AMD Ryzen 9 PRO 9965 vs Intel Xeon 676X Comparison

The Intel Xeon 676X and AMD Ryzen 9 PRO 9965 represent two fundamentally different interpretations of high-end x86 computing, and the benchmark data shows a stark division of labor. The Xeon 676X wins 9 of 11 head-to-head tests, often by massive margins, while the Ryzen 9 PRO 9965 claims only the single-threaded crown. The Xeon is the clear choice for multi-threaded, data-heavy server workloads, whereas the Ryzen is a better fit for lightly-threaded, latency-sensitive tasks where raw clock speed and per-core efficiency matter more than sheer core count.

The Verdict

The data points to two distinct buyers. The Intel Xeon 676X is the pick for anyone running heavily parallel, throughput-oriented workloads. Its lead in multithreaded tests is not incremental; it is overwhelming, with a 154.3% advantage in the physics test and a 102.7% lead in finding prime numbers. For databases, compression, or encryption pipelines that can use every thread available, the Xeon’s 32 cores and 64 threads provide a decisive edge. Its average benchmark score of 158540 places it in the 98th percentile of all CPUs, and it sits just 1.1% behind the AMD EPYC 9355P, a direct server competitor.

The AMD Ryzen 9 PRO 9965 is the pick for single-threaded performance and power-conscious environments. Its 5.50 GHz boost clock and 4.30 GHz base clock are the highest in this comparison, and it delivers a 14.2% better single-thread score. While its average benchmark score of 145728 is lower, it still lands in the same 98th percentile. This is a processor for workloads that are latency-bound or that rely on a few fast cores, not for scaling across many. Its 16 cores and 32 threads are half the Xeon's count, but its higher clock speeds mean it will feel snappier in tasks that cannot use more than one or two threads.

Architecture Differences

The two chips are built on different nodes and employ different design philosophies. The Intel Xeon 676X uses a 5 nm process fabricated by Intel itself, with a massive die size of 2x 598 mm². It is based on the Granite Rapids architecture, part of the Xeon 600 generation, and fits into the Intel Socket 4710. This is a server-class design with an eight-channel memory bus, delivering 409.6 GB/s of memory bandwidth. Its cache hierarchy is generous, with 144 MB of shared L3 cache and 2 MB of L2 per core.

The AMD Ryzen 9 PRO 9965 uses a smaller 4 nm process from TSMC, with a die size of just 2x 70.6 mm² and 16,630 million transistors. It is based on the Granite Ridge codename, part of the Ryzen 9 series with Zen 5 cores, and fits the AMD Socket AM5. Its memory bus is dual-channel, yielding 89.6 GB/s of bandwidth, a fraction of the Xeon’s. Its L3 cache is 64 MB, less than half the Xeon’s, and L2 is 1 MB per core. The Ryzen includes integrated Radeon Graphics, while the Xeon has none. The Xeon offers 128 PCIe Gen 5 lanes, while the Ryzen offers 24. Both support ECC memory and DDR5, but the platform differences could not be starker.

Where Each One Wins

The Xeon 676X wins every multi-threaded and computationally intensive test. In data compression, it scores 1,355,807 versus 908,293, a 49.3% lead. In data encryption, it leads by 50.6%. In floating-point math, the Xeon’s 283,570 score is 76.4% higher than the Ryzen’s 160,746. The pattern is consistent: the more parallel the workload, the larger the Xeon’s advantage. This is a processor for rendering farms, scientific simulation, and enterprise data processing where the answer is to throw more cores at the problem.

The Ryzen 9 PRO 9965 wins only the single-thread tests, with a score of 4,682 versus the Xeon’s 4,015, a 14.2% improvement. This is the only category where the Ryzen’s higher clock speeds and newer Zen 5 architecture translate into a win. This makes it suitable for legacy software that is poorly threaded, for real-time control applications, or for interactive workloads where a single thread's speed is the bottleneck. Its lower TDP of 170 watts versus the Xeon’s 275 watts also suggests it will be easier to cool and run in more compact systems, though the data does not directly test power efficiency.

FAQ

Q: Which CPU has a higher single-core performance?

A: The AMD Ryzen 9 PRO 9965 is faster in single-threaded tests, scoring 4,682 compared to the Intel Xeon 676X’s 4,015, a 14.2% difference.

Q: How much faster is the Intel Xeon 676X in multithreaded workloads?

A: The Xeon leads by 36.7% in the passmark multithread test (91,115 vs 66,655), but the gap is much wider in specific tasks. It is 154.3% faster in physics and 102.7% faster in finding prime numbers.

Q: Does the AMD Ryzen 9 PRO 9965 have integrated graphics?

A: Yes, it includes Radeon Graphics. The Intel Xeon 676X has no integrated graphics, meaning a discrete GPU is required for display output.

Q: What is the memory bandwidth difference between the two?

A: The Intel Xeon 676X has a substantial advantage, with 409.6 GB/s of bandwidth from its eight-channel memory bus. The AMD Ryzen 9 PRO 9965 offers 89.6 GB/s from its dual-channel bus.

Q: Which processor has more PCIe lanes?

A: The Intel Xeon 676X provides 128 PCIe Gen 5 lanes, while the AMD Ryzen 9 PRO 9965 provides 24 Gen 5 lanes.

Q: Are both processors in the same performance percentile?

A: Yes, both are in the 98th percentile of all CPUs, though the Xeon’s average score of 158540 is higher than the Ryzen’s 145728.

Head-to-Head Benchmarks

The largest win for the Intel Xeon 676X comes in the physics test, where it scores 8,281 against the Ryzen’s 3,256. That is a 154.3% advantage, more than double the competitor’s output. This suggests a massive difference in computational throughput for physics simulations or similar floating-point heavy tasks. The next biggest margin is in finding prime numbers, where the Xeon’s 738 score beats the Ryzen’s 364 by 102.7%. This test is often sensitive to integer performance and memory latency, and the Xeon’s larger cache and memory bandwidth likely contribute to the result.

In floating-point math, the Xeon’s 283,570 score is 76.4% better than the Ryzen’s 160,746. Data encryption shows a 50.6% lead for Intel (67,638 vs 44,920), and extended instructions show a 47.8% lead (105,231 vs 71,210). The Xeon also leads in data compression by 49.3% (1,355,807 vs 908,293), integer math by 45.8% (354,777 vs 243,280), and random string sorting by 45.4% (137,976 vs 94,915). The multithread test shows a 36.7% lead (91,115 vs 66,655), which is actually one of the smaller gaps, indicating that the Ryzen’s core efficiency is decent but cannot overcome the core count deficit.

The only wins for the AMD Ryzen 9 PRO 9965 are the two single-thread entries, both scoring 4,682 versus 4,015 for the Intel, a 14.2% improvement. This is a significant margin in single-threaded terms, but it is the only territory where AMD can claim victory. Across the board, the Xeon’s 9 wins versus the Ryzen’s 2 wins paint a clear picture of a dominant multi-threaded performer facing a strong single-threaded contender.

Specification Differences

The core count is the most obvious difference: the Intel Xeon 676X has 32 cores and 64 threads, while the AMD Ryzen 9 PRO 9965 has 16 cores and 32 threads. Clock speeds favor AMD, with the Ryzen boosting to 5.50 GHz and idling at 4.30 GHz, while the Xeon boosts to 4.90 GHz and has a 2.80 GHz base clock. The Xeon has a 275 TDP, and the Ryzen has a 170 TDP. The Xeon uses the Intel Socket 4710, and the Ryzen uses the AMD Socket AM5. The manufacturing process differs, with Intel on 5 nm and AMD on 4 nm, and the foundries are Intel and TSMC, respectively.

The cache configurations are also different. The Xeon has 112 KB of L1 per core, 2 MB of L2 per core, and 144 MB of shared L3. The Ryzen has 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. Memory bandwidth is heavily skewed toward Intel, with 409.6 GB/s versus 89.6 GB/s. The Xeon is an eight-channel memory design; the Ryzen is dual-channel. Both support ECC memory and DDR5. The Xeon offers 128 PCIe Gen 5 lanes, while the Ryzen offers 24. The Xeon has no integrated graphics; the Ryzen includes Radeon Graphics. The Xeon’s multiplier is unlocked, while the Ryzen’s is not. The Xeon was released earlier, in February 2026, while the Ryzen came later, in June 2026. The Xeon has a launch MSRP of $2499, while the Ryzen’s launch MSRP is not listed.