Thursday, February 02, 2012

HP's memristor delusion

Recently EETimes (link)  has picked up on e-mail correspondences I have had with a few industry scientists regarding HP's so-called "memristor." I had recently posted an article on ArXiv debunking the mathematical basis for Leon Chua's memristor theory from the 1970's.  HP is using the memristor theory to claim credit for a new form of non-volatile memory called RRAM which is under development by Panasonic, Toshiba, Sharp, Unity Semiconductor, and a variety of other companies. It turns out that many of the scientists involved in legitimate research in RRAM view HP's memristor as a foolish gimmick which is being perpetuated by naive academics and science writers.

The story behind the memristor delusion is actually quite interesting and I went into more detail in a paper I posted on scribd available at this link. The article directly addresses public comments made by Stan Williams, one of the lead researchers of HP.  The article also points out some evidence of plagarism in HP's original 2008 Nature paper based on an earlier patent application of Samsung and a paper published in 1968 on resistance switching in titanium oxide thin films.