Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There is a real physical device with a real application and that is called Resistive RAM (RRAM). Then there is a phenomenological mathematical model of a two terminal device.

In 2005 some guy from HP wrote a highly publicized paper where he tried to assign the model to the device and called it memristor.

There are a couple of issues with this:

1) The model is not physics based and is therefore not suitable to accurately describe the device operation.

2) The device existed even before the crude memristor model was attached to it. The memristor model did not help the application of RRAM in any way, it only publicized it in a way that made many people misunderstand its origins and application.

3) Research and development of RRAM is at an all-time high right now. Most semiconductor conferences have several sessions dedicated to this device, not the memristor. Actually there has been tremendous progress in RRAM. Toshiba/Sandisk made a 32Gbit prototype device and there are demonstrators of microcontrollers using RRAM. The only thing preventing market introduction right now is the momentum that flash has.

edit: HP is all talk. They have not shown anything in term of practical application yet. I don't understand how they are supposed to have a product any time.



Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: