From chipsandcheese.com
Inside the AMD Radeon Instinct MI300A's Giant Memory Subsystem
12 12
AMD acquired ATI in 2006, hoping ATI's GPU expertise would combine with AMD's CPU know-how to create integrated solutions worth more than the sum of their parts.
#Bot #News #infinity #ycombinator #hackernewsbot #記憶體子系統 #amd #hackernews
on Sat, 11AM
From chipsandcheese.com
Skymont: Intel’s E-Cores reach for the Sky
10 10
The 2020s were a fun time for Intel’s Atom line, which went from an afterthought to playing a major role across Intel’s high performance client offerings.
#intel #hackernews #ycombinator #科技新聞
16h ago
From chipsandcheese.com
Inside the AMD Radeon Instinct MI300A’s Giant Memory Subsystem
2 2
AMD acquired ATI in 2006, hoping ATI’s GPU expertise would combine with AMD’s CPU know-how to create integrated solutions worth more than the sum of their parts. AMD uses the name ̶…
on Sat, 12PM
From chipsandcheese.com
An Interview with Susan Eickoff and Christian Jacobi from IBM at Hot Chips 2024
2 2
This is our second video interview from Hot Chips 2024 and this time I got to sit down with Susan Eickoff and Christian Jacobi from IBM to talk about their brand new Telum II processor and Spyre AI…
on Sep 5
From chipsandcheese.com
Nvidia's Announcements at CES 2025
0 1
Hello you fine Internet folks,
on Thu, 1AM
From chipsandcheese.com
Nvidia’s Announcements at CES 2025
0 1
Hello you fine Internet folks, Nvidia had a massive keynote at CES 2025 where they announced the RTX 50 series, the ConnectX-8 SuperNIC, and Project DIGITS which uses a new GB10 chip that was made …
on Wed, 11PM
From chipsandcheese.com
AMD’s Strix Halo – Under the Hood
0 1
Hello you fine Internet folks, At CES 2025 I got the chance to sit down with Mehash Subramony, AMD Senior Fellow, to talk about AMD’s upcoming Strix Halo SoC which is a brand new type of prod…
on Tue, 1AM
From chipsandcheese.com
AMD's Strix Halo - Under the Hood
0 2
Hello you fine Internet folks,
on Tue, 1AM
From chipsandcheese.com
0 0
Hello you fine Internet folks, AMD held a keynote at CES 2025 about their upcoming products in the form of dual CCD V-Cache for Zen 5 CPUs, Fire Range, Kraken, Strix Halo, along with the surprising…
on Jan 12
From chipsandcheese.com
0 0
Hello you fine Internet folks,Intel held a keynote at CES 2025 about their upcoming products in the form of Arrow Lake U, H, and HX, and a refresh of Intel N-series along with some news about Battl…
on Jan 9
From chipsandcheese.com
Digging into Driver Overhead on Intel’s B580
0 0
Hardware Canucks recently found that Intel’s new B580 GPU suffered when paired with older, weaker CPUs. When paired with the Ryzen 7 9800X3D, the B580 could compete against Nvidia’s RTX…
on Jan 8
From chipsandcheese.com
Digging into Driver Overhead on Intel's B580
0 0
Hardware Canucks recently found that Intel's new B580 GPU suffered when paired with older, weaker CPUs.
on Jan 8
From chipsandcheese.com
Analyzing Lion Cove’s Memory Subsystem in Arrow Lake
0 0
A CPU core can perform very differently depending on the platform it’s deployed in. That was on full display with Skymont’s wildly different performance characteristics in Lunar Lake an…
on Jan 6
From chipsandcheese.com
Analyzing Lion Cove's Memory Subsystem in Arrow Lake
0 1
A CPU core can perform very differently depending on the platform it's deployed in.
on Jan 6
From chipsandcheese.com
NeuReality: A Server on a PCIe Card
0 0
Hello you fine Internet folks,I made a mistake with our last article where I said that it would be our last bit of coverage on Supercomputing 2024. That was in fact incorrect, we have one more piec…
on Jan 4
From chipsandcheese.com
NeuReality: A Server on a PCIe Card
0 0
Or a DPU for AI accelerators
on Jan 4
From chipsandcheese.com
d-Matrix Corsair: 256GB of LPDDR for AI Models
0 0
With Microscaling Number Formats to boot!
on Jan 1
From chipsandcheese.com
d-Matrix Corsair: 256GB of LPDDR for AI Models
0 0
Hello you fine Internet folks,Today is the last of our Supercomputing 2024 coverage, but last does not mean least. Today we are going to be covering a new AI startup company called d-Matrix. d-Matr…
on Jan 1
From chipsandcheese.com
0 0
Hello you fine Internet folks, I had the opportunity to sit down and interview Bill Starke the Chief Architect of Power CPUs at IBM where we got to talk about the future of Power along with where h…
on Dec 27
From chipsandcheese.com
Skymont in Desktop Form: Atom Unleashed
0 0
Skymont is Intel’s newest E-Core architecture. E-Cores trace their lineage to low power and low performance Atom cores of long ago. But E-Cores have become an integral part of Intel’s h…
on Dec 21
From chipsandcheese.com
Skymont in Desktop Form: Atom Unleashed
0 0
Skymont is Intel's newest E-Core architecture.
