Breaking News
Product News

ARM announces 8-way graphics core

11/10/2011 08:52 AM EST
11 comments
NO RATINGS
Page 1 / 2 Next >
More Related Links
View Comments: Threaded | Newest First | Oldest First
help.fulguy
User Rank
Author
re: ARM announces 8-way graphics core
help.fulguy   11/10/2011 4:24:48 PM
NO RATINGS
Poor ARM.....Looks like they are set to loose in this battle. Samsung, who was one of the biggest ARM Mali license, announced today that they licensed Imagination Technologies SGX Core. A fe months back, ST Ericsson made the same switch. ARM is losing....big time....

Scott Sutton
User Rank
Author
re: ARM announces 8-way graphics core
Scott Sutton   11/10/2011 7:18:20 PM
NO RATINGS
Recent A15 announcements from ARM (Freescale, HiSilicon, TI...) do not include any mention of ARM GPUs. Given ARMs PR push on next generation Mali GPUs these must be Imagination or Vivant design wins? As fulguy mentions, Samsung/ST are switched from Mali to Imagination. NVIDIA, Broadcom and Qualcomm have traditionally developed their own GPUs.

AJones
User Rank
Author
re: ARM announces 8-way graphics core
AJones   11/10/2011 9:37:59 PM
NO RATINGS
Samsung haven't announced anything! Imagination issued an RNS to the LSE. The fact there is no press release on the Imagination website, shows there is no real substance behind this "announcement" and Samsung would have done the same if there was something worth shouting about.

help.fulguy
User Rank
Author
re: ARM announces 8-way graphics core
help.fulguy   11/12/2011 3:20:31 AM
NO RATINGS
here you go: http://www.imgtec.com/corporate/newsdetail.asp?NewsID=656

AJones
User Rank
Author
re: ARM announces 8-way graphics core
AJones   11/13/2011 11:40:36 AM
NO RATINGS
This is not a typical press release that is approved by both parties. It's an RNS as i mentioned. I'm trying to find the a counterpart that Samsung has posted. There isn't one. It seems this is a mere attempt to try and disrupt ARM's announcement.

dirk.bruere
User Rank
Author
re: ARM announces 8-way graphics core
dirk.bruere   11/10/2011 7:46:16 PM
NO RATINGS
ARM + nVidia seems to be the winning combination

Neo10
User Rank
Author
re: ARM announces 8-way graphics core
Neo10   11/11/2011 3:33:27 AM
NO RATINGS
This could be make it or break it for ARM graphics business which bought the small graphics chip design company 3 years ago. This could have the performance which the market is looking for unlike their previous offerings which were on the lower end of the scale. A small power efficient graphics core would certainly boost the handling of visual data on the handhelds wihtout burning a significant hole in the power budget.

t.alex
User Rank
Author
re: ARM announces 8-way graphics core
t.alex   11/11/2011 12:59:25 PM
NO RATINGS
Just a basic question. What does it mean by 8way graphics core? Is it a marketing term?

AJones
User Rank
Author
re: ARM announces 8-way graphics core
AJones   11/13/2011 11:41:39 AM
NO RATINGS
8way means 8-Core

Peter Clarke
User Rank
Author
re: ARM announces 8-way graphics core
Peter Clarke   11/14/2011 9:55:19 AM
NO RATINGS
@t.alex I was trying to avoid saying "8-core graphics core" which although logically correct reads oddly. It is indeed a graphics core with up to 8 shader cores inside it.

Peter Clarke
User Rank
Author
re: ARM announces 8-way graphics core
Peter Clarke   11/14/2011 9:55:34 AM
NO RATINGS
@AJones You got it

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