News

Taiwan Semiconductor Manufacturing Company, the world’s No. 1 contract semiconductor manufacturer, has announced 55nm process technology, an optical shrink for its 65nm fabrication process, which allows to make chips smaller and more energy efficient and does not require designers of processors to fully redevelop their products.

“TSMC’s half-node process, including 55nm, is the quickest and simplest way for our customers to be cost competitive in the rapidly changing marketplace. TSMC continues to combine manufacturing superiority with a comprehensive design ecosystem to support customers of any size, from startups to multinational giants,” said Jason Chen, vice president of corporate development of TSMC.

TSMC’s 55nm process technology is a 90% linear-shrink process from 65nm including I/O and analog circuits. The process delivers reportedly significant die cost savings from 65nm, while offering the same speed and 10 to 20% lower power consumption. Because the 55nm process is a direct shrink, chip designers can leverage existing libraries and port their 65nm designs with minimal risk and effort. The 55nm logic family includes general purpose (GP) and consumer (GC) platforms. Initial production of the 55GP begins this quarter, followed later in the year by 55GC.

Even though many of TSMC’s customers have already adopted the 65nm process, the semiconductor maker will allow them, as well as other clients, to test production using its CyberShuttle prototyping program that allows multiple customers and IP suppliers to share the costs of a single mask set and prototype wafers on a pilot run. The 55nm CyberShuttle runs are expected to be offered on a bi-monthly basis starting from the beginning of May this year.

TSMC’s customers who make cost-efficient products, such as ATI, graphics product group of AMD, and Nvidia Corp. are projected by unofficial sources to release first chips made using 55nm process technology in early 2008.

Discussion

Comments currently: 2
Discussion started: 03/28/07 08:52:27 AM
Latest comment: 03/28/07 10:58:04 AM

[1-2]

1. 
HELLO EVERYONE!!
I know this isn't the subject & the right place but I have some intresting tidbits & info for you to enjoy with!!!

THE SPECS FOR THE UPCOMING nVIDIA GPUs
IS HERE!!!(provided to me by some unknown source...
I have some intresting information to youL:
nVIDIA is releasing their 8800 Ultra and 8950GX2 as well as other GPUs in the 8300,8400 and 8600 series on the 17 of April...prices and specs are as follows :

8950 GX2 (192 Stream Processors working at 1.70GHz)
Core Clock : 575MHz
Dual GPU design comprised out of two GPUs each of which has 96 SPs for a total of 192 SPs clocked at 1.70GHz..
640-bit memory bus (320bit*2)
96 TF/48 TA
Memory Size : 1280 MB GDDR4 Memories at 2.0GHz
160 GB/s Memory Bandwidth
8950GX2 comes with optional Water Cooling thogh water cooling won't be the retail version
The version of the 8950GX2 sold in stores will not have water cooling of any type!!!
price : 649 USD

PROS:
8950 GX2 will be 50% faster than a 8800GTX
CONS:
high PSU requirement and power consumption..
At 649 USD the 8950GX2 is VERY expensive

........................................................
8800 Ultra (128 Stream Prcessors at 1.80GHz)
Still on a 90nm process at TSMC
modified to support higher clocks than before
64 TF/32 TA
Core : 650MHz/
768MB GDDR4 at 2.2GHz /105.6 GB/s memory Bandwidth
8800 Ultra will be 30% FASTER than a 8800GTX in real world apps & Video games
& optimizations to the core to allow higher clocks (Remember 7800GTX 512 ??)
price for an 8800 Ultra : 549 USD
NO WATER COLLING FOR 8800 ULTRA!!
Cons: prices might be higher due to low supplies.(supplies will get better overtime & prices will go down)
price:549USD
......................................................
8600 GTS Spes & prices:(G84)
midrange GPU.
80nm 300 million transistors at TSMC
64 Stream Processors at 1.65GHz
32 TF/16 TA
Core Clock at 675MHz
Memory Clock at 2.0GHz (1ns)
128 bit Memory Bus /256MB-512MB GDDR4/32 GBs Bandwidth
price : 229USD for 512MB version and 199USD for 256MB version..
pros: 8600GTS is 50% faster than a Radeon X1950 XTX 512MB in real world apps & video games..
Cons : narrow memory bus
pros: amazing performance & best bang for the buck!!! (8600 GTS 3Dmark2006 score is
6500 points in 1280x1024 res on a stock X6800 Conroe CPU)
......................................................
8600 GT Specs:(G84)
same as 8600 GTS just with lower clocks
64 SPs at 1.30GHz
Core at 540MHz
128 bit memory bus
Memory Clock :1.40GHz using 1.4ns 256MB GDDR3 memories
performance for an 8600GT: same as an X1950 XTX & 7900GTX..
Cons : narrow memory bus
price : 169 USD

.....................................................
8500 GT Specs (Cut down G84)
48 Stream Processors at 1.20GHz
Core : 470 MHz
256MB GDDR3
performance : same as an 7900/7950 GT
cons: no cons for its class...
price 129 USD
................................................
8400 GS (Cut-down G84) Specs
48 SPs at 1.0 GHz
Core at 400MHz
Memory at 1.20GHz
256MB GDDR3
performance: in the range of 7900GS
Cons: No cons for its class
price: 99 USD
...................................................
8300 GS (c G86) Spes
32 Stream Processors at 1.0 GHz
Core : 400MHz
128/256MB GDDR3
performance : in the range of a 7600 GT
price: 79 USD
...............................................
After the this product release the prices MSRP for the 8800 GTX will go down to nice 449 USD..While the price for the 8800 GTS will go down for an impressive 299USD for the 8800GTS 640MB version and 249 USD for the 8800GTS 320MB version...

source : http//www.vr-zone.com/?i=8659
[Posted by: Erik Larsson  | Date: 03/28/07 08:52:27 AM]

2. 
"The 55nm logic family includes general purpose (GP) and consumer (GC) platforms. Initial production of the 55GP begins this quarter, followed later in the year by 55GC. "

Cool since this quarter have 3 days left =)
[Posted by: Silver  | Date: 03/28/07 10:58:04 AM]

[1-2]

You must log in to add comments.

Forgot password? Registration

remember me



Related news

Latest News

Monday, November 9, 2009

5:15 pm | Intel to Launch “Clarkdale” Chips on January 7, 2010. Intel to Begin Shipments of Clarkdale Chips in a Month

3:54 pm | Microprocessor Unit Shipments Grow 23% in Q3 2009 – Market Tracking Firm. AMD Continues to Lose Share to Intel As Market Shows Gains

6:56 am | Globalfoundries Set to Become World’s Second Largest Foundry in Two Years – ATIC. ATIC: The First Chip Foundry in Abu Dhabi to Emerge in Four Years

Sunday, November 8, 2009

11:56 pm | Nokia to Replace Faulty Chargers. Nokia Initiates Charger Exchange Program

10:51 pm | Nvidia’s Chief Executive Says “No” to Globalfoundries, Microprocessors. Nvidia Denies Intentions to Use Globalfoundries, Develop Own Central Processing Units