# Multisim vs LTspice : Different Result

I was using the simple darlington circuit below as the test circuit. DC operating point analysis was run on both LTspice and Multisim, with V2 as the manipulating variable and the base voltage of Q2 as the responding variable.

After the experiment, I found out that the result were different for Multisim and LTSpice when V2 is near to zero voltage. Noted that I had modified the BJT model in Multisim to match that in LTspice.

Belows are the result of Multisim and LTspice when V2 was set to 0V and 2V respectively. For 2V, the result matched but for 0V, the result showed huge difference. Any idea causing this issue?

I have also attached the convergence/accuracy parameter lists for LTspice and Multisim.

V2 = 0V

V2 = 2V

Paramater

LTSpice

Multisim

Which one is more accurate? Is there anything I miss?

The voltage difference is actually rather inconsequential in practice because it's like two leakage currents fighting. Even 1 or 2 G$$\\Omega\$$ input impedance on the virtual meter could cause that amount of voltage difference.