2024-08-31
AnandTech will stay online so readers can continue to access articles from our archive, and the forums will remain active to serve our community. Our sister site Tom's Hardware, will also continue to publish all the latest news, reviews and more from the PC world. Thank you all [image]
AnandTech
AnandTech shuts down after covering PC and mobile hardware for over 27 years; publisher Future plans to keep AnandTech's website and articles live indefinitely
Ryan Smith, with some very sad news: — It is with great sadness … Shawn Knight / TechSpot : AnandTech ends its 27-year run in online tech publishing John Callaham / Neowin : The ...
After 27 years of providing in-depth coverage of the amazing world of PC and mobile hardware, AnandTech is saying farewell. We want to thank everyone from the AnandTech community for their support and passion for what we've done over the years https://www.anandtech.com/...
AnandTech
AnandTech shuts down after covering PC and mobile hardware for over 27 years; publisher Future plans to keep AnandTech's website and articles live indefinitely
Ryan Smith, with some very sad news: — It is with great sadness … Shawn Knight / TechSpot : AnandTech ends its 27-year run in online tech publishing John Callaham / Neowin : The ...
2024-08-30
After 27 years of providing in-depth coverage of the amazing world of PC and mobile hardware, AnandTech is saying farewell. We want to thank everyone from the AnandTech community for their support and passion for what we've done over the years https://www.anandtech.com/...
AnandTech
AnandTech shuts down after covering PC and mobile hardware for over 27 years; publisher Future plans to keep AnandTech's website and articles live indefinitely
It is with great sadness that I find myself penning the hardest news post I've ever needed to write here at AnandTech.
AnandTech will stay online so readers can continue to access articles from our archive, and the forums will remain active to serve our community. Our sister site Tom's Hardware, will also continue to publish all the latest news, reviews and more from the PC world. Thank you all [image]
AnandTech
AnandTech shuts down after covering PC and mobile hardware for over 27 years; publisher Future plans to keep AnandTech's website and articles live indefinitely
It is with great sadness that I find myself penning the hardest news post I've ever needed to write here at AnandTech.
2024-06-14
Ahead of next week's retail launch of Snapdragon X laptops, this morning we have our deep dive into the new SoC's CPU and GPU architectures. At long last it's finally time to get to know the new Oryon CPU and Adreno X1 GPU, and see what makes them tick https://www.anandtech.com/... [image]
AnandTech
A deep dive into the Snapdragon X Elite/Plus architecture including Oryon, Qualcomm's custom-built Arm CPU cores, as first laptops with the SoC ship next week
Ryan Smith / AnandTech :
2023-11-13
Kicking off SC23, NVIDIA has a slate of server product announcements. Among these are the updated HBM3E-equipped H200 accelerator, new HGX boards, and a new supercomputer win - Jupiter - which will be the first NVIDIA HPC-grade exascale supercomputer http://www.anandtech.com/... [image]
AnandTech
Nvidia announces the H200, an update to its standalone H100 accelerator that swaps 5.3Gbps/pin HBM3 memory for 6.5Gbps/pin HBM3E memory and ships in Q2 2024
With faster and higher capacity HBM3E memory set to come online early in 2024, NVIDIA has been preparing its current-generation server GPU products to use the new memory.
2023-06-14
And finally, AMD will be producing a super-sized, GPU-only version of the MI300 accelerator, the MI300X. Using 24GB HBM3 stacks and 8 CDNA 3 GPU chiplets, the flagship chip will offer 192GB of local memory, ideal for AI/LLM training and other AI workloads https://www.anandtech.com/... [image]
Reuters
Amazon VP of EC2 Dave Brown says AWS is considering using AMD's new MI300 chips and declined to use Nvidia's DGX Cloud chips; Oracle is Nvidia's first partner
Amazon Web Services (AMZN.O), the world's largest cloud computing provider, is considering using new artificial intelligence chips …
And finally, AMD will be producing a super-sized, GPU-only version of the MI300 accelerator, the MI300X. Using 24GB HBM3 stacks and 8 CDNA 3 GPU chiplets, the flagship chip will offer 192GB of local memory, ideal for AI/LLM training and other AI workloads https://www.anandtech.com/... [image]
AnandTech
AMD previews its Instinct MI300 accelerator family, including the MI300X, a high-performance GPU with 192GB of HBM3 memory aimed squarely at the LLM market
Ryan Smith / AnandTech :
2023-06-06
This morning Intel is taking the wraps off of its PowerVia backside power delivery tech for Intel 20A & 18A process nodes. Intel's BS-PDN implementation is up and running in the labs, with a test CPU showing promising results ahead of manufacturing in 2024 https://www.anandtech.com/... [image]
AnandTech
Intel details PowerVia, a backside power delivery network technology promising performance and manufacturing benefits, set to be introduced in its chips in 2024
Ryan Smith / AnandTech :
2022-09-27
It's here at last! Today we're taking a look at AMD's new Ryzen 7000 processors. Based on the company's updated Zen 4 architecture and associated AM5 platform, AMD is bringing out DDR5, PCIe 5, and more as part of a push to retake the high-end CPU market https://www.anandtech.com/... https://twitter.com/...
AnandTech
Review of the $699 AMD Ryzen 9 7950X, built on Zen 4 with 16 cores: higher single- and multi-threaded performance and superb efficiency helps beat Intel's chips
AnandTech :
2022-08-02
The CPU interconnect wars are finally over - and CXL has won. This afternoon the OpenCAPI consortium is announcing that they're transferring everything to the CXL consortium, bowing-out and leaving CXL as the winning next-gen I/O standard for server CPUs https://www.anandtech.com/... https://twitter.com/...
