• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Samsung Announces Breakthrough in Building Blocks of 3nm Circuits, Updates Roadmap

btarunr

Editor & Senior Moderator
Staff member
Joined
Oct 9, 2007
Messages
47,281 (7.53/day)
Location
Hyderabad, India
System Name RBMK-1000
Processor AMD Ryzen 7 5700G
Motherboard ASUS ROG Strix B450-E Gaming
Cooling DeepCool Gammax L240 V2
Memory 2x 8GB G.Skill Sniper X
Video Card(s) Palit GeForce RTX 2080 SUPER GameRock
Storage Western Digital Black NVMe 512GB
Display(s) BenQ 1440p 60 Hz 27-inch
Case Corsair Carbide 100R
Audio Device(s) ASUS SupremeFX S1220A
Power Supply Cooler Master MWE Gold 650W
Mouse ASUS ROG Strix Impact
Keyboard Gamdias Hermes E2
Software Windows 11 Pro
Samsung Electronics Co., Ltd., a world leader in advanced semiconductor technology, today announced its ongoing commitment to foundry innovation and service at the Samsung Foundry Forum 2019 USA, providing the silicon community with wide-ranging updates on technology advances that support the most demanding applications of today and tomorrow.

The event, held today in Santa Clara, California, features top Samsung executives and industry experts reviewing progress on semiconductor technologies and foundry platform solutions that enable developments in artificial intelligence (AI), machine learning, 5G networking, automotive, the Internet of Things (IoT), advanced data centers and many other domains.



"We stand at the verge of the Fourth Industrial Revolution, a new era of high-performance computing and connectivity that will advance the daily lives of everyone on the planet," said Dr. ES Jung, President and head of Foundry Business at Samsung Electronics.

"Samsung Electronics fully understands that achieving powerful and reliable silicon solutions requires not only the most advanced manufacturing and packaging processes as well as design solutions, but also collaborative foundry-customer relationships grounded on trust and shared vision. This year's Foundry Forum is filled with compelling evidence of our commitment to progress in all those areas, and we're honored to host and converse with our industry's best and brightest," Dr. Jung added.

The new 3nm GAE PDK version 0.1 is ready
Samsung's 3nm Gate-All-Around (GAA) process, 3GAE, development is on track. The company noted today that its Process Design Kit (PDK) version 0.1 for 3GAE has been released in April to help customers get an early start on the design work and enable improved design competitiveness along with reduced turnaround time (TAT).

Compared to 7nm technology, Samsung's 3GAE process is designed to provide up to a 45 percent reduction in chip area with 50 percent lower power consumption or 35 percent higher performance. The GAA-based process node is expected to be widely adopted in next-generation applications, such as mobile, network, automotive, Artificial Intelligence (AI) and IoT.

Conventional GAA based on nanowire requires a larger number of stacks due to its small effective channel width. On the other hand, Samsung's patented version of GAA, MBCFET (Multi-Bridge-Channel FET), uses a nanosheet architecture, enabling greater current per stack.

While FinFET structures must modulate the number of fins in a discrete way, MBCFET provides greater design flexibility by controlling the nanosheet width. In addition, MBCFET 's compatibility with FinFET processes means the two can share the same manufacturing technology and equipment, which accelerates process development and production ramp-up.

Samsung recently taped out the 3GAE test vehicle design and will focus on improving its performance and power efficiency going forward.

The launching of a new SAFE-Cloud program
As part of its ongoing efforts to support and enhance customers' entire design workflow, Samsung Electronics launched the Samsung Advanced Foundry Ecosystem Cloud (SAFE -Cloud) program. It will provide customers with a more flexible design environment through collaboration with major public cloud service providers, such as Amazon Web Services (AWS) and Microsoft Azure, as well as leading Electronic Design Automation (EDA) companies, including Cadence and Synopsys.

To date, most foundry customers have built and managed design infrastructure on their own servers. The SAFE-Cloud program reduces this burden and supports easier, faster and more efficient design efforts by providing an excellent turnkey design environment with extensive process information (PDK, design methodologies), EDA tools, design assets (IP, library) and design services.

