SSDs might be getting less interesting, but that’s not necessarily bad

Ars Technica has a story about SSD news coming out of CES.

Basically, they’re predicting that the big news this year will be consolidation and lower prices. That may be bad news for someone who writes about SSDs for a living (I don’t), but good news for consumers.

The article cites the HDD market of 20-30 years ago, when there were dozens of companies making hard drives and there wasn’t much difference between them. Over time, those companies consolidated, to the point where now we’re down to three manufacturers: Seagate, Western Digital, and Toshiba.

The complexities of getting controllers and memory to work together favor the companies who make the devices themselves, rather than integrators like Kingston, Corsair, and the other 30-something companies who make memory modules and SSDs. This means companies like Intel, Samsung, and Toshiba have an inherent advantage over everyone else because they make controllers and memory. Since they make both, they’ll know the peculiarities of their own controllers and memory better than anyone else and can make design decisions based on that.

This makes me wonder if Toshiba is going to decide to try to make a bigger splash in the SSD space, much like Samsung did last year. So far, Toshiba has been content to sell SSDs to OEMs, especially Apple, rather than directly to consumers. Given Apple’s uncomfortable relationship with Samsung, it could be that Toshiba does very well just selling drives to Apple and doesn’t really need you and me. I don’t own a Toshiba drive, but I do own a Kingston drive with a Toshiba controller and Toshiba memory in it, and the biggest complaint I ever had about it was that Kingston stopped making it soon after I bought it. My Samsung drives are faster, but that Toshiba-powered Kingston delivered a good balance of price and performance in its day, and it’s been exceptionally reliable too. For what it’s worth, SSDLife thinks that this 13-month-old drive has seven years left in it. I’m not sure how useful the drive will be in 2020.

Besides consolidation, prices will continue to fall due to process shrinks and increased production. That’s good for everybody. While I’m always in favor of greater speed, what we need more than speed right now is stability and increased affordability.

Right now a Samsung 840 delivers 500 GB for around $350, which is close to 70 cents per GB. One reason the price is low is because Samsung is using a new memory technology, TLC. If SSD pricing is going to reach 50 cents per GB before mid-2014, it’s going to be aggressive moves like that that get it there.
Another possibility is that we may eventually see budget SSDs that do things like pair TLC memory with cheaper, lower-performing controllers like the Phison PS3105 used in the Crucial v4 drive, to get costs down faster. Even a cheap SSD is far faster than any hard drive, so there’s probably room for both budget SSDs and enthusiast-grade SSDs, the same way there are budget-grade CPUs and hard drives (think 5400-RPM HDDs, Celeron and Pentium and AMD CPUs) and enthusiast-grade CPUs and hard drives (think 7200- and 10,000-RPM hard drives and Core i7 CPUs).

%d bloggers like this:
WordPress Appliance - Powered by TurnKey Linux