After all the critics these guys received last time I guess it needed to show that they fixed the mistakes.
I agree with the maths now... my first estiamte give me a ~7.3m for Xbone but I changed to 7.5m because my bias and the possible chance to 360 have sold less than 700k in last quarter and 50/50 in the previous one.
I guess 7.35m estimate close to real situation.