Hey Kbrad, you are right that jlgrimes was technically incorrect but they were talking casually, and in a general sense they are in the right ballpark; running at a higher sample rate and oversampling both reduce aliasing for nonlinear processing of the same input signal.
You are also correct that some plugins might react differently at different sample rates. But that's a design and coding issue specific to each plugin, not a general principle.
However, I think you have misunderstood the actual difference between oversampling and higher sample rates.
For reducing aliasing, oversampling is equal to, and usually better than, high sample rates, in real-world usage.
Why?
The nonlinear process itself will be mathematically identical whether using 2x sample rate or 2x oversampling. However, an oversampled plugin will then downsample again, keeping the audible frequencies. On the other hand, working at higher sample rates can leave the ultrasonic content in. This can cause undesirable artifacts if you go into a long chain of effects, so you will have to manually add a low pass filter after each plugin. With oversampling you do not have any unwanted ultrasonic content going in to the next plugin in the chain.
There is another advantage; Oversampling can be customized and targeted. It will use less CPU than a higher sample rate for the whole project, and if you have a very aggressive process (heavy saturation for example) then you can set just one plugin to x8 oversampling, which would be impossible to do for the whole project, and get an even bigger reduction of aliasing. The only real disadvantage is higher latency. So I recommend oversampling for post-processing, not for recording.
tl,dr: Oversampling is good, because plugin designers know what they are doing. (most of them)
And as usual: in the real world, you very quickly hit the limit of what you can actually reproduce on speakers or headphones or hear with human ears. 2x and sometimes 4x oversampling in a 48khz project is enough bang-for-the-buck in most cases.
kbard wrote: ↑13 Aug 2021
jlgrimes wrote: ↑12 Aug 2021
Going to a high sample rate basically forces all plug-ins to oversample, which can produce a cleaner sound in most cases.
No it doesn't forces them to do that. That is technically incorrect way of explaining things. If you increase sample rate in your DAW then the plugin processes at an increased (higher) sample rates. As simple as that. End result is less aliasing (in some cases).
"Oversampling" is a term used as when a plugin is working at a given rate but in it's internal engine plugin can work at a higher sample rates (hence the oversample). Which is not the same as using absolutely same plugin at a higher sample rates.