Extrapolating Diablo III console sales

Back in February during their fiscal 4th quarters earnings call, they announced they had sold 12 million copies of Diablo III (this was of course before the console versions went on sale).
Note that 10 million of those copies were sold in the period of May 2012-July 2012 (Mike Morhaime also noted that this exceeded their forecast for the entire year which means that Diablo III afforded them a large profit windfall – this is actually kind of similar to when they underestimated WoW subscriptions back in 2004 where that original monthly fee they had was based on the expenses they would incur on those underestimated numbers; that huge demand provided them a huge windfall even after they had to stop sales for a bit in order to upgrade their backend infrastructure; WoW subscriptions has been a huge gravy train for them ever since and that doesn’t even include the microtransaction angle; thus even though they continue to lose subscriptions, they have a huge cushion before the profitability of that franchise becomes an issue).
During their Q1 2013 (covers the 3 month period ending March 31) conference call, they announced their #1 selling PC game was HoTS at 1.1 million copies sold (this of course puts D3 sales somewhere below that mark).  
Thus between then and now, they’ve sold the balance of that approximately 2 million between desktop and console versions.
Total unit sales wise, Diablo III has been a huge success for Activision Blizzard, Inc.  But all one has to look at is where the bulk of those desktop version sales occurred to get a pretty good picture of how much of those numbers were based on sheer pent up demand versus the continuing sell through.  Basically, a huge number of pre-order sales of Diablo III prior to its May 2012 release and subsequent channel sell through during the first three months was due to a decade plus of goodwill legacy which the previous two titles in the franchise, had managed to garner.
From August 2012 to the end of the year, they managed to sell another 2 million.  That is still a spectacular amount of unit sales by the way even though the trend is a declining one.  Furthermore, the declining unit sales for 2013 isn’t surprising for a game that is over an year old.  With that said though, it isn’t an exaggeration to say that the poor state of how the game ended up shipping (and the time it took to fix some of the initial problems; plus the current wait for the pre-expansion system level changes they are working on), has eroded that goodwill legacy.
What is telling is the combined unit numbers of console and desktop combined (slightly under 2 million split between the two) which one can extrapolate as to how poor the console numbers actually were.  That has to be taken into the context of the amount of development and engineering time they spent on getting the game onto two separate console platforms, let alone the marketing.  Furthermore, consoles are usually a separate target demographic but it is clear that part of those sales were taken up by those who also owned the desktop versions.  The key is to gain new customers in that other demographic which further puts into perspective, how the console version, hasn’t been that great in accomplishing that objective.
With all this said though, the desktop version of Reaper of Souls will still sell several million copies once it goes on sale.  A good metric for success that I would personally consider is 5 million in the first three months based on half of the initial sales that occurred during the first three months (according to the 1 year infographic that Blizzard released, 2.1 million players still logged in daily with the peak being 5.8 million: http://us.battle.net/d3/en/blog/9691895/).  Anything less than that though would be a good indicator as to just how much goodwill has been lost with the franchise, and will be a truer reflection of the actual active population of the player base.