AnandTech
The CXL consortium plans to absorb the OpenCAPI consortium's assets and standards, setting CXL to become the dominant server CPU-to-device interconnect standard
With the 2022 Flash Memory Summit taking place this week, not only is there a slew of solid-state storage announcements in the pipe … Source: GlobeNewswire News Room .
2022-06-18
Delivering a major update to their roadmap, TSMC today has revealed that they are developing a total of 5 versions of their N3 process. Joining N3 will be N3E, N3P, N3S, and N3X, giving more design options, as well as FinFlex tech for low-level flexibility https://www.anandtech.com/... https://twitter.com/...
AnandTech
TSMC details its roadmap: a high volume 3nm chip production starting in H2 2022, with five 3nm-class N3 node iterations, and 2nm-class N2 introduction in 2025
Anton Shilov / AnandTech :
2022-06-07
Apple this afternoon has taken the warps off of its second-generation Apple Silicon SoC for Macs: M2. With improved CPU cores, LPDDR5, and a larger 10-core GPU, the 20B transistor M2 has all the makings of a nice upgrade over the well-received M1 https://www.anandtech.com/... https://twitter.com/...
AnandTech
An in-depth look at Apple's M2 chip: 20B transistors, up from 16B in M1 and 15B in A15, alongside a 10-core GPU, up from an 8-core in M1, and 8K video decoding
Though primarily a software-focused event, Apple's WWDC keynotes are often stage for an interesting hardware announcement or two as well …
2022-05-20
Qualcomm this morning is announcing the Snapdragon 8+ Gen 1, their annual mid-gen flagship SoC refresh. Porting the original S8G1 to TSMC's superior 4nm process, this year's update promises both big power efficiency improvements, and good performance gains https://www.anandtech.com/... https://twitter.com/...
The Verge
Qualcomm unveils the Snapdragon 8 Plus Gen 1, says it will offer 10% faster CPU performance, 10% faster GPU clocks, and have up to 30% better power efficiency
Bragging rights (and battery life?) for gaming phones — Qualcomm's Snapdragon 8 Gen 1 set the stage for the biggest Android smartphones …
2022-03-23
NVIDIA has just announced their next-generation GPU architecture: Hopper. The first part out of the gate, the H100 server accelerator, is comprised of 80B transistors, and along with trebling tensor throughput, will draw an eye-popping 700 Watts of power https://www.anandtech.com/... https://twitter.com/...
VentureBeat
Nvidia announces its next-gen Hopper GPU architecture, optimized for AI workloads, and the first graphics card to use it, Hopper H100, which has 80B transistors
NVIDIA has just announced their next-generation GPU architecture: Hopper. The first part out of the gate, the H100 server accelerator, is comprised of 80B transistors, and along with trebling tensor throughput, will draw an eye-popping 700 Watts of power https://www.anandtech.com/... https://twitter.com/...
VentureBeat
Nvidia unveils the Grace CPU Superchip, an ARM-based discrete data center CPU with 144 high-performance cores and 1TB of secondary memory, coming in 2023
Nvidia offered details on its Grace central processing unit (CPU) “Superchip” during CEO Jensen Huang's keynote speech at its virtual Nvidia GTC 2022 event.
2022-03-10
Today @IanCutress is talking with Intel's Raja Koduri, SVP and GM of their Accelerated Computing Systems and Graphics (AXG) Group. Topics include Intel's Zettascale initiative, OneAPI, as well as where the frontman sees dedicated accelerators going next https://www.anandtech.com/... https://twitter.com/...
AnandTech
Q&A with Intel SVP and GM of Accelerated Computing Raja Koduri on Intel's Zettascale Initiative, zettascale architectural innovations, the metaverse, and more
Dr. Ian Cutress / AnandTech : Tweets: @anandtech . Thanks: @ryansmithat Tweets: @anandtech : Today @IanCutress is talking with Intel's Raja Koduri, SVP and GM of their Accelerated...
2022-03-09
Saving the best for last, Apple today has announced the fourth and final M1 SoC, the M1 Ultra. Packing two M1 Max dies on a single chip with a 2.5TB/sec interface for even more performance, the M1 Ultra is unlike any other SoC on the market today https://www.anandtech.com/... https://twitter.com/...
AnandTech
A deep dive into Apple's M1 Ultra: two M1 Max dies combined into a single chip, 800GB/second memory bandwidth, 64 GPU cores to outperform the RTX 3090, and more
As part of Apple's spring “Peek Performance” product event this morning, Apple unveiled the fourth and final member of the M1 family of Apple Silicon SoCs, the M1 Ultra.
2022-02-18
Intel discloses a new roadmap - the future of Intel Xeon, going out three generations. Bundled in the roadmap is going to be a new class of Xeon, featuring only DC-optimized E-cores, coming on the Intel 3 process node. Tip @Techmeme. Details 👇 https://www.anandtech.com/... https://twitter.com/...
AnandTech
Intel Xeon roadmap: Emerald Rapids, based on Intel 7 process, coming in 2023, Granite Rapids and new Efficient-core Sierra Forest in 2024, both built on Intel 3
It's no secret that Intel's enterprise processor platform has been stretched in recent generations.
2022-01-12
The PCI Express 6.0 spec has been ratified and released! @pci_sig's latest gen of PCIe tech is once again doubling bus bandwidth - now to 128GB/sec/direction for a x16 link - thanks to a combination of PAM4 signaling, forward error correction, and FLIT http://www.anandtech.com/... https://twitter.com/...
AnandTech
PCI Special Interest Group finalizes its PCIe 6.0 specification, expected in servers in 2023, doubling PCIe 5.0's speeds and reaching 128GBps in x16 slots
This morning the PCI Special Interest Group (PCI-SIG) is releasing the much-awaited final (1.0) specification for PCI Express 6.0.