I did not notice any deviation from the standard quant section. What really bothers me is my insanely fluctuating verbal score: on
mgmat I scored 35-39, veritas 32-36, gmat prep 35-37. On this pilot I got verbal 23 - this is abnormal and is not consistent with my verbal ability across the tests and previous gmats. My verbal just cannot flucatuate that wildely!
This makes me to conclude that verbal algorithm is different from quant algortihm (which actually rated me quite consistently from q40 to q43 and q 44 this last pilot exam) on gmat - in a manner that reconfirms the gmat prep experiments taken by guys here on the forum where they answered first 10 questions incorrect and with the rest q correct they got verbal 23. This is only explanatation that comes to my mind.
So imho first 10 verbal questions weigh more for a good verbal score than first 10 quant q do for the quant one. Something is just really inconsistent about the gmat verbal algo that gives me consistently 36-37 and then dumps to 23: from above 80% down to 30%.
Wheras the quant gmat algo is really consistent and shows the quant ability. Overall I am realizing now that tthe order of sections is not a great factor. GMAT measures your overall ability, and does it quite fairly. At least for the quant section.