The 90 did worked so i use the zero degres file and so positionned the mic .... I know not optimal but quite enougth for testing and appreciating how dirac is great !!!

Basically if you have a 0 degree file and use the mic in a 90 degree position what you'll have issues with is the high end. A guy over at the MiniDSP forums did some testing and calculation and found out the general transfer function for mics line that. You can see it here (3rd graph) the net effect is it starts rolling off at around 2kHz. What that means is your high frequency sounds will be emphasized more than they should, and you aren't going to get the focus and sound stage that you would with an accurate calibration.
 
Now if you used a 0 degree calibration and then set the mic at zero degrees (meaning pointing straight at the speaker) it'll work great. You do need to move the mic for each sample though, you have to point it at the left speaker, take the sample, then point it at the right speaker, and take the sample.

Yep it was zero degre file and i pointed the center ...not a specific speaker ...it did n t allow me for on point of measurement to get 3 measurement ....as you desçribe

At the center it works OK, but then you need a 45 degree calibration file. Best to just look in to getting a mic with a 90 degree file in the long run, or having Cross Spectrum Labs calibrate your mic.