Customers can be assured of as much server and storage space as they need, as well as a safe environment optimized for chip design, due to Samsung Electronics' verification of SAFE -Cloud's security, applicability and expandability.

Utilizing the SAFE -Cloud platform, Samsung was able to accelerate the development of its 7nm and 5nm cell libraries in collaboration with Synopsys. In addition, Samsung, Gaonchips - a fabless design company in Korea - and Cadence have successfully completed design verification based on the platform.

"Making up-front investments in high-performance computing (HPC) servers and systems can be a challenge for a company like us," said Kyu Dong Jung, CEO of Gaonchips. "SAFETM-Cloud offers us a very flexible design environment without requiring investment in additional infrastructure, as well as reduced design TAT. I expect this program to provide more tangible business and technical benefits to us and the entire fabless industry."

Process technology roadmap and advanced packaging updates
Samsung's roadmap includes four FinFET-based processes from 7nm down to 4nm that leverage extreme ultraviolet (EUV) technology as well as 3nm GAA, or MBCFET .

In the second half of this year, Samsung is scheduled to start the mass production of 6nm process devices and complete the development of 4nm process.

The product design of Samsung's 5nm FinFET process, which was developed in April, is expected to be completed in the second half of this year and go under mass production in the first half of 2020.

Extensions of the company's FD-SOI (FDS) process and eMRAM together with an expanded set of state-of-the-art package solutions were also unveiled at this year's Foundry Forum. Development of the successor to the 28FDS process, 18FDS, and eMRAM with 1Gb capacity will be finished this year.

View at TechPowerUp Main Site
 

Space Lynx

Astronaut
Joined
Oct 17, 2014
Messages
17,413 (4.69/day)
Location
Kepler-186f
Processor 7800X3D -25 all core
Motherboard B650 Steel Legend
Cooling Frost Commander 140
Video Card(s) Merc 310 7900 XT @3100 core -.75v
Display(s) Agon 27" QD-OLED Glossy 240hz 1440p
Case NZXT H710 (Red/Black)
Audio Device(s) Asgard 2, Modi 3, HD58X
Power Supply Corsair RM850x Gold
nom nom
 
Joined
Jun 3, 2010
Messages
2,540 (0.48/day)
MBCFET 's compatibility with FinFET processes means
...at last, we can have FDS + FinFET + graphene transistors. Back-body biasing and all! Just imagine how hot these chips will run, I say 125°C will be the norm.
 

cdawall

where the hell are my stars
Joined
Jul 23, 2006
Messages
27,680 (4.12/day)
Location
Houston
System Name All the cores
Processor 2990WX
Motherboard Asrock X399M
Cooling CPU-XSPC RayStorm Neo, 2x240mm+360mm, D5PWM+140mL, GPU-2x360mm, 2xbyski, D4+D5+100mL
Memory 4x16GB G.Skill 3600
Video Card(s) (2) EVGA SC BLACK 1080Ti's
Storage 2x Samsung SM951 512GB, Samsung PM961 512GB
Display(s) Dell UP2414Q 3840X2160@60hz
Case Caselabs Mercury S5+pedestal
Audio Device(s) Fischer HA-02->Fischer FA-002W High edition/FA-003/Jubilate/FA-011 depending on my mood
Power Supply Seasonic Prime 1200w
Mouse Thermaltake Theron, Steam controller
Keyboard Keychron K8
Software W10P
...at last, we can have FDS + FinFET + graphene transistors. Back-body biasing and all! Just imagine how hot these chips will run, I say 125°C will be the norm.

If they are engineered to run at that temp, I could not care less how hot they run.
 
Joined
Jun 3, 2010
Messages
2,540 (0.48/day)
If they are engineered to run at that temp, I could not care less how hot they run.
I was earnest for a GF FDX + FinFET attempt, but Samsung is the top player - it would be better for them to lead the field. Funny how no one else could plant the greatest silicon technology advancement, the graphene, on the drawing board until Samsung the recent 2nd player ruled all doubt obsolete as to how cnt-graphene foils would be integrated. Well done Samsung, great stuff.:)
PS: we now have infinite gate surface area.

