The “no improvement” aspect of this would seem like a mis-perception that obscures the answer to the question. Of the 30K product listings mentioned, only 19K or so of them are active and if one further divides that by ten to account for ancillary considerations such as package, packaging, etc. which don’t usually figure strongly figure in the design selection process,there’s only a couple thousand products left that represent the cumulative solutions to (say) a 10-dimensional optimization problem rendered over the last ~60 years. Compared to nearly a million active SMT resistor products or half-million MLCCs, one might remark that the op amp solution set seems comparatively sparse.
The moment one allows for application-dependent selection criteria, the notion of a “best” pretty much goes out the window. The discussion AP references stems from an observation shared with me by an analog Graybeard maybe 15 years ago, that past the gatekeeping considerations of supply compatibility, there’s usually only one or two dominant factors that end up controlling the remainder of the selection process, even though the solution space may subtend 10 to 20 possible dimensions of variation.
I’d respectfully vote in the negative.
This is false, or at least incomplete. The reason being that a significant number of single-supply op amp applications involve amplifying small signals near ground potential, in which cases it is the common-mode input range that limits suitability for purpose. Accepting the guidance above would result in exclusion of products such as the TLV34xx, which might well serve such needs when configured for a gain of about 1.5 or more.
To the best of my knowledge, there’s no formal definition of the “jellybean” concept by the IEC, JEDEC, DIN, or similar standards body, but if it’s understood to mean parts that are old enough to have become multi-source commodities, then the ask for one that operates down to 1.5v pretty much forces a null result because that simply wasn’t a thing until quite recently.
Rather than alerting to the inherent flaw in the query, the machine engaged a polite argument over the definition of a subjective colloquialism. While that may offer some cheap entertainment, simply using the available parametric tools to stipulate a few of the other criteria mentioned would very rapidly give a user a clue that there’s some trouble with the ask.
Some editorial points:
- I’m seeing two “Example 3” and no “Example 2.”
I would respectfully suggest that this be cast as an “OR”, rather than an “AND”. This post (Also mentioned as related info) offers some explanation. It’s dated, but remains largely relevant. Observers are directed to more authoritative sources for a fuller explanation.