REGISTER | LOGIN
Breaking News
News & Analysis

Better DRAM with No New Materials

10/11/2016 08:00 AM EDT
12 comments
NO RATINGS
Page 1 / 2 Next >
More Related Links
View Comments: Newest First | Oldest First | Threaded View
Page 1 / 2   >   >>
resistion
User Rank
Author
vs. MRAM?
resistion   10/16/2016 6:34:24 AM
NO RATINGS
So the title hints to ditch MRAM with its new materials?

Ron Neale
User Rank
Author
Re: D? RAM
Ron Neale   10/16/2016 5:40:43 AM
NO RATINGS
Resistion:-Yes they do describe it as a thyristor structure, so by definition that might be interpreted as a three terminal device. In the figure there appears to be a p type well for a possible common connection, although it does not appear to be connected to the base of the npn as would be required for thyristor latching. It would be nice to see an example of a section of their proposed array from Kilopass and the way they propose to write/erase.

resistion
User Rank
Author
Re: D? RAM
resistion   10/15/2016 9:54:13 PM
NO RATINGS
Maybe kilopass intended more than 2 terminal use.

Ron Neale
User Rank
Author
Re: D? RAM
Ron Neale   10/15/2016 11:51:21 AM
Resistion:- Simple answer No, The pnpn will (should) offer rectifying action for reverse voltage, assuming its reverse breakdown voltage is higher than its threshold switching voltage. The chalcogenide threshold switch will be symmetrical. In an array the pnpn might help with nearest neighbour sneak path problems. However, it will be interesting to see how Kilopass plan to deal with the problem of writing a device in the off state when all of its nearest neighbours are held in the conducting state.

resistion
User Rank
Author
Re: D? RAM
resistion   10/15/2016 10:40:58 AM
NO RATINGS
Does it operate the same way?

Ron Neale
User Rank
Author
Re: D? RAM
Ron Neale   10/15/2016 6:05:38 AM
NO RATINGS
Micron/Intel 3DXPoint ™ teams must be breathing a collective sigh of relief. If the memory part of 3DXPoint does not work, i.e. they are unable to get "stable working media" * they can just use the chalcogenide threshold switch, their matrix isolation device, in exactly the same way as the VLT-DRAM.

* http://www.eetimes.com/document.asp?doc_id=1330285&page_number=2

resistion
User Rank
Author
junction leakage
resistion   10/12/2016 7:58:42 PM
NO RATINGS
I suppose before using it I should check spec of junction leakage vs. temperature.

resistion
User Rank
Author
Re: D? RAM
resistion   10/12/2016 11:05:04 AM
NO RATINGS
Yes you are right, dynamic refers to always needing refresh.

zman_tekinsil
User Rank
Author
Re: T-RAM
zman_tekinsil   10/12/2016 9:56:35 AM
NO RATINGS
That is correct, TRAM worked on similar concept starting late 90's. TRAM uses a mosfet to trigger the thyristor. Mosfet gate surrounds the p-region of the 4 layer-pillar of the thyristor. Mosftet triggered devices have been around for a while. Known as IGBT in power/high voltage circles.

sw guy
User Rank
Author
D? RAM
sw guy   10/12/2016 8:51:20 AM
NO RATINGS
I always believed the D in DRAM was a remainder of the refresh thing.
Or, this is a reference to interface at pin level.

Any clue someone ?




Page 1 / 2   >   >>
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed