George Papadimitriou
LightNews LightNews
banner
gpapad.bsky.social
George Papadimitriou
@gpapad.bsky.social
55 followers 72 following 12 posts
Assistant Professor at the Computer Engineering and Informatics Department (CEID), University of Patras, Greece
Posts Replies Media Videos
gpapad.bsky.social
Intel’s 18A Process Reportedly Shows “Disappointing” Yield Rates As They Are Now At 20%-30%, Making Mass-Production Impossible

#Intel #SoCs

wccftech-com.cdn.ampproject.org/c/s/wccftech...
Intel's 18A Process Reportedly Shows "Disappointing" Yield Rates As They Are Now At 20%-30%, Making Mass-Production Impossible
Intel's 18A yields are now to be around 20%- 30%, as the company is now claimed to face a massive roadblock to switching to mass production.
wccftech-com.cdn.ampproject.org
March 22, 2025 at 7:11 AM
gpapad.bsky.social
#TSMC, #Intel, #Samsung Foundry ready to battle over the start of 2nm production next year

#semiconductors #processor #Apple #HPC #AI #AMD #Nvidia

www.phonearena.com/news/battle-...
TSMC, Intel, Samsung Foundry ready to battle over the start of 2nm production next year
There are several foundries looking to dominate with the mass production of 2nm chips starting in 2025.
www.phonearena.com
March 16, 2025 at 7:50 PM
Reposted by George Papadimitriou
safari-eth.bsky.social
We're excited to share our recent ACM SIGARCH blog post highlighting our real-world PIM tutorials!

Processing-in-Memory Tutorials: Experiences from Past Two Years and Thoughts Looking Forward:
www.sigarch.org/processing-i...
Geraldo Oliveira, Juan Gómez-Luna, Onur Mutlu
@ethzurich.bsky.social
Processing-in-Memory Tutorials: Experiences from Past Two Years and Thoughts Looking Forward
Processing-in-Memory (PIM) is a computing paradigm that aims to overcome the data movement bottleneck (i.e., wasted execution cycles and energy, resulting from the back-and-forth data movement betw…
www.sigarch.org
January 29, 2025 at 11:29 AM
gpapad.bsky.social
China's #SpacemiT develops 64-core RISC-V datacenter CPU — 12nm chip allegedly performs like a 10-year old Xeon or Opteron but with higher core count

#riscv #intel #amd #datacenter #cpu #ai #meltdown #spectre

www.tomshardware.com/pc-component...
China's SpacemiT develops 64-core RISC-V datacenter CPU on 12nm
Another datacenter-oriented processor from China.
www.tomshardware.com
January 14, 2025 at 8:18 PM
gpapad.bsky.social
Check out our latest article at @OpenAccessGov for the value of simulation frameworks in optimizing the development and integration of photonic and electronic systems

#simulation #gem5 #neuropuls #photonic #RISC-V #AI #accelerators #technology

www.openaccessgovernment.org/article/acce...
Accelerating the development of secure and reliable photonic and electronic systems
The value of simulation frameworks in optimizing the development and integration of photonic and electronic systems
www.openaccessgovernment.org
January 8, 2025 at 9:30 AM
gpapad.bsky.social
30-year-old #Pentium FDIV bug tracked down in the silicon — Ken Shirriff takes the microscope to Intel's first-ever recall

#Intel #bug #transistors #srt #division #error #cpu

www.tomshardware.com/pc-component...
30-year-old Pentium FDIV bug tracked down in the silicon — Ken Shirriff takes the microscope to Intel's first-ever recall
Hardware historian finds the specific transistors that cost Intel $475 million
www.tomshardware.com
January 6, 2025 at 12:40 AM
gpapad.bsky.social
#Intel on the Brink of Death | Culture Rot, Product Focus Flawed, Foundry Must Survive

