How Intel’s 13-TOPS NPU Might Find Purpose in a World of Copilot+ PCs

How Intel’s 13-TOPS NPU Might Find Purpose in a World of Copilot+ PCs

Analyst(s): Olivier Blanchard
Publication Date: April 11, 2025

Intel’s Core Ultra 200HX processors use their built-in NPU to offload AI assistant workloads during gaming, showcasing up to a 14.8% frame rate boost in Black Myth: Wukong. While not meeting Copilot+ PC thresholds, Intel’s chip showed it could assist users with game hints and boss fights in real-time through voice-based AI tools, and this niche use case could open up new markets for Intel’s sub-Copilot+ processors.

What is Covered in this Article:

  • Intel introduces Core Ultra 200HX and 200H mobile processors with integrated NPU.
  • AI assistant helps gamers defeat tough bosses in Black Myth: Wukong using voice or text input.
  • Intel claims up to 14.8% FPS improvement when AI assistant runs on the NPU instead of GPU.
  • Core Ultra 9 275HX also delivers up to 40% better power efficiency than its predecessor at 55W.
  • Intel’s AI performance with these processors remains limited by its 13 TOPS ceiling, far below Copilot+ PC standards.

The News: At a recent event in China, Intel officially unveiled its Core Ultra 200H and 200HX mobile processors—hardware tailored for gamers and power users running demanding workloads on the go. What stands out with these chips is the integration of a Neural Processing Unit, or NPU, which handles local AI-driven tasks on the device to boost performance without relying on cloud-based processing.

In a demo centered around the title Black Myth: Wukong, Intel illustrated how offloading AI assistant tasks from the GPU to the NPU had a tangible impact on performance. Specifically, the frame rate jumped from 94 FPS to 108 FPS – an improvement of roughly 14.8%. The AI assistant, which interacts through both voice and text, provides real-time help during gameplay, offering boss strategies or timely in-game advice. It effectively acts as a kind of digital co-pilot for players.

How Intel’s 13-TOPS NPU Might Find Purpose in a World of Copilot+ PCs

Analyst Take: Intel’s latest Core Ultra 200HX processors underscore the company’s broader ambition to bring locally processed AI into everyday gaming. With 13 TOPS NPU on board, the goal seems to be closing the gap between performance and smart features. That means higher frame rates, yes, but also a more responsive, hands-on gaming experience. Still, there are trade-offs. While the demo shows promise, performance limits, and strategic market positioning will play a major role in whether this feature truly shifts the needle for gamers and creators. Also worth keeping in mind is that with 18A and Panther Lake set to enter the market as early as H1 2026, if these processors are perceived as little more than short-term stopgaps between now and next year, gamers may opt to just wait until 2026 to upgrade their systems.

Frame Rate Gains With the NPU and Power Efficiency

Nevertheless, what Intel is showcasing here is a relatively straightforward product with clear and immediate value: By offloading specific agentic workloads from the GPU to the NPU, the system can free up more of the GPU’s horsepower to focus on rendering frames, translating into helpful FPS boosts and better overall UX for players.

Another subtle win with this system is power efficiency. According to the company, the Core Ultra 9 275HX offers up to 40% greater efficiency compared to its predecessor, the i9-14900HX, when both are running at 55W. Translated into real-world value, that kind of leap could mean longer gaming sessions without thermal throttling or battery drain.

Performance gains were also highlighted across a few other benchmarks. Intel notably pointed to an 8% improvement in Counter-Strike 2 and up to a 31% jump in POV-Ray, although naturally, those results will fluctuate depending on the nature of the task. Some workloads will benefit more than others, but the trend points upward.

AI Gaming Assistant Functionality

The built-in assistant can be triggered via voice commands or text inputs, and it’s designed to offer helpful suggestions during gameplay. That might include strategy tips, explanations of enemy behavior, or guidance on how to tackle tricky bosses. While it’s not yet compatible with every title, Intel showed it already working seamlessly within Black Myth: Wukong.

In many ways, this mirrors NVIDIA’s Project G-Assist – though Intel seems to be focusing more on native, on-device performance rather than cloud-powered solutions. Whether this assistant will become a staple across games remains to be seen, but as a feature demo, it signals Intel’s intent to play in the interactive AI space and delivers yet another concrete use case for on-device AI capabilities, even outside of the Copilot+ ecosystem.

