Anton Shilov

Anton Shilov

Contributing Writer

Anton Shilov has been in the PC industry since 1990s playing games, building PCs, and writing stories about pretty much everything that relates to PCs, Macs, smartphones, tablets, and even fab equipment. Over his career, he has worked at a variety of high-ranking websites, including AnandTech, EE Times, TechRadar, X-bit labs, and now Tom's Hardware. When Anton is not reading or writing about something high-tech, he is probably watching a good movie, playing a video game, or spending time with his family.

Articles by: Anton Shilov

Intel

Intel Xeon chief architect leaves just 8 months after appointment — Ronak Singhal latest departure in ongoing shakeup

By Anton Shilov published

Ronak Singhal, Intel Xeon chief architect, departs from the company just eight months after his appointment and after 28 years at the company.

Kioxia LC9 123TB SSD

Nvidia and Kioxia target 100 million IOPS SSD in 2027 — AI server drives aim to deliver 33 times more performance

By Anton Shilov published

Kioxia is working with Nvidia to develop a 100 million IOPS SSD for AI servers by 2027, likely leveraging unconventional NAND architectures.

Nvidia H200 Hopper

Top China silicon figure calls on country to stop using Nvidia GPUs for AI — says current AI development model could become 'lethal' if not addressed

By Anton Shilov published

Develop AI-specific ASICs instead.

A19 Pro

Apple's A19 Pro beats Ryzen 9 9950X in single-thread Geekbench tests — iPhone 17 Pro chip packs 11-12% CPU performance bump, GPU performance up 37% over predecessor

By Anton Shilov published

Apple's latest A19 Pro offers an 11% - 12% higher CPU performance compared to its predecessor and even beats desktop-class CPUs. However, the main advantage of the A19 Pro is its GPU.

Huawei

China's chip champions ramp up production of AI accelerators at domestic fabs, but HBM and fab production capacity are towering bottlenecks

By Anton Shilov published

Millions of China-made AI accelerators are incoming from multiple companies in 2025 - 2026, but they may not be enough to meet the performance demands of local AI companies.

ASML

ASML makes $1.5 billion investment in Mistral AI — ASML becomes the largest shareholder for the French AI start-up

By Anton Shilov published

ASML has invested €1.3 billion in s Mistral AI allegedly in a strategic move to deepen its use of AI across lithography and chipmaking software.

Intel

Intel could sell up to 49% of its foundry business to external investors, but a full IPO or spin-off is unlikely

By Anton Shilov published

News Analysis But would it make any sense?

Intel

Intel CFO confirms that 14A will be more expensive than 18A due to High-NA EUV tool — Intel expects 14A process to offer 15-20% better performance-per-watt or 25-35% lower power consumption than 18A

By Anton Shilov last updated

Intel's 14A process, designed for both internal and foundry use, will be significantly more expensive than 18A due to its reliance on ASML's $380 million High-NA EUV tools.

Micron

U.S. government considers annual permits for Samsung and SK hynix to supply equipment to their Chinese fabs

By Anton Shilov published

What about TSMC?

AMD

AMD lists mystery Radeon 9060 XL model in ROCm documents, but it is more likely to be a typo than a new SKU

By Anton Shilov published

AMD's ROCm 6.4.2 software stack lists a non-existent Radeon RX 9060 XL GPU, likely a typo confusing internal 'Navi 44 XL' die naming with official product branding. But who knows?

Nvidia Grace Hopper superchips

Nvidia says 'We never deprive American customers in order to serve the rest of the world' — company says GAIN AI Act addresses a problem that doesn't exist

By Anton Shilov published

Nvidia says that the U.S. legislators' attempt to restrict exports unnecessarily is counterproductive, as the U.S. is already its top priority.

Broadcom

OpenAI widely thought to be Broadcom's mystery $10 billion custom AI processor customer — order could be for millions of AI processors

By Anton Shilov published

That's a lot of hardware.

Synopsys

Synopsys adds generative AI for chip development with Synopsys.ai Copilot design software

By Anton Shilov published

Having designed an AI-enhanced EDA toolset, Synopsys now adds Assistive and Creative capabilities for chip designers that can accelerate chip design across all phases.

ASML

China injects tens of billions of dollars in chipmaking tools, but it's easily more than a decade behind the market leaders — Here's why

By Anton Shilov published

China’s chip industry excels at mature nodes but remains at least a decade behind in advanced lithography.

AMD

AMD preps Mega Pod with 256 Instinct MI500 AI GPUs, Verano CPUs — Leak suggests platform with better scalability than Nvidia will arrive in 2027

By Anton Shilov published

First details about AMD's next-generation AI rack solution emerge: three racks, 64 compute racks, 256 Instinct MI500-series GPUs, 64 EPYC 'Verano' CPUs, 18 switch trays.

Nvidia Hopper H100 die shot

'Doomer science fiction': Nvidia criticizes proposed US bill designed to give American buyers 'first option' in AI GPU purchases before selling chips to other countries — GAIN AI Act debuts in defense spending bill

By Anton Shilov last updated

American buyers should be prioritized ahead of others, according to U.S. legislators

ASML

ASML and SK hynix assemble industry-first 'commercial' High-NA EUV system at fab in South Korea

By Anton Shilov published

Initially for R&D use, until later this decade where it will transition to production

TSMC

U.S. gov't revokes TSMC's authorization to ship tools to its fabs in China — special export license to be pulled by end of 2025

By Anton Shilov published

But TSMC vows to continue making chips on the mainland.

TSMC building

TSMC to reportedly raise quotes on advanced process nodes by up to 10% next year to pay for new fabs

By Anton Shilov published

Premium TSMC will raise 3nm–5nm chip prices by up to 10% in 2026 to fund new fabs and sustain margins as it expands 2nm production.

Intel

Intel files patent for 'Software Defined Supercore' — increases single-thread performance and IPC by mimicking ultra-wide execution using multiple cores

By Anton Shilov published

Intel's Software Defined Supercore (SDC) lets multiple CPU cores fuse into a virtual high-performance core to boost single-thread speed without building wider physical cores.

Intel Raptor Lake

Intel admits its high-end desktop PC chips 'fumbled the football,' disusses 18A yield challenges and performance, Panther Lake ramp

By Anton Shilov published

Intel sees performance of Panther Lake processors as competitive, but it admits lower-than-desired 18A yields and weak offerings on desktop and data center markets.

AMD Epyc processor performance

AMD's SP7 platform could enable CPUs with up to 1,400W of peak power consumption — chillers tested to keep heat in check

By Anton Shilov published

CPUs are going the way of GPUs?

Intel

Intel, Samsung, and SK hynix hit by another abrupt US policy change — government revokes waivers for advanced chipmaking tools at companies' China-based fabs

By Anton Shilov published

They face a 120 day deadline

AMD

According to a Linkedin profile, AMD is working on another chiplet-based GPU — UDNA could herald the return of 2.5D/ 3.5D chiplet-based configuration

By Anton Shilov published

AMD's upcoming RDNA 5 (Navi 5x) GPUs could mark the company's first consumer graphics architecture to use a multi-tile GPU design.

AMD

AMD details how it built a product line-up with just two RDNA 4 dies — Flexible design and asymmetric harvesting enables production of multiple models without new silicon

By Anton Shilov published

Premium AMD details how it can sell multiple SKUs without the need for more silicon

Other versions of this page are available with specific content for the following regions: