CDO

Hi all
This is Suman from IIT Kharagpur, India. I am regular user of cdo. Recently I am doing eof analysis using cdo. I am facing a serious problem. It is as follows:

I have monthly precipitation data (rainfall.nc) for Indian landmass during 1951-1980 (30 years=12*30=360 months). I have used that datasets for eof analysis using cdo with the following code:

cdo sub rainfall.nc -timmean rainfall.nc anomaly.nc
export MAX_JACOBI_ITER=100
cdo eoftime,3 anomaly.nc eig.nc eof.nc

Now when I am plotting eof1(1st major variability pattern)from eof.nc, it seems to me that it is showing simply the pattern of the climatological mean picture. I couldn't understand whether I am doing it correctly or not.
Please reply me whether the steps I am following correct or not.

I have also seen that with the same anomaly file, computation of eof with cdo and matlab shows two different results (patterns are similar but magnitude is different). Explained variance is also different.Why it is so?

Any kind of Help is highly appreciated.