gate of a single device? (This is much more commonly done with tubes than transistors.) Is that "local feedback"?

c) When talking about a complete circuit that has only unbypassed source (emitter, cathode) resistors and/or followers, common practice by people in the business is to call that a circuit with "no feedback".

So now we are stepping into more doggy-do. In one application it is accepted practice to say that a circuit element has "local feedback" but in slightly different context, it is accepted practice to say that there is "no feedback" with essentially the same circuit.

So already you can see that there is plenty of room for mischief and disagreement. So let's go ahead and step into one more piece of doggy-do. Returning to the single (say) common-source FET with NO source resistor, most textbooks would say that this circuit has "no local feedback". But this is clearly incorrect. There is an internal Rs that is equal to the reciprocal of the transconductance. Clearly this internal Rs creates "local feedback" in the same way as an external unbypassed source resistor does. (If it didn't, the circuit would have infinite gain.) Now we cannot escape the doggy-do. Because in this sense, *every* circuit ever made has feedback. But if *every* circuit has feedback, then the term loses all meaning completely. It would be like saying there is no such thing as an insulator because all insulators conduct *some* current, or that there is no such thing as a conductor, as all "conductors" have *some* resistance.

For the purposes of a complete circuit, I prefer to use the commonly accepted terminology of professionals. If the circuit only contains unbypassed source resistors and/or followers, I call this a "no feedback" circuit. It could also be called a "no loop feedback" circuit or a "zero feedback circuit". Personally, I prefer the term "zero feedback" as it is literally impossible to build a circuit with any less feedback than this. Also, it operates differently than other traditional circuits, as at *no* point is the signal fed from a downstream point to a point further back upstream. (I have found that the larger the loop encompassed by the feedback loop, the more deleterious the sonic effect.)

But anytime there are qualifiers attached to the term, alarm bells go off for me. When I see the term "no global feedback", this is sign of an essentially meaningless claim. For example, take the Marantz 9 tube amp. It is a traditional tube amp circuit with feedback taken from the transformer secondary back to the cathode of the input tube, EXCEPT that they add a cathode follower in front of the input tube. That cathode follower is *not* inside the feedback loop, so they could legitimately claim "no global feedback", but it would be extremely misleading to do so. Another example are the Boulder power amps. They are built by putting two (discrete) op-amps in series. Of course, there is a ton of feedback around each op-amp, but no feedback from the second op-amp back to the first op-amp. So they could also claim "no global feedback", but again it would be exceedingly misleading to do so.

OK, you asked for it. Let the semantics wars begin...
 
Status
Not open for further replies.