AMD Ryzen 9 PRO 9965 vs Intel Xeon 678X Comparison

Where Each One Wins

The benchmark data splits cleanly along workload type. The Intel Xeon 678X dominates every multithreaded and throughput-oriented test in the head-to-head comparison, winning 9 of 11 recorded benchmarks. The AMD Ryzen 9 PRO 9965 takes exactly two: the single-thread tests, where it posts a 4682 score against the Xeon's 3758, a 19.7% advantage in the database's measurement.

The Xeon's wins are not marginal. In passmark_find_prime_numbers it scores 1010 versus 364, a 177.5% delta. Floating point math shows 362070 against 160746, a 125.2% gap. Physics simulation delivers 7809 versus 3256, 139.8% ahead. These are not close contests; they represent fundamentally different capability tiers.

Multithread performance, the aggregate metric, sits at 98559 for the Xeon versus 66655 for the Ryzen, a 47.9% lead. Data compression shows 1690896 against 908293, an 86.2% margin. Integer math lands at 405075 versus 243280, 66.5% ahead. Encryption, extended instructions, and random string sorting all follow the same pattern, with the Xeon leading by 86.1%, 98.6%, and 72.9% respectively.

The Ryzen's wins are exclusively in single-threaded throughput. That 4682 score places it in a different class for lightly threaded workloads, but it is the only area where the AMD part takes the lead. Every other recorded test belongs to Intel.

The Verdict

The data points to a clear use-case split. The Intel Xeon 678X is the choice for workloads that scale across many cores: rendering, simulation, data compression, encryption, and heavy floating-point math. Its 48 cores and 96 threads give it a structural advantage that shows up in every parallel benchmark. The 99th percentile ranking among all CPUs, with an average benchmark score of 193477, confirms its position near the top of the database.

The AMD Ryzen 9 PRO 9965 is the pick for single-threaded responsiveness and lightly threaded applications. Its 4682 single-thread score beats the Xeon's 3758 by a wide margin. The 98th percentile ranking and average score of 145728 show it is no slouch overall, but its 16 cores and 32 threads cannot match the Xeon's raw parallel throughput.

For a server or workstation running heavily threaded batch jobs, the Xeon 678X is the data-backed recommendation. For a workstation where single-thread performance matters most, or where the lower core count is sufficient, the Ryzen 9 PRO 9965 holds the advantage in the only tests it wins. Mixed workloads will favor the Xeon because the magnitude of its multithreaded leads dwarfs the Ryzen's single-thread edge.

Head-to-Head Benchmarks

The biggest win for the Intel Xeon 678X comes in passmark_find_prime_numbers. The Xeon scores 1010 against the Ryzen's 364, a 177.5% delta. This test is highly sensitive to core count and memory bandwidth, both areas where the Xeon's eight-channel DDR5 setup and 48 cores overwhelm the Ryzen's dual-channel configuration.

Physics simulation shows a 139.8% lead, 7809 versus 3256. Floating point math follows at 125.2%, 362070 versus 160746. Extended instructions deliver a 98.6% gap, 141431 versus 71210. These are the largest margins and indicate workloads that benefit from the Xeon's 192 MB shared L3 cache and massive thread count.

Data compression and encryption show 86.2% and 86.1% leads respectively. The Xeon's 1690896 compression score versus 908293, and 83598 encryption score versus 44920, suggest memory bandwidth plays a significant role. The Xeon's 409.6 GB/s memory bandwidth is substantially higher than the Ryzen's 89.6 GB/s, and these tests appear to exploit that difference.

Random string sorting gives the Xeon a 72.9% lead, 164102 versus 94915. Integer math shows 66.5%, 405075 versus 243280. The multithread aggregate test, passmark_multithread, gives a 47.9% lead, 98559 versus 66655. Even the closest multithreaded margin is substantial.

The AMD Ryzen 9 PRO 9965 wins passmark_single_thread and passmark_singlethread, both recording 4682 versus 3758, a 19.7% delta in AMD's favor. This is the only benchmark category where the Ryzen leads, and it reflects the higher boost clock of 5.50 GHz against the Xeon's 4.90 GHz, along with the Ryzen's newer 4 nm process node versus Intel's 5 nm.

FAQ

Q: Which CPU has the higher single-thread performance?

A: The AMD Ryzen 9 PRO 9965, with a score of 4682 in passmark_single_thread compared to the Intel Xeon 678X's 3758, a 19.7% advantage.

