I simply meant to comment upon the differences between the 2 differing topologies/methodologies (not to imply that 3pin ic's don't have opamps in them)....as I have found the discrete style of Mr.Jung's favorable within the conditions and applications i have employed them, in comparison to the more "traditional" 3pin ic.
And of course I never meant to imply that the use of said 3pin ic's necessitate a noisy supply, simply that i have found otherwise to have less noise in comparison (again in this type of application, and depending upon the implementation the opposite could occur)....as for example the Drefahl supply is quite low in noise.
Absolutely, appropriate filtering of any ac-dc conversion-regulation circuit is paramount to well performing power supplies, regardless of regulation topology and should span an entire circuit consideration as well...(this is also something i feel has been sometimes executed to a lowest common denominator)
...my apologies for not being more specific and possibly adding confusion.
What about a bi-polar discrete shunt...maybe Bruno Putzeys could chime in...I believe he might use this methodology in his Grimm audio tube power supply.