Ori
Member of the Trade: Oritek Audio
- Joined
- Oct 23, 2005
- Posts
- 674
- Likes
- 11
Quote:
Yes, in the DC servo loop. Not in the signal path, which is a zero-feedback circuit.
Quote:
As a "blanket statement", that assumption is wrong. However, you have to ask what's the best achievable with both circuits and in that regard I full agree with marroyo's comments.
Most opamps tie the designer's hands in a sense that he must reduce their inherent high gain, usually by means of a negative feedback. We can argue forever about the virtues of NFB. It does reduce distortion and equalizes circuits and in general it is an appropriate device to use in high volume consumer electronics. It is be analogous to a Honda 4 cylinder engine: well-behaved, gives you excellent gas milage, but also a tad boring for most people. One can say that's all you need...
On the other hand you have a "wild" discrete circuit that can be designed "on the edge". It allows the designer to use stages with less gain and therefore avoid or reduce NFB. It also allows the designer to skip buffer stages, cut down on power regulation based on the quality of his supply, combine functions in "tricky" circuits, etc.
This "hardly tamed" circuit is more analogous to an Alfa Romeo engine (I'm tired of Ferraris...) in a sense that it works, but is not as smooth as the Honda... Sometimes it might sputter, so you would expect in cases more issues (mostly when not designed right...) but it does give you a better performance. It's hot-rodded to do that, not to drive you to the grocery store in an uneventful manner.
Then you have the discrete circuit that is well-designed and well-behaved. That's a Honda S-2000...
If the NFB concept had only virtues and no drawbacks then I'd be the first to advocate it.
[size=large]In my subjective judgement[/size] negative feedback has some drawbacks which translate to (for me) slight sonic anomalies, therefore IMO (!) the NFB circuit with opamps is less than ideal for typical line level applications.
Other people may not agree with me on this issue and that's fine with me. My ears are what they are as are my sonic preferences, so I choose a certain direction with my circuits.
I use opamps where I feel there are no such issues, so I do not look at the whole topic as a heresy issue. To me, using a particular component or circuit is an engineering decision, based on my subjective goals for optimization. That criteria is not different than an engineer at Sony using opamps. The only difference is the set of goals.
A consumer should be interested in the final result and for that matter there is no option but to listen to more equipment and judge for yourself. I don't mean listen for 5 minutes but try to listen at length, preferably at home with familar tracks.
I do that all the time. I cannot dismiss competent opamp-based products just due to the use of opamps. Fact is that some of them sound very good indeed. However - in my judgement - they could be even better and I can tell you that in many cases people do not agree with me. It doesn't make me right or the other way around. That is my taste and my preferences.
Which brings us to the conclusion... Buyers of (hopefuly) quality audio equipment would assume that the designer has some consistency in the "voicing" of his products. Not because they are tuned to sound artificially this way or that way but because the designer listens to the gear and tries to optimize its sonic "signature" to what he believes is the "right sound".
If such consistency exists (and I would hope so!) then you are really buying not just a functional product but one that is tailored to a specific taste. If that taste happens to match your own sonic preferences and the product is reliable and presentable then you have found a good option for you.
If it doesn't - then who cares if it's made of opamps or transistors or directly-heated triodes?! These are just components. One shouldn't be judging musicality based on basic electrical components...
| Originally Posted by athenaesword Does my omz have opamps at all? |
Yes, in the DC servo loop. Not in the signal path, which is a zero-feedback circuit.
Quote:
| Originally Posted by StanleyB1 One would be easily fooled by such talk that discrete is better than opamp |
As a "blanket statement", that assumption is wrong. However, you have to ask what's the best achievable with both circuits and in that regard I full agree with marroyo's comments.
Most opamps tie the designer's hands in a sense that he must reduce their inherent high gain, usually by means of a negative feedback. We can argue forever about the virtues of NFB. It does reduce distortion and equalizes circuits and in general it is an appropriate device to use in high volume consumer electronics. It is be analogous to a Honda 4 cylinder engine: well-behaved, gives you excellent gas milage, but also a tad boring for most people. One can say that's all you need...
On the other hand you have a "wild" discrete circuit that can be designed "on the edge". It allows the designer to use stages with less gain and therefore avoid or reduce NFB. It also allows the designer to skip buffer stages, cut down on power regulation based on the quality of his supply, combine functions in "tricky" circuits, etc.
This "hardly tamed" circuit is more analogous to an Alfa Romeo engine (I'm tired of Ferraris...) in a sense that it works, but is not as smooth as the Honda... Sometimes it might sputter, so you would expect in cases more issues (mostly when not designed right...) but it does give you a better performance. It's hot-rodded to do that, not to drive you to the grocery store in an uneventful manner.
Then you have the discrete circuit that is well-designed and well-behaved. That's a Honda S-2000...
If the NFB concept had only virtues and no drawbacks then I'd be the first to advocate it.
[size=large]In my subjective judgement[/size] negative feedback has some drawbacks which translate to (for me) slight sonic anomalies, therefore IMO (!) the NFB circuit with opamps is less than ideal for typical line level applications.
Other people may not agree with me on this issue and that's fine with me. My ears are what they are as are my sonic preferences, so I choose a certain direction with my circuits.
I use opamps where I feel there are no such issues, so I do not look at the whole topic as a heresy issue. To me, using a particular component or circuit is an engineering decision, based on my subjective goals for optimization. That criteria is not different than an engineer at Sony using opamps. The only difference is the set of goals.
A consumer should be interested in the final result and for that matter there is no option but to listen to more equipment and judge for yourself. I don't mean listen for 5 minutes but try to listen at length, preferably at home with familar tracks.
I do that all the time. I cannot dismiss competent opamp-based products just due to the use of opamps. Fact is that some of them sound very good indeed. However - in my judgement - they could be even better and I can tell you that in many cases people do not agree with me. It doesn't make me right or the other way around. That is my taste and my preferences.
Which brings us to the conclusion... Buyers of (hopefuly) quality audio equipment would assume that the designer has some consistency in the "voicing" of his products. Not because they are tuned to sound artificially this way or that way but because the designer listens to the gear and tries to optimize its sonic "signature" to what he believes is the "right sound".
If such consistency exists (and I would hope so!) then you are really buying not just a functional product but one that is tailored to a specific taste. If that taste happens to match your own sonic preferences and the product is reliable and presentable then you have found a good option for you.
If it doesn't - then who cares if it's made of opamps or transistors or directly-heated triodes?! These are just components. One shouldn't be judging musicality based on basic electrical components...
I like food analogies!
