What's going on
With negative feedback, an op-amp adjusts its output until its two inputs are at the same voltage. That makes the gain depend only on the resistors. An inverting amplifier has a gain of −Rf ÷ Rin, and a non-inverting amplifier has a gain of 1 + Rf ÷ Rin.
The output can't go beyond the supply rails. A rail-to-rail op-amp gets close to them, while older designs stop a volt or two short. Ask for more and the tops of the wave are cut off, which is called clipping.
Every op-amp also has a gain-bandwidth product (GBW). The higher the gain you ask for, the lower the frequency at which the amplifier runs out of speed. The bandwidth is GBW divided by the noise gain, 1 + Rf ÷ Rin, for both circuits. Our op-amps guide explains these ideas in more depth.
How this simulation works
The op-amp is modelled as a single-pole amplifier with the GBW shown, plus hard clipping at the output limit. The "classic" option stops 1.5 V short of each rail. It doesn't model slew rate, input offset, noise or the extra phase shift of a real op-amp, so large fast signals in a real circuit can look worse than shown here. Check your op-amp's datasheet for its real output swing and slew rate.