How do you know if you need to add a sub (without auditioning one, I mean)?

First of all, a lot of bass in some music is entirely unrealistic.  I've been with many people listening to recorded music where some said bass was great and others including myself said the bass was over emphasized.

Way too much.

So the best way was suggested before, get a test CD or album that plays pink noise or frequencies from zero to 20 kHz and a pretty good DB meter.  have the meter at the listener's level and measure the readings over the entire frequency range.

See where the gaps, valley, peaks are.  This also helps tell you the room effects and helps solve room issues.

if your system's bass response is what it is suppose to be, IOW, is it flat?, then no, you do not need subs at all.  your system and speakers are doing their job.

If it isn't flat, that does not automatically mean you need subs, it may be the room.  Or a combination of room and system.  

It is definitely not a simple as "I need subs".

A test CD and a sound level meter are not expensive.

enjoy

An interesting test is to do the measurements in a system that has subs with the test CD and sound level meter and see if it shows too much bass.

enjoy

Minorl: I like your idea of testing DB at all the frequencies. Aside from my wondering about subs, I'm sure I have room problems I don't know about since I have almost no points of reference as to how rooms/systems without problems sound.
It's certainly possible (and maybe probable) that I don't need anything more than what I've got, given that I can maximize/improve without adding devices or components.

By the way, I'm pleased that so many posters are really trying to help me vs promote other agendas or make me feel like a dumbass.

Do the measuring with a sound meter app and white noise to see if/what you are missing or not. Only then will you know for sure. Having that lowest octave reasonably flat in response along with all the rest makes a significant listening difference ranging from huge on many recordings to minor or even none on others.

Room acoustics is a different issue and greatly impacts bass. Not a bad idea to get a handle on that first with what you have before adding anything.

So the best way was suggested before, get a test CD or album that plays pink noise or frequencies from zero to 20 kHz and a pretty good DB meter. have the meter at the listener’s level and measure the readings over the entire frequency range.

I do not want to be rude "the best way" and then suggesting something that were the best way at the 80-ties. 🤔

Maybe a computer with free software like REW instead of a CD. And a inexpensive AND calibrated measurement microphone ~70-100$ much better than any DB meter..

But a dB meter is better than nothing and it is almost as nothing also in comparison. But I understand that not everyone are handy with computers and software. (No pain no gain.)

You will get graphs with levels and frequency at any point you desire in your room and move your speakers and see exactly what happens for each change you make.

You also get so much more for example can get help to find crossover point and delays for your subwoofers to time align them with your mains automatically generated.

You can do multipoint measurements and also do moving microphone measurements and you can see decay times for across the whole frequency band to see what your treatment do for you... And so much more that you can’t do at all with a DB meter.. so much more so it is not even funny. 😉💖

And remember our hearing is NOT the same as a flat graph.✋ Even if we would like it to be.. it is convenient to think so. This is regarding to get something flat.. it LOOKS nice with a flat line.. anyway microphones measure accurately.. let me explain below.

Study this graphs that show how our hearing works across the frequency band. Note that they are far from horizontal or linear/flat for that matter..
https://en.m.wikipedia.org/wiki/File:Lindos1.svg

If we measure with a microphone the same shape as the graph shows sound coming from a speaker. Then we would experience that speaker as truly flat. With equal loudness from 20 to 20000 Hz.

So as you see in the graph, if we follow the 60 line that is 60 dB at 1 kHz so for us humans we perceive that at 20 Hz at 107 dB is AS LOUD as 60 dB @1 kHz.
Because our hearing is less sensitive at 20 Hz.👂

In other words if you strive for as a goal to make your frequency response flat so you have 60 dB also at 20 Hz then you precive when you listo, that you have attenuated the level at 20 Hz by 47 dB!

And that in it self make us to not be able to hear ANYTHING at all at 60 dB @20Hz! 😱
When you can clearly see that 60 dB @20 Hz is BELOW your hearing threshold (the other read line named the same).

I hope this is food for thoughts. 🤔😍🌷❤️

(This is just science worked on since 1933 and then reassessment and revised in 1956, which became the basis for an ISO 226 standard later on.)