Q: How much faster is the Intel Xeon 678X in multithreaded workloads?

A: The Xeon leads by 47.9% in passmark_multithread, scoring 98559 versus the Ryzen's 66655. In the most extreme case, passmark_find_prime_numbers, the Xeon is 177.5% ahead.

Q: What drives the Xeon's large leads in memory-sensitive tests?

A: The recorded memory specifications differ significantly. The Xeon has eight-channel DDR5 with 409.6 GB/s bandwidth, while the Ryzen has dual-channel DDR5 with 89.6 GB/s. Tests like data compression and encryption, where the Xeon leads by roughly 86%, appear to reflect this bandwidth gap.

Q: Does the Ryzen 9 PRO 9965 win any benchmark?

A: Yes, it wins both single-thread tests recorded: passmark_single_thread and passmark_singlethread, each with a score of 4682 against the Xeon's 3758.

Q: How do the core counts compare?

A: The Intel Xeon 678X has 48 cores and 96 threads. The AMD Ryzen 9 PRO 9965 has 16 cores and 32 threads. The Xeon's threefold core advantage is consistent with its multithreaded benchmark dominance.

Q: Which CPU ranks higher overall in the database?

A: The Intel Xeon 678X sits in the 99th percentile of all CPUs with an average benchmark score of 193477. The AMD Ryzen 9 PRO 9965 sits in the 98th percentile with an average score of 145728.

Architecture Differences

The two processors come from different design philosophies. The Intel Xeon 678X uses the Granite Rapids architecture on a 5 nm process from Intel's own foundry. The AMD Ryzen 9 PRO 9965 uses the Granite Ridge codename, built on Zen 5 architecture at TSMC's 4 nm node.

Core counts diverge sharply. The Xeon packs 48 cores and 96 threads, while the Ryzen offers 16 cores and 32 threads. This explains the multithreaded performance gap. Cache hierarchies also differ. The Xeon has 112 KB of L1 and 2 MB of L2 per core, with 192 MB of shared L3. The Ryzen has 80 KB of L1 and 1 MB of L2 per core, with 64 MB of L3.

The Xeon's die size is listed as 2x 598 mm², while the Ryzen is 2x 70.6 mm². The Ryzen's transistor count is recorded at 16,630 million. The Xeon's transistor count is not recorded in the database.

Memory architecture is a major differentiator. The Xeon supports eight-channel DDR5 with a recorded bandwidth of 409.6 GB/s. The Ryzen supports dual-channel DDR5 with 89.6 GB/s. Both support ECC memory. PCIe lanes also differ: the Xeon provides 128 Gen 5 lanes from the CPU, while the Ryzen provides 24 Gen 5 lanes.

The Xeon has no integrated graphics. The Ryzen includes Radeon Graphics. Both target the server and workstation market segment. The Xeon's multiplier is unlocked, while the Ryzen's is not.

Specification Differences

The clock speeds differ notably. The Intel Xeon 678X has a base clock of 2.40 GHz and a boost clock of 4.90 GHz. The AMD Ryzen 9 PRO 9965 has a base clock of 4.30 GHz and a boost clock of 5.50 GHz. The Ryzen's higher clocks contribute to its single-thread win.

Power draw is recorded differently. The Xeon has a TDP of 300, while the Ryzen has a TDP of 170. This aligns with the Xeon's higher core count and memory bandwidth.

Sockets are incompatible. The Xeon uses Intel Socket 4710, while the Ryzen uses AMD Socket AM5. The Xeon's part number is SA2CX, the Ryzen's is 100-000002001.

The Xeon's cache configuration lists L1 at 112 KB per core, L2 at 2 MB per core, and L3 at 192 MB shared. The Ryzen lists L1 at 80 KB per core, L2 at 1 MB per core, and L3 at 64 MB. No 3D V-Cache is recorded for either.

Release dates differ. The Xeon was released on 2026-02-01, the Ryzen on 2026-06-29. The Xeon has a recorded launch MSRP of $3749, while the Ryzen has no launch MSRP recorded. The Xeon's production status is Active, and the Ryzen's is also Active.

The Xeon's die size is 2x 598 mm² versus the Ryzen's 2x 70.6 mm². The Ryzen's process node is 4 nm from TSMC, while the Xeon's is 5 nm from Intel. The Ryzen has integrated Radeon Graphics, the Xeon has N/A for integrated graphics.