In the OFF-region, the FinFETs have over an order of magnitude larger OFF-current that those of the equivalent GAA NWs. In the ON-region, the 25/10.7 nm gate length FinFETs deliver 20/58% larger ON-current than the 22/10 nm gate length GAA NWs.
Mind you: the FinFETs here are not your regular bulk, they are;
Si SOI FinFETs
...so indeed GAA's, apart from process variability, are going to be the stepping stone towards cost effective performance that is even higher than the feasibly inconvenient contemporary limits.
 
Joined
Jul 4, 2018
Messages
245 (0.10/day)
When I see news like this I think that the level of technology we are achieving is mind blowing. wishing to be able to live as long as possible to see future achievements... but at the same time realizing that there are people still killing each other over religions.
 
Joined
Oct 25, 2018
Messages
278 (0.12/day)
Good news for the AI robots that will control us.

a few more years and they will have the chips powerful
enough to put in our AI robot overloads. They can learn
as they go; then transmit that knowledge to a main system
data base run by a huge quantum computer "skynet"
If they lack the computational power to make a decision they can "phone
home" with their new 26GHZ 5G connections to seek further
guidance and instructions.

And don't worry they can recreate as many robots they want
since IVP6 will offer endless addresses for everything
around us!!!

Eventually Skynet will have gathered so much information on us
from our personal devices; the robots can take our places
and assume our roles in life for us.

The robots will then wonder why humans created machines to be their
masters. And ponders the need for their existence.....


anyways.....


3nm is pretty small. I do wonder about the heat?
if IPC is higher; than can we just reduce the clock speed to bring it under control?
 
Joined
Jun 3, 2010
Messages
2,540 (0.48/day)
Good news for the AI robots that will control us.

a few more years and they will have the chips powerful
enough to put in our AI robot overloads. They can learn
as they go; then transmit that knowledge to a main system
data base run by a huge quantum computer "skynet"
If they lack the computational power to make a decision they can "phone
home" with their new 26GHZ 5G connections to seek further
guidance and instructions.

And don't worry they can recreate as many robots they want
since IVP6 will offer endless addresses for everything
around us!!!

Eventually Skynet will have gathered so much information on us
from our personal devices; the robots can take our places
and assume our roles in life for us.

The robots will then wonder why humans created machines to be their
masters. And ponders the need for their existence.....


anyways.....


3nm is pretty small. I do wonder about the heat?
if IPC is higher; than can we just reduce the clock speed to bring it under control?
The more I read, the more I am struck. Process variations are less and you get a nice & even On charge. Basically, what we knew about dark silicon and binning does not apply past FinFETs.
 
D

Deleted member 67555

Guest
This is impressive as fuck.
I knew someone would get there but with Intel getting stuck at 10 and AMD just going on 7 I didn't think it would be anytime soon.. Honestly thought it'd be at least 2022 before this... Just wow
 
Joined
Mar 10, 2010
Messages
11,878 (2.20/day)
Location
Manchester uk
System Name RyzenGtEvo/ Asus strix scar II
Processor Amd R5 5900X/ Intel 8750H
Motherboard Crosshair hero8 impact/Asus
Cooling 360EK extreme rad+ 360$EK slim all push, cpu ek suprim Gpu full cover all EK
Memory Corsair Vengeance Rgb pro 3600cas14 16Gb in four sticks./16Gb/16GB
Video Card(s) Powercolour RX7900XT Reference/Rtx 2060
Storage Silicon power 2TB nvme/8Tb external/1Tb samsung Evo nvme 2Tb sata ssd/1Tb nvme
Display(s) Samsung UAE28"850R 4k freesync.dell shiter
Case Lianli 011 dynamic/strix scar2
Audio Device(s) Xfi creative 7.1 on board ,Yamaha dts av setup, corsair void pro headset
Power Supply corsair 1200Hxi/Asus stock
Mouse Roccat Kova/ Logitech G wireless
Keyboard Roccat Aimo 120
VR HMD Oculus rift
Software Win 10 Pro
Benchmark Scores 8726 vega 3dmark timespy/ laptop Timespy 6506
This is impressive as fuck.
I knew someone would get there but with Intel getting stuck at 10 and AMD just going on 7 I didn't think it would be anytime soon.. Honestly thought it'd be at least 2022 before this... Just wow
It is version .01 of a Sdk to make chips, it probably will be 2022 before you see any unless your in military or academic circles , 7nm is said to be a long run node ,as is 5 Nm ie 2+years minimum though nodes run concurrent this is just a release of definitive but .01 edition ground rules for circuit design.

were some way off working silicon that's also affordable.
 