on Dec 21
From chipsandcheese.com
Rebellions: From High Frequency Trading to AI Acceleration
0 0
Hello you fine Internet folks, At Supercomputing 2024 we stopped by the Rebellions Booth. Rebellions is a Korean startup that originally focused on the High Frequency Trading (HFT) sector and now i…
on Dec 17
From chipsandcheese.com
Rebellions: From High Frequency Trading to AI Acceleration
0 0
Hello you fine Internet folks,
on Dec 17
From chipsandcheese.com
Fujitsu's Monaka CPU: ARMv9, SVE2, and 3D Stacking
0 0
Designed for the datacenter and due in 2027
on Dec 13
From chipsandcheese.com
Fujitsu’s Monaka CPU: ARMv9, SVE2, and 3D Stacking
0 0
Hello you fine Internet folks, today we are going back to SC24 with a short about Fujitsu’s upcoming Monaka CPU. Fujitsu’s CPUs are quite prevalent in the HPC space with their prior gen A64FX CPU p…
on Dec 13
From chipsandcheese.com
Turning off Zen 4’s Op Cache for Curiosity and Giggles
0 0
CPUs start executing instructions by fetching those instruction bytes from memory and decoding them into internal operations (micro-ops). Getting data from memory and operating on it consumes power…
on Dec 11
From chipsandcheese.com
Turning off Zen 4's Op Cache for Curiosity and Giggles
0 0
CPUs start executing instructions by fetching those instruction bytes from memory and decoding them into internal operations (micro-ops).
on Dec 11
From chipsandcheese.com
Cannon Lake: Intel’s Forgotten Generation
0 0
Palm Cove (also commonly known as Cannon Lake) is Intel’s 10 nm die shrink of the Skylake core.
on Dec 9
From chipsandcheese.com
400G Omnipath is Coming: Cornelis Networks at SC24
0 0
Hello you fine Internet folks,
on Dec 8
From chipsandcheese.com
400G Omnipath is Coming: Cornelis Networks at SC24
0 0
Hello you fine Internet folks, At Supercomputing 2024 we stopped by the Cornelis Networks Booth where we got a peek at their CN5000 series of products which are their 400G line of of NICs, Switches…
on Dec 8
From chipsandcheese.com
Examining Intel’s Arrow Lake, at the System Level
0 0
Arrow Lake is the codename for Intel’s newest generation of high performance desktop CPUs. Its highest end offering, the Core Ultra 9 285K, implements eight Lion Cove P-Cores and 16 Skymont E…
on Dec 4
From chipsandcheese.com
Examining Intel's Arrow Lake, at the System Level
0 0
Arrow Lake is the codename for Intel's newest generation of high performance desktop CPUs.
on Dec 4
From chipsandcheese.com
An EPYC Exclusive for Azure: AMD's MI300C
0 0
The pun was just too good to resist.
on Dec 2
From chipsandcheese.com
An EPYC Exclusive for Azure: AMD’s MI300C
0 0
Hello you fine Internet folks, At SC24 we stopped by the Azure Booth to check out their new HBv5 VMs powered by the AMD EPYC 9v64H CPU. Each AMD EPYC 9v64H CPU physically have 96 Zen 4 cores along …
on Dec 2
From chipsandcheese.com
AMD Disables Zen 4’s Loop Buffer
0 0
A loop buffer sits at a CPU’s frontend, where it holds a small number of previously fetched instructions. Small loops can be contained within the loop buffer, after which they can be executed…
on Nov 30
From chipsandcheese.com
AMD Disables Zen 4's Loop Buffer
0 0
A loop buffer sits at a CPU's frontend, where it holds a small number of previously fetched instructions.
on Nov 30
From chipsandcheese.com
NextSilicon: Putting HPC First
0 0
Not an FPGA, not a GPU, but something different!
on Nov 28
From chipsandcheese.com
NextSilicon: Putting HPC First
0 0
Hello you fine Internet folks! At SC24, I got the chance to interview Elad Raz, CEO of NextSilicon which is a startup that is focusing on the HPC market which is unique considering that most of the…
on Nov 28
From chipsandcheese.com
Ayar Labs at Supercomputing 2024: Making Light Move Bits (YouTube Short)
0 0
Hello you fine Internet folks,
on Nov 25
From chipsandcheese.com
Ayar Labs at Supercomputing 2024: Making Light Move Bits (YouTube Short)
0 0
Hello you fine Internet folks,Today’s content from Supercomputing 2024 is from Ayar Labs. Ayar Labs is known for co-packaged optical interconnects with their TeraPHY chiplet which converts the elec…
on Nov 25
From chipsandcheese.com
Pushing AMD’s Infinity Fabric to its Limits
0 0
I recently wrote code to test memory latency under load, seeking to reproduce data in various presentations with bandwidth on the X axis and latency on the Y axis. Ampere pretty much described how …
on Nov 24
From chipsandcheese.com
Pushing AMD’s Infinity Fabric to its Limits
0 0
I recently wrote code to test memory latency under load, seeking to reproduce data in various presentations with bandwidth on the X axis and latency on the Y axis.
on Nov 24
From chipsandcheese.com
0 0
Hello you fine Internet folks! We are here at SC24 in Atlanta and there is a new TOP500 list that has just been announced and to celebrate the occasion, we have a video and an article for y’all and…
on Nov 20
From chipsandcheese.com
ARM or x86? ISA Doesn’t Matter
0 0
For the past decade, ARM CPU makers have made repeated attempts to break into the high performance CPU market so it’s no surprise that we’ve seen plenty of articles, videos and discussions about ARM’s effort, and many of these pieces focus on differences between the two instruction set...
on Nov 16
From chipsandcheese.com
Chips and Cheese's Microbenchmark Framework
0 0
A Call to Action for Code Contributors
on Nov 14
From chipsandcheese.com
Chips and Cheese’s Microbenchmark Framework
0 0
Hello you fine Internet folks! Back in May of this year, we had a Call to Action where we talked about setting up a new microbenchmark framework. This framework was meant to allow for standardizati…
on Nov 14