AMD Exec: Intel’s Multithreading Misstep Will ‘Help Us Gain Even More Market Share’

Why It Matters

Intel’s multithreading omission creates a clear competitive opening for AMD, potentially accelerating its market‑share gains in data‑center CPUs and influencing enterprise total‑cost‑of‑ownership calculations.

Key Takeaways

• Intel's upcoming Xeon Diamond Rapids likely lacks SMT.
• AMD expects performance gap to widen with Venice EPYC launch.
• AMD server share rose to 28.8% Q4, record high.
• Multithreading boosts cloud workloads up to 50% performance.
• ARM-based CPUs add competitive pressure to AMD and Intel.

Pulse Analysis

Intel’s decision to postpone simultaneous multithreading on its Diamond Rapids Xeon line signals a strategic pivot that could reverberate across the data‑center market. While Intel’s CEO Lip‑Bu Tan has pledged to re‑introduce SMT in later generations, the immediate absence leaves a performance gap that AMD can exploit. AMD’s upcoming Venice EPYC chips, built on the Zen architecture, retain SMT and promise higher throughput for multi‑threaded workloads, positioning the company as the default choice for cloud providers and enterprises that prioritize raw performance and efficiency.

The market impact is already evident. Mercury Research reports AMD’s server‑CPU shipment share climbed to 28.8% in the fourth quarter, up 3.1 percentage points year‑over‑year, while its revenue share reached 41.3% of the server segment. Beyond raw performance, SMT delivers tangible cost benefits: many enterprise software licenses are core‑based, so two logical threads per core effectively double licensing efficiency with minimal silicon overhead. Analysts estimate up to a 50% performance uplift on typical cloud workloads, translating into lower total‑cost‑of‑ownership for customers and a compelling value proposition for AMD’s sales teams.

However, the competitive landscape is widening. Arm‑based offerings from AWS, Nvidia’s upcoming Vera CPU, and Arm’s own AGI processor introduce alternative paths for hyperscalers seeking custom silicon. AMD counters this by emphasizing collaborative engineering and custom silicon services, aiming to become the "single most collaborative silicon provider" for hyperscale customers. As the industry balances performance, power, and supply‑chain resilience, AMD’s ability to leverage Intel’s SMT gap while navigating emerging Arm competition will shape data‑center CPU dynamics through 2027.