Why the problem with NOR flash might just be NAND flash and what next-generation memory might prove to be an even bigger competition.
The mobile market might be an engine of growth for NAND flash but it's not nearly as friendly to NOR flash.
Indeed, IHS iSuppli analyst Ryan Chien expects the one-time powerhouse wireless NOR flash market to shrink by 28 percent in 2013, reaching $725 million by year end, compared to $1 billion in 2012. It's a state of affairs that's likely to continue -- the report calls for wireless NOR flash revenues in 2014 to drop further, to $480.4 million, reaching just $190.4 million by the end of the forecast period in 2017.
There was a time NOR flash was the go-to memory for mobile apps, supporting high-speed read operations for run-in-place code execution. Back in the day when the most exciting things users did with their cellphones was store phone numbers and send text messages, NOR flash played an important role in helping handsets power up quickly. With the advent of smartphones, however, the focus shifted from downloading boot code to managing lots of data, and that meant changing to a nonvolatile memory with high density and greater write speed than NOR flash can offer. Says Chien:
In smartphones, parallel NOR has been squeezed out by NAND flash, which can emulate NOR flash capabilities at far higher densities. On the low end, where entry-level handsets are involved, parallel NOR is feeling pressure from serial peripheral interface (SPI) NOR, which is increasingly important to handset manufacturers interested in keeping costs down. Meanwhile, in media tablets, NOR is used only in Android-based models but not in the Apple iPad, the industry leader.
In a more interesting observation, he sees NOR flash also pressured by competition from phase-change memory (PCM), with the non-volatile memory technology expected to surge from $67 million in revenues for 2013 to $138 million in 2014. With limited uptake by only a handful of manufacturers, however, he expects that growth to flatten out, finishing the forecast period in 2016 at $170 million.
Meanwhile, as previously reported, the mobile device market for DRAM should consume $11.6 billion of DRAM by 2015, 15 percent less than the $9.9 billion in DRAM consumed by PCs.
Elsewhere, NAND flash pricing rose 2 percent to 5 percent over the second half of June compared to the first half, reports TrendForce's DRAMeXchange. They expect third-quarter capacity to rise by 6.1 percent compared to the third quarter. Contrast that with the expected demand of 10 percent to 15 percent in the same period, driven by mobile computing products. Take it with a grain of salt, however, since the numbers appear to come from module manufacturers. They, too, are banking on the enterprise SSD market to buoy NAND flash demand, driving what they anticipate to be a healthy third quarter.