HP pushing back memristor technology debut until at least 2014

By on July 12, 2012, 2:00 PM

HP is now estimating they won’t be able to ship products containing memristor technology until at least 2014, going against predictions late last year that such technology would be feasible in 2013. The problem isn’t with the technology itself, as Stan Williams of the Memristor Research Group at HP puts it, but rather the economics and timing behind the introduction.

During a recent presentation at a conference in Oxnard, California, Williams said that HP would have something technologically viable by the end of next year. He goes on to explain that the science and technology are the easy parts and that literally any foundry could make memristors tomorrow.

The issue lies in the economics, investment and market readiness. HP is working with Hynix on the new technology, a major producer of flash memory. If they launched memristors now, it would seriously cannibalize Hynix’s flash memory division so the timing has to be just right.

Memristor technology is an electrical component that controls the stream of electrons flowing through it but can additionally “remember” its last charge. This is significant as one memristor is described as being as computationally powerful as 10 transistors.

HP and Hynix announced a partnership in 2010 to use this technology to develop Resistive Random Access Memory, or ReRAM. This is the same type of technology that Japanese researchers showcased last month as part of a hybrid solid state drive. The scientists claim the combination of NAND and ReRAM could result in SSDs that can write up to 11 times faster than today’s fastest drives.




User Comments: 5

Got something to say? Post a comment
MilwaukeeMike said:

So, HP doesn't want to sell the new stuff because the old stuff is still selling well. But they say the technology is the 'easy part'. So why doesn't someone else make these things? Does HP own the patent or something?

Anyway, if they wanted to prevent cannabalizing their own stuff, why not price it very high? people would pay a premium for a faster drive, regardless of manufacturing cost. Or is the technology bound by the limits of SATA III?

Seems like there are a lot of unanswered questions here.

spydercanopus spydercanopus said:

They're probably too excited about wasting time with a Surface tablet to be interested in game changers like memristors. HP needs to focus on one really good product instead of blowing their billions on things like palm and webos.

waterytowers said:

Investment and market readiness is the problem? so does that means they have not fully recovered their costs on their currently selling products? I say start developing now and stop wasting the worlds limited resources on products that are no longer the best technology. Our landfills are already overflowing with old technology. Latest tech always carries a premium and as such investments will be recouped. Some companies spend too much time developing incremental releases rather than just skipping straight to the high end version they have had in the wings for 5 years. I understand this is to keep selling and make a profitable/viable business over the long term, but this is at the expense of our resources and slows down progress.

I think I will hold off buying an SSD for a few more years. When current speed devices (~520Mbps/~520Mbps) hit around the same price per GB as platter drives I may consider buying.

Read/write speeds over 5Gbs (10 x 520Mbps) for an SSD drive means I can start looking at upgrading my home network to 10Gbps in the next few years. Copying large files will be nearly instantaneous

MilwaukeeMike said:

Investment and market readiness is the problem? so does that means they have not fully recovered their costs on their currently selling products? I say start developing now and stop wasting the worlds limited resources on products that are no longer the best technology. Our landfills are already overflowing with old technology. Latest tech always carries a premium and as such investments will be recouped. Some companies spend too much time developing incremental releases rather than just skipping straight to the high end version they have had in the wings for 5 years. I understand this is to keep selling and make a profitable/viable business over the long term, but this is at the expense of our resources and slows down progress.

I think I will hold off buying an SSD for a few more years. When current speed devices (~520Mbps/~520Mbps) hit around the same price per GB as platter drives I may consider buying.

Read/write speeds over 5Gbs (10 x 520Mbps) for an SSD drive means I can start looking at upgrading my home network to 10Gbps in the next few years. Copying large files will be nearly instantaneous

Exactly... wtf, right? If we had all the information we'd have the answer to these questions. And you won't have an instanteous network because the SATA interface is only 6Gbps, which is why I wondered if that was the reason they're not making these drives.

Guest said:

Is there any response of HP's research to the recent publication "Fundamental Issues and Problems in the Realization of Memristors" (on arXiv.org)? There are strong arguments that the physics behind the "memristive systems" approach might be in conflict with Landauer's principle of the minimum possible amount of energy required to change stored information in a system.

Load all comments...

Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...
Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.