mos noise

We know the channel noise of MOSFET can be expressed as 4kTγgd0(γ=2/3 for long channel). When it comes to deep sub-micro technology,its value is much higher than 2/3.
The question is when I simulate the LNA noise figure in spectre RF, according to the simulated results, I found γ is much closer with 2/3. I wonder how the simulator calculates this parameter. and whether we could trust this result.
Thanks

It depends on the technology model file that you are using. If you are using deep submicron technology the value of gamma is close to 4/3.
and for long channel devices as you mentioned it is equal to 2/3.

I use TSMC0.13um technology,but it seems the spectre RF underestimates the noise figure,i.e. it underestimates the value of γ...