Joined
Nov 4, 2005
Messages
12,000 (1.72/day)
System Name Compy 386
Processor 7800X3D
Motherboard Asus
Cooling Air for now.....
Memory 64 GB DDR5 6400Mhz
Video Card(s) 7900XTX 310 Merc
Storage Samsung 990 2TB, 2 SP 2TB SSDs, 24TB Enterprise drives
Display(s) 55" Samsung 4K HDR
Audio Device(s) ATI HDMI
Mouse Logitech MX518
Keyboard Razer
Software A lot.
Benchmark Scores Its fast. Enough.
So... We're at 15 atoms now. Isn't the silicon limit 4 atoms for a transistor?
Yup, we are at roughly 7 or 8 atom thick process. 3D chips with better design a couple layers thick with a change to the mounting will probably be next as we are far up the diminishing returns ladder, and we have it working with memory chips.
 
Joined
Sep 15, 2011
Messages
6,753 (1.39/day)
Processor Intel® Core™ i7-13700K
Motherboard Gigabyte Z790 Aorus Elite AX
Cooling Noctua NH-D15
Memory 32GB(2x16) DDR5@6600MHz G-Skill Trident Z5
Video Card(s) ZOTAC GAMING GeForce RTX 3080 AMP Holo
Storage 2TB SK Platinum P41 SSD + 4TB SanDisk Ultra SSD + 500GB Samsung 840 EVO SSD
Display(s) Acer Predator X34 3440x1440@100Hz G-Sync
Case NZXT PHANTOM410-BK
Audio Device(s) Creative X-Fi Titanium PCIe
Power Supply Corsair 850W
Mouse Logitech Hero G502 SE
Software Windows 11 Pro - 64bit
Benchmark Scores 30FPS in NFS:Rivals
So... We're at 15 atoms now. Isn't the silicon limit 4 atoms for a transistor?
where did you get that number?? A quick search show that 1nm = ~2000 atoms minimum, depending on the material.
 
Joined
Mar 13, 2012
Messages
396 (0.08/day)
Location
USA
where did you get that number?? A quick search show that 1nm = ~2000 atoms minimum, depending on the material.
Diameter of a silicon atom is 210 picometers, (according to google - other sources have it between 110ish to 240ish) so roughly 5 silicon atoms across per nanometer
 
Joined
Nov 4, 2005
Messages
12,000 (1.72/day)
System Name Compy 386
Processor 7800X3D
Motherboard Asus
Cooling Air for now.....
Memory 64 GB DDR5 6400Mhz
Video Card(s) 7900XTX 310 Merc
Storage Samsung 990 2TB, 2 SP 2TB SSDs, 24TB Enterprise drives
Display(s) 55" Samsung 4K HDR
Audio Device(s) ATI HDMI
Mouse Logitech MX518
Keyboard Razer
Software A lot.
Benchmark Scores Its fast. Enough.
Diameter of a silicon atom is 210 picometers, (according to google - other sources have it between 110ish to 240ish) so roughly 5 silicon atoms across per nanometer


"The thing is, atoms are very, very small, but they still have a finite size. A hydrogen atom, for example, is about 0.1 nanometers, and a caesium atom is around 0.3nm. The atoms used in silicon chip fabrication are around 0.2nm. Now, you would be right in thinking that you can get hundreds of atoms into 22 or 16nm — but that’s not the size of individual transistors; that’s actually a measure of the distance between discrete components on a chip. In the case of the 22nm chips — a process that only Intel has mastered and will come to market with Ivy Bridge — the high-κ dielectric layer is just 0.5nm thick; just two or three atoms."

My bad, I thought the layers were still 5 to 7 thick. Even more awesome.
 
Top