And that's the clincher. op amps can do everything in a preamp bipolars can, yet in the end, the bipolars end up sounding just that little bit better.
An op amp is what it is, you can do little but adjust yourself to their capabilities, whereas in a discrete design, you sure have to work a lot harder, but you keep the freedom of adjusting your design much more felxibly simply because you control each and ever little bit of the ciruit. Thus you can choose the best part for the job, including the often unmentioned passive parts. The price you pay is in increased work, more parts and more real estate on the PCB, all of which can be translated into a higher price.
And, of course, you can mix-'n-match. People use low noise discrete FETs cascoded with low noise BJTs, and have them driving an op amp for the required high gain in a RIAA phono board.
An op amp is what it is, you can do little but adjust yourself to their capabilities, whereas in a discrete design, you sure have to work a lot harder, but you keep the freedom of adjusting your design much more felxibly simply because you control each and ever little bit of the ciruit. Thus you can choose the best part for the job, including the often unmentioned passive parts. The price you pay is in increased work, more parts and more real estate on the PCB, all of which can be translated into a higher price.
And, of course, you can mix-'n-match. People use low noise discrete FETs cascoded with low noise BJTs, and have them driving an op amp for the required high gain in a RIAA phono board.
maybe we're having a translation effect here. An op-amp is a topology, generally its that big a** triangle with + and - inputs. It can be either discrete or integrated. Most of the amps talked about here are op-amps, both integrated and discrete. Remember the original op-amps were tube!
Cheers
Alan
Cheers
Alan
So, you suppose that those who think there is a difference are or liars, or crazy ?
(A very respectful attitude ?)
(On my side, I have no conclusion while I am very happy with some linear ICs.)
Last edited:
......and not very good. For all practical purposes the best opamps by far are ICs, for very good reasons. This is why it's taken as read that opamps are ICs: though one can implement them discretely, there's seldom any point and plenty of good reasons not to.
No, I'm confident there are plenty very, very, sincere and otherwise sane people that are rife with biases/etc. I certainly am one of them (well, ask the right people and both my sincerity and sanity might come into question 
).
At the same time, if we assume both the discrete and IC forms are well engineered (or misbehave very similarly), I'd bet you get a null in any sort of ABX.
Welcome to being human.
).At the same time, if we assume both the discrete and IC forms are well engineered (or misbehave very similarly), I'd bet you get a null in any sort of ABX.
Welcome to being human.
I have one of those on my desk, plugged into an octal Amphenol socket adaptor to keep it upright.
Heathkit also made some knockoffs for their analog computer.
How about betting on a null between a discrete on 2-side clad board, and the same discrete on a 4-layer board ?
(both AP optimised, NO half inch thick alloy case)
- Status
- Not open for further replies.