About a year ago I penned a Seeking Alpha article titled, "Will Micron's Phase Change Memory Win The Flash Replacement Race?" Perhaps my crystal ball has been shattered:
- I received comments and private emails telling me to forget it. "PCM is dead so far as Micron Technology (NASDAQ:MU) is concerned" was the common theme.
- Not long after I wrote the article, Micron pulled the parts from its catalog. It had been producing millions of a combined PCM/DRAM part which was powering the popular Nokia Asha cellphone, which was getting good reviews. A Numonyx slide deck dating to just prior to the company's purchase by Micron showed a series of upcoming node shrinks on the PCM parts. Last seen, Micron was at an antique 90nm transitioning to 45nm. Here's a press release about the 45nm 1gb PCM + 512 Mb LPDDR2.
So what's happened to PCM at Micron? The company has been pretty quiet on the disappearance of PCM from its catalog. Here's a snippet from the third quarter earnings call:
(Micron CFO Ron Foster speaking)And then there is a $30 million effect related to some expected sell-through of legacy technology inventory in the fourth quarter. And that will be basically - float out in the third quarters, but I wanted to call that out, because it's a somewhat unique effect. It's a technology (acquired) from our Numonyx acquisition and we will be replacing that with new emerging technology over time, but we did have some legacy inventory that we are selling off and that's related to a specific approach to phase change that we are no longer pursuing in favor of other (variants) of the technology
And then there were a scanty two slides from the 2014 Winter Analyst Day which mentioned PCM:
Writing in The Register, Chris Mellor captioned this slide with his own interesting remarks (emphasis added):
The Emerging Technology timeline is interesting. PCM is Phase Change Memory. With New Memory A Gen 1 and 2 and New Memory B it's clear Micron has not given up on PCM.
So we have the company CFO indicating it was pursuing other "...(variants) of the technology...." and a technical journalist magically determining that two blue blobs on a chart are PCM.
What about the rumors? There is a juicy rumor that (one variant?) of the Hybrid Memory Cube (HMC) contains both PCM and its technological cousin the Ovonic Threshold Switch. Since the HMC has been sampling and there hasn't been a whisper of this in either the technical press or Wall Street sell-side research, I initially discounted the rumor. But off to Google I went and found a very recent IBM (NYSE:IBM) paper which, while not specific to HMC, contained some interesting snippets. The paper is titled "Access Points for 3D Crosspoint Memory" and has a rather long winded explanation of "access points":
The emergence of new nonvolatile memory (NVM) technologies - such as phase change memory, resistive, and spin-torque-transfer magnetic RAM - has been motivated by exciting applications such as storage class memory, embedded nonvolatile memory, enhanced solid-state disks, and neuromorphic computing. Many of these applications call for such NVM devices to be packed densely in vast "crosspoint" arrays offering many gigabytes if not terabytes of solid-state storage. In such arrays, access to any small subset of the array for accurate reading or low-power writing requires a strong nonlinearity in the IV characteristics, so that the currents passing through the selected devices greatly exceed the residual leakage through the non-selected devices. This nonlinearity can either be included explicitly, by adding a discrete access device at each crosspoint, or implicitly with an NVM device which also exhibits a highly nonlinear IV characteristic.
With 30 mentions of "PCM" and 5 of "Ovonic Threshold Switch" this article would seem to lend some credence to the rumor.
Conclusions. So PCM seems to have been put in an induced coma within Micron, but it is not dead. The technology is too bright to kill but its ultimate application may be hybridized with other technologies, as the HMC rumor would indicate.
The memory industry has been involved in a decade or so of evolutionary change rather than explosive innovations and new technologies. I believe the next decade will be different as the blizzard of new memory types, eg. RRAM, spin torque, PCM, etc., are winnowed down. Past technical battles like Edison's DC current vs. Tesla's AC, betamax vs VHS tape, CDMA vs GSM cell technology, etc. have resulted in high prices until production volumes solidified behind a "winner." Micron has a commanding intellectual property position in PCM and if the technology is awakened either in a standalone or hybridized implementation Micron should do very well.
Disclosure: The author is long MU, INTC, SNDK, IBM. The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.