#CPU #x86 #chip #manufacturing #AMD #TSMC #datacenter #AI #hyperscalers

semianalysis.com/2024/12/09/i...
Intel on the Brink of Death | Culture Rot, Product Focus Flawed, Foundry Must Survive
Intel’s board is incompetent and its horrible decisions over the decades are going to push it towards death. The decision to fire Pat Gelsinger, put in charge a CFO + career sales and marketing lea…
semianalysis.com
January 5, 2025 at 11:59 PM
gpapad.bsky.social
Why Chips Fail, And What To Do About It

#SDC #Synopsys #AI #Reliability #Siemens #automotive #Technology #Chip #redundancy #performance #power #failure

semiengineering.com/why-chips-fa...
Why Chips Fail, And What To Do About It
Improving reliability in semiconductors is critical for automotive, data centers, and AI systems.
semiengineering.com
January 3, 2025 at 8:08 PM
gpapad.bsky.social
#Intel terminates x86S initiative — unilateral quest to de-bloat #x86 instruction set comes to an end

#AMD #CPU #Arm #datacenters

www.tomshardware.com/pc-component...
Intel terminates x86S initiative — unilateral quest to de-bloat x86 instruction set comes to an end
A pivot here, a pivot there.
www.tomshardware.com
January 1, 2025 at 8:00 AM
gpapad.bsky.social
Co-packaged optics deliver high-speed connectivity to supercharge generative #AI computing

research.ibm.com/blog/co-pack...

#IBM #circuits #LLMs #EnergyEfficiency
Co-packaged optics can supercharge generative AI computing
IBM Research scientists are bringing the speed and capacity of fiber optics where they haven’t gone before: inside data centers and onto circuit boards.
research.ibm.com
December 10, 2024 at 6:31 AM
Reposted by George Papadimitriou
blabbermouthnet.bsky.social
IRON MAIDEN's NICKO MCBRAIN Announces Retirement From Touring: Tonight Will Be 'My Final Gig' With The Band blabbermouth.net/news/iron-ma...
IRON MAIDEN's NICKO MCBRAIN Announces Retirement From Touring: Tonight Will Be 'My Final Gig' With The Band
IRON MAIDEN drummer Nicko McBrain has announced his retirement from touring. The 72-year-old musician, who has been a member of MAIDEN for more than four decades, revealed his decision to "step back" ...
blabbermouth.net
December 7, 2024 at 2:08 PM
gpapad.bsky.social
Next-gen #Nvidia #GPU "Rubin" is ahead of schedule, uses 3nm manufacturing and HBM4

www.techspot.com/news/105852-...

#AI #HBM
Next-gen Nvidia GPU "Rubin" is ahead of schedule, uses 3nm manufacturing and HBM4
According to the Taiwan Economic Daily, Nvidia's Rubin series AI graphics chips will be available in the second half of 2025, half a year earlier than previously...
www.techspot.com
December 7, 2024 at 3:19 PM
gpapad.bsky.social
#TSMC is reportedly in talks with #Nvidia to make Blackwell #GPUs in Arizona — Blackwell #silicon needs to be shipped back to Taiwan for assembly

www.tomshardware.com/pc-component...

#AI #Fab21
TSMC is reportedly in talks with Nvidia to make Blackwell GPUs in Arizona — Blackwell silicon needs to be shipped back to Taiwan for assembly
TSMC could move production of some of Nvidia's Blackwell GPUs to America.
www.tomshardware.com
December 5, 2024 at 10:21 PM
gpapad.bsky.social
#Intel 18A node reportedly stuck at 10% yields, SRAM density also trails #TSMC upcoming 2nm tech www.techspot.com/news/105844-... #CPU #AI #manufacturing
Intel 18A node reportedly stuck at 10% yields, SRAM density also trails TSMC upcoming 2nm tech
Recent reports indicate that Intel is facing significant yield challenges with its 18A node, potentially delaying its mass production timeline. According to the South Korean newspaper Chosun,...
www.techspot.com
December 5, 2024 at 9:59 PM