Yeah it's about using the maximum ffb available.
Most Logitech users (me included) use 100% all the time and when they upgrade to a better wheel, they want more force.
I've found that my feel for "real life forces" is somewhere between the TS-PC and the Fanatec CSW 2.5.
A friend of mine wanted to upgrade from his Logitech DFGT to a T300 but sent it back due to the fan noise and the belt noise while steering.
I tried it too and definitely would've went for 100% ffb strength.
With the TS-PC I've found that 100% at the base and about 60-70% gain in the games felt good.
With the CSL Elite of a friend, I've found the same as with the TS-PC I've had for 2 weeks:
100% and 60-70% gain in the games.
Now with my CSW 2.5, I'm often running a bit less.
Mostly 80% in the base and 50-60% gain in the games.
CSW 2.5 has 8Nm.
80% base = 6.4 Nm
60% game gain = let's say this equals to 60% of 6.4 Nm for average steering forces.
So 3.84 Nm constantly at the hands while driving.
Now let's just use these values:
G27 = 2.5 Nm maximum (so definitely 100% all the time...)
T300 = similar to the TX = 3.9 Nm.
So the T300 at 100% base strength and 100% game gain is about what I'm experiencing constantly with my CSW 2.5.
But with the CSW, I have peaks for kerbs/details/working aerodynamics at higher speeds that go up to 6.4 Nm.
The TS-PC seems to be a bit stronger than the T500 so probably around 6 Nm, which is a little bit less than what I'm using with my CSW 2.5.
3 of my online racing buddies have a direct drive wheel. OSW small mige and simucube 2.
They are both using 8-10 Nm as base strength.
So a little bit more than I'm using, but they are using lower game gain.
So probably the same average force as I'm using but more peak dynamic.
jfyi: I did a driver training with my road car on tarmac (and rain + snow simulation surfaces). It's a 1995 MB C-Class and the forces when cornering at the limit felt pretty similar to my CSW 2.5 settings.