Just finished up GMATPrep 2. I took it with all the real testing conditions, and I didn't recognize any of the questions, so I'm guessing it's pretty accurate. The one problem I had with it was that I completed it in an absurdly cold room, and I found the temperature a bit distracting. Lesson learned -- I'll bring a sweater to the real testing center.
Here was the result:
GMATPrep 2 = 770 - Q50 V45
I was a bit disappointed to see my verbal score drop down to a 45 (I missed 3 questions), but I was glad to see my quant score stay up at 50 (I missed 7 questions). I have a feeling that none of the verbal questions I have answered incorrectly during my three GMATPrep attempts (1, 1-retake, and 2) have been experimental, because all of the questions have docked me serious verbal points. I wonder how GMATPrep deals with experimental questions -- does the problem even have them? I'd be willing to bet that this could be a reason why most people score higher on the real GMAT than on GMATPrep, a trend that I have noticed after reading the last 5-10 pages of the "Share Your GMAT Experience" forum.
My test is Thursday evening, and at this point I think I'm ready to take it. I'm definitely getting used to taking one GMAT exam every evening, so I think that the real thing will just feel like another GMATPrep exam.
Here are all of my scores thus far (in chronological order):
Princeton Review 1 = 710 - Q47 V41
Princeton Review 2 = 750 - Q51 V43
Princeton Review 3 = 690 - Q43 V43
Princeton Review 4 = 730 - Q48 V44
MGMAT 1 = 750 - Q48 V45
MGMAT 2 = 780 - Q51 V45
MGMAT 3 = xxx - Q51
MGMAT 4 = 730 - Q51 V39
GMAT Prep 1 = 780 - Q50 V48 (with AWA)
PowerPrep 1 = 780 - Q50 V47
GMAT Prep 1 = 780 - Q50 V48 (retake)
GMAT Prep 2 = 770 - Q50 V45 (with AWA)
I seem to have plateaued around the 770-780 range, so I'm hoping for that type of result on the real thing.
I'm going to go through the GMATPrep SC document to make sure my verbal score is as high as possible.