You get the point, but gaming won't get that big again. It may have dropped like a rock but it still holds #1 of last gen, don't forget. Mobile gaming isn't only hurting Nintendo you know. It won't reach past 100M. I'm just saying, big profit, mass appeal, and tons of multi million sellers.

you don't think PS4 will hit 100M??? I think it will and with ease, espeically considering how well it's doing compared to PS3. 

I have had access to Mercury Research Group pricing reports, and cannot post them here (nothing since late summer anyway due to my business structure), but can tell you this :

GDDR5 prices plummeted after the *coin mining rushes failed. They were already falling anyway. This is why you now see many 4GB GPUs slotting in at prices previously reserved for 2GB and 3GB GPUs. Also due to the mining craze, capacity for GDDR5 production increased massively from previous years. So the production facilities have to either be idled, or produce more for less $. I saw prices fall from ~$12/GB to ~$7/GB to ~$4/GB over a relatively short period of time.

This is the usual cycle with memory pricing anyway. Here are the stages relevant to your questions :

Stage 0 : Engineering samples are made after the design is more or less finalized, and sent to OEMs to test with their PCBs or other hardware on their ends. Basically a compatibility/testing/quality/performance measurement stage. VERY high price to get this done.

Stage 1 : First production for release in the market, the 'bleeding edge'era. DDR4 is still roughly in this stage. Production numbers are low, implementation is low, and price/GB remains high for awhile.

Stage 2 : Early mass production. Some major market actors have moved this memory into mass market items, and a lot is needed out there, so production increases. Price/GB is hugely lower than stage 1, but still moderately high. Last-gen memory still being produced in decent volume.

Stage 3 : Mainstread mass production. This memory type is now the de-facto standard for whatever, and pricing is incrediblycompetitive. There may be some price fixing on this due to just how cheap it gets. Cheapest stage for Price/GB. Think last year when you saw sometimes 8GB DDR3 kits for $30-$35 retail.

Stage 4 : Downhill of mass production. Combination of low margins and lowering demand overlaps when OEMs are moving some of their demand to a different standard, which results in manufacturers retooling for the new memory in some cases, idling production, looking at stockpiled inventory, and trying to cut a good balance. This is the current stage of DDR3. Pricing increases steadily.

Stage 5 : Basic EOL of mass production. Due to the stage 3/4 timeframes, enough manufacturers have achieved large inventories of these ram chips, which can be put onto various PCBs with ease, but demand for new memory types along with massive dropoff in demand for the old means production is largely shuttered on this kind (for purposes of this example at least). Pricing increases again fairly significantly.

So that is it. With DDR3, it's already passed the time that it is at its cheapest, and is now on track to steadily increase over time. DDR4 is rising steadily in high-end areas, but will truly make the leap to mass market next year, which will cause DDR4 prices to plummet, and DDR3 prices to rise sharply.

If you want an idea of how it ends, look at this (just purely using consumer-available parts as an example) : 

4GB DDR2-667 $119.99 http://www.crucial.com/usa/en/memory/ct51264aa667

4GB DDR3-1600 $39.99 http://www.crucial.com/usa/en/memory/ct51264ba160b

4GB DDR4-2133 $57.99 http://www.crucial.com/usa/en/memory-ddr4/ct4g4dfs8213

Over the next 6 months, watch the prices on DDR3 and DDR4 start to overlap and move in different directions. DDR3 will eventually be as insanely priced as that DDR2 stick up there, and DDR4 will become very cheap, then start ramping up all over again as the cycle repeats.

GDDR5 is a bit different than all of this. It's more specialized, but also lives in a similar cycle. Pricing is still falling on GDDR5, but eventually this will rise as well when it's no longer mainstream.

I think they are showing some real restraint and foresight.

The $399 diskinect SKU in June started out big, then trailed off to the same ratio as before in a few days to weeks. We clearly saw sales fall off a cliff in May due to the announcement, so you had evidence of people holding off buying their X1s, then there was a big slowdown after the initial rush (July/August were terrible for X1), which showed strong evidence that the deal might have convinced many to simply move their purchases up to June instead of waiting.

This $329 and special holiday bundles / gift cards / etc happened, and the effect may end up being similar to even worse : a lot of people might have been waiting for the rumored big holiday specials from Microsoft instead of spending $399 with no game on one, so you had a bubble of holdouts, and likewise, people who might have otherwise waited until December or even beyond might have been convinced to  go ahead and jump in while the mega specials were available.

There are many ways it can go from here. Microsoft might want to remain silent for now if they truly ARE going  to go back to $399 in January, with no game to boot. That's worst-case scenario, and would mean a rapid return to around 2:1 PS4 advantage month to month, perhaps even 2.5:1.

I think even at $349, a bare X1 will fall behind PS4 again easily for at least the 1st half of 2015, but remain very competitive (say stick between 5%-15% per month on average)

At $329 and bundled AAA games, they could stay ahead all the way up until Sony decided to lower the price of the PS4.

Anyway, reasons for MS being pretty quiet :

(A)- They are going to return the prices to normal, because losing a huge amount of $ per console sold is simply bad business. It's particularly bad if they're getting a lot of '2nd console' buyers who have zero interest in XBLG subs, and who are really only getting the system for exclusives. With the typical multiplat favoring PS4, multi-console owners are not a good 'get'for Microsoft.

(B)- They want to look past the initial hysteria of the $329 crazy firesale sales rates to see what their projected baselines and competitive stature end up being. If they walk into 2015 with both November and December under their belts, and have a solid plan at keeping momentum (again, $349 is not enough imho), then they can start having more confident PR reports.

Think how dumb they would look if they went all-out trumpeting November NPD and/or December NPD, only to get killed 2:1 in the following months.

It has a high chance of hitting 100M but not 150M... that's never happening again.