NPU Performance Lags Behind Rivals

Of course, there’s a catch. Despite the new capabilities, this Intel NPU doesn’t meet the Microsoft Copilot+ PC standard, which demands a minimum of 40 TOPS. In contrast, AMD’s Strix Point Ryzen AI 9 HX 370 hits 50 TOPS, putting this particular processor at a numerical disadvantage in the Copilot+ arms race.

The most interesting point this raises for Intel, at least this year, is whether or not AI PC processors that fall short of achieving minimum Copilot+ performance thresholds have a place in the market. Obviously, Intel’s answer here is that they do. Not all AI PCs need to meet the Copilot+ standard – at least not until agentic AI software catches up to the category’s on-device capabilities. With gaming PCs powered by Core Ultra 200HX and 200H mobile processors, Intel points to lean, purpose-built use cases for AI-enabled PCs at more accessible price points.

Granted, one could argue that this feels like a solution looking for a problem, but I do believe that Intel might be onto something here. While Copilot+ class processors may be all the rage and a case can easily be made in favor of future-proofing for AI by buying early into systems powered by an AI-enabled processor capable of delivering 40 TOPS on the NPU, a lot of PC users might not care that much about agentic AI outside of very niche use cases – like gaming. If anything, what Intel is delivering here is a valuable market test that will help determine the viability of sub-Copilot+ AI-enabled processors, so long as their value is clearly defined for a well-defined audience.

In the grand scheme of things, 13 TOPS on the NPU feels more like a stepping stone than a destination, and that single spec may underwhelm customers. But if enough of them buy into the notion that a lean, purpose-built system is a better value for them than a pricier, high performance but general use system, this exercise may yet pay off for Intel and its PC OEM partners, at least through 2025.

What to Watch:

  • Intel’s NPU delivers useful gaming assistance, but limited performance could hinder broader adoption.
  • Copilot+ PCs set a higher benchmark that Intel does not currently meet with its 13 TOPS AI hardware.
  • Competitive pressure from AMD and Nvidia may escalate as AI capabilities become central to laptop differentiation.
  • Intel’s success will depend on broader game developer adoption of AI assistant features.

Read more about Intel Ultra Core on Intel’s website.

Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.

Other insights from The Futurum Group:

Dell’s Innovative Approach with AI PCs – Six Five On The Road

HP’s New EliteBook Series Brings AI to the Forefront of Business Laptops

AI PCs: Key Considerations for Procurement – Six Five On the Road

Image Credit: Intel

Author Information

Olivier Blanchard

Research Director Olivier Blanchard covers edge semiconductors and intelligent AI-capable devices for Futurum. In addition to having co-authored several books about digital transformation and AI with Futurum Group CEO Daniel Newman, Blanchard brings considerable experience demystifying new and emerging technologies, advising clients on how best to future-proof their organizations, and helping maximize the positive impacts of technology disruption while mitigating their potentially negative effects. Follow his extended analysis on X and LinkedIn.

SHARE:

Latest Insights:

Synopsys Deepens NVIDIA Collaboration to Accelerate EDA Workloads on Grace Blackwell Platform
Richard Gordon, VP & Practice Lead, Semiconductors at The Futurum Group, examines how Synopsys and NVIDIA aim to accelerate chip design with Grace Blackwell, targeting 30x EDA speedups and enhanced AI productivity.
Custom Arm Neoverse V2 Chip Posts Gains in AI, HPC, and General Compute Across C4A VMs
Richard Gordon, VP & Practice Lead, Semiconductors at The Futurum Group, unpacks Google Axion’s strong benchmarks across AI, HPC, and cloud workloads, showing how Google’s custom Arm CPU could reshape enterprise infrastructure.
Intel’s New CEO Lip-Bu Tan Spotlighted Physical AI and Its Importance at Intel Vision in Providing First Glimpses of His Unfolding Strategy
Ron Westfall, Research Director at The Futurum Group, shares insights on why Intel possesses the key portfolio building blocks, such as AI accelerators, edge computing expertise, and a legacy of powering complex systems, to become an integral player in the nascent Physical AI market.
Google’s AI Ambitions Expand Beyond the Cloud as It Builds an Interoperable Ecosystem for Agent Communication, Application Management, and On-Premises AI Deployment
Nick Patience, AI Practice Lead at The Futurum Group, shares his insights - along with those of colleagues - on Google’s Cloud Next 25 event, held last week in Las Vegas.

Book a Demo

Thank you, we received your request, a member of our team will be in contact with you.