Good. These companies have been fixing prices for years, making the market attractive to countries like China who see an opportunity for $$$. And when you export your production to countries like Thailand and China, dont be surprised when your tech is stolen.
At the very least, this will force these companies to drop their price fixing garbage.
When the companies involved export their production to insecure asian countries and fix the market causing large price increases and big fat margins, anyone could have told them china would be trying to do this. When companies abuse their position for long enough, many people lose sympathy for them when their tech gets stolen.
For the most part, DRAM was/is a commodity product - pricing largely depends on market conditions. There are probably exceptions. DRAM is also extremely volatile in pricing (see attached link) - in 2016, DRAM demand dropped 8%, only for it to surge 77% in 2017. The forecast for 2018 is 30%. As per the IC insights article, bit growth for the big three are expected to average 20%.
Here's the problem. DRAM is cyclical. In the mid-2000's everyone made a ton of DRAM, expecting huge demand. That never materialized, ASP dropped, and everyone lost a ton of money. I remember when we made a $1 billion dollar profit, and I remember when we lost close to half a billion dollars. In either case, business as normal. There was never an attempt to decrease capacity. Our job was to move wafers, and we moved as many as we could.
Fabs are also stupid expensive to build these days, and take a while to actually build, then start up. Back in the day, a couple billion dollars was a reasonable amount of money for a fab. Today it's probably more. Profit drives the ability to expand. That's why over the years, the amount of companies actually making DRAM has sharply declined.
Furthermore, just because your factory is online, doesn’t mean you can expect profits day 1. You're not going to sell the first DRAM wafer that comes out the fab. I also remember hearing that some company took a capacity hit when they transitioned to 3D NAND because of extra tooling, production/yield ramps, etc.
That comes to patents. IP is important, and protects your investment, which then drives growth. If it took you $10 billion dollars to develop a product only for China to steal your patent, that might put them $5 billion dollars ahead (i came up with the numbers, but you get the point).
The current state of semiconductors is that everything is hard, and expensive - see last link.
“3nm will cost $4 billion to $5 billion in process development, and the fab cost for 40,000 wafers per month will be $15 billion to $20 billion,” IBS’ Jones said.
So, I get the frustration with DRAM/NAND pricing. I remember the days of $10 RAM. However, there’s a lot that goes into DRAM production, and by extension cost. From the articles i’ve been reading, it seems like many of these fabs are running at full capacity, and any increase requires expansion - which involves, time, money, and people. None of which happens overnight.
http://www.icinsights.com/news/bulletins/are-the-major-dram-suppliers-stunting-dram-demand/
https://www.electronicsweekly.com/b...7-will-grow-30-2018says-dramexchange-2018-02/
https://semiengineering.com/big-trouble-at-3nm/
