Breaking News
News & Analysis

MIPS Moves Multicore Battle to Clustering

9/2/2014 04:50 PM EDT
4 comments
Page 1 / 2 Next >
More Related Links
View Comments: Oldest First | Newest First | Threaded View
Kinnar
User Rank
Author
Very nice step to promote MIPS 64
Kinnar   9/3/2014 6:49:59 AM
NO RATINGS
It is very nice to know that support for MIPS 64 is made available on QEMU. This will really promote the use of MIPS 64 in new as well as existing systems. Does QEMU require any specific hardware to simulate MIPS 64 on a normal PC

TarraTarra!
User Rank
Author
Who are the licensees of this design?
TarraTarra!   9/4/2014 3:06:59 PM
NO RATINGS
"All this aims at the larger networking and storage applications for customers such as Broadcom and Cavium"

 

But haven't both Broadcom and Cavium announced that their next generation processors will be ARM based?

 

Nick, did Imagination indicate that this new core will be licensed by Cavium and Broadcom? That would be worthwile to clarify and fact-check. Otherwise this is a just a "press-release" article for imagination.

nickflaherty
User Rank
Author
Re: Who are the licensees of this design?
nickflaherty   9/8/2014 5:38:01 AM
NO RATINGS
The latest Cavium part is ARM-based yes as mentioned in the ARM server story, but Imagination is also including Cavium as a continuing MIPS customer without giving away its customer's product roadmap and the 6400 is very much aimed at Cavium as a customer - this is why it has been an interesting set of stories over the last week

bobdvb
User Rank
Author
Re: Who are the licensees of this design?
bobdvb   9/29/2014 3:37:27 PM
NO RATINGS
I saw Broadcom's CE division at a show recently, they are still shipping new MIPS products but I saw no use of new core designs. A shame really because I have a soft spot for MIPS.

Most Recent Comments
michigan0
 
SteveHarris0
 
realjjj
 
SteveHarris0
 
SteveHarris0
 
VicVat
 
Les_Slater
 
SSDWEM
 
witeken
Most Recent Messages
9/25/2016
4:48:30 PM
michigan0 Sang Kim First, 28nm bulk is in volume manufacturing for several years by the major semiconductor companies but not 28nm FDSOI today yet. Why not? Simply because unlike 28nm bulk the LDD(Lightly Doped Drain) to minimize hot carrier generation can't be implemented in 28nm FDSOI. Furthermore, hot carrier reliability becomes worse with scaling, That is the major reason why 28nm FDSOI is not manufacturable today and will not be. Second, how can you suppress the leakage currents from such ultra short 7nm due to the short channel effects? How thin SOI thickness is required to prevent punch-through of un-dopped 7nm FDSOI? Possibly less than 4nm. Depositing such an ultra thin film less then 4nm filum uniformly and reliably over 12" wafers at the manufacturing line is extremely difficult or not even manufacturable. If not manufacturable, the 7nm FDSOI debate is over!Third, what happens when hot carriers are generated near the drain at normal operation of 7nm FDSOI? Electrons go to the positively biased drain with no harm but where the holes to go? The holes can't go to the substrate because of the thin BOX layer. Some holes may become trapped at the BOX layer causing Vt shift. However, the vast majority of holes drift through the the un-dopped SOI channel toward the N+Source,...

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Like Us on Facebook
EE Times on Twitter
EE Times Twitter Feed