AMD Ryzen 7 PRO 9745 vs Intel Xeon 6724P Comparison

The Intel Xeon 6724P and AMD Ryzen 7 PRO 9745 occupy different corners of the server/workstation market, and the benchmark data reflects a stark division of labor. The Xeon 6724P, a 16-core Granite Rapids part, dominates nearly every multi-threaded and compute-intensive workload in this comparison, while the Ryzen 7 PRO 9745, an 8-core Zen 5 part, claims a decisive victory in single-threaded performance. The data shows a 9-to-2 split in benchmark wins, but the margins tell a more nuanced story about which processor suits which task.

Head-to-Head Benchmarks

The most lopsided result in the entire comparison is in the PassMark physics test, where the Intel Xeon 6724P scores 5004 against the AMD Ryzen 7 PRO 9745's 2026. That is a 147% advantage for Intel, the largest delta of any benchmark in this matchup. The physics workload is highly parallel and scales with core count and memory bandwidth, both of which favor the Xeon's 16 cores and eight-channel DDR5 interface.

In floating-point math, the Xeon again demonstrates overwhelming superiority. Its score of 135404 versus 82461 for the Ryzen represents a 64.2% lead. This is a core-heavy workload, and the Xeon's 32 threads provide more than double the parallelism of the Ryzen's 16 threads. Similarly, in the find prime numbers test, the Xeon scores 372 versus 188, a 97.9% advantage that reflects the same thread-count disparity.

Data encryption is another area where the Xeon runs away with the result. Its 34332 score beats the Ryzen's 22848 by 50.3%. This workload benefits from the Xeon's larger L3 cache (72 MB shared versus 32 MB shared) and its higher memory bandwidth of 409.6 GB/s compared to 89.6 GB/s. Extended instructions follow suit, with the Xeon at 54101 versus 36800, a 47% margin.

The multi-threaded PassMark score, which aggregates many workloads, shows the Xeon at 51345 versus 38201 for the Ryzen, a 34.4% lead. Integer math tells a similar story: 172216 versus 124167, a 38.7% advantage. Data compression favors the Xeon at 627185 versus 457211, a 37.2% difference, while random string sorting shows a 39.1% gap (68477 versus 49219).

The single bright spot for AMD comes in the single-thread tests. The Ryzen 7 PRO 9745 scores 4624 in PassMark single-thread, while the Xeon scores 3279. That is a 29.1% advantage for AMD, and it is the only category where the Ryzen wins. The Ryzen's higher boost clock of 5.40 GHz versus 4.30 GHz for the Xeon, combined with its newer 4 nm process node from TSMC, explains this outcome.

Across the eleven head-to-head benchmarks, the Intel Xeon 6724P wins nine, with margins ranging from 34.4% to 147%. The AMD Ryzen 7 PRO 9745 wins two (the two identical single-thread entries), with a margin of 29.1%.

The Verdict

The data presents a clear choice based on workload type. For any task that scales with core count, the Intel Xeon 6724P is the definitive winner. Its multi-threaded benchmark victories are not marginal; they are substantial, often exceeding 50% and reaching as high as 147% in physics simulations. The Xeon's 16 cores and 32 threads, paired with 72 MB of L3 cache and eight-channel memory, make it the superior processor for render farms, scientific computing, and heavily parallel server workloads.

The AMD Ryzen 7 PRO 9745 wins decisively where single-thread performance matters most. Its 29.1% lead in single-thread tests means it will feel snappier in lightly threaded applications, database queries that are latency-bound, or any workload that cannot utilize more than a few cores. The Ryzen also offers a much lower thermal envelope at 65 W TDP versus 210 W for the Xeon, which can be a deciding factor in dense deployments or power-constrained environments.

Benchmark results indicate that the Xeon 6724P sits at the 94th percentile of all CPUs, while the Ryzen 7 PRO 9745 sits at the 95th percentile. Their average benchmark scores are close — 72396 for Intel versus 74761 for AMD — but this aggregate masks the divergent strengths. The Xeon's nearest rivals include the Intel Xeon 6517P (0.1% ahead) and Intel Core Ultra 7 265KF (0.7% behind), showing it is competitive within Intel's own lineup. The Ryzen's nearest rivals include the AMD Ryzen 9 9950X (0.2% ahead) and AMD EPYC 4545P (0.8% ahead), placing it firmly in the upper tier of AMD's offerings.

For a server that runs many concurrent virtual machines or batch processing jobs, the Xeon 6724P is the data-backed choice. For a workstation running interactive applications or single-threaded engineering tools, the Ryzen 7 PRO 9745 delivers better responsiveness per the benchmarks.

Architecture Differences

The two processors are built on fundamentally different architectures. The Intel Xeon 6724P uses the Granite Rapids microarchitecture, specifically the Xeon 6 (Granite Rapids-SP) generation, fabricated on Intel's 5 nm process. The AMD Ryzen 7 PRO 9745 uses the Granite Ridge microarchitecture (Zen 5), part of the Ryzen 7 9000 series, fabricated on TSMC's 4 nm process.

Core configurations differ significantly. The Xeon offers 16 cores and 32 threads, while the Ryzen offers 8 cores and 16 threads — exactly half. This core disparity is the primary driver of the multi-threaded benchmark results. Clock speeds also favor AMD on the boost side: the Ryzen boosts to 5.40 GHz versus 4.30 GHz for the Xeon, though the Xeon has a slightly higher base clock of 3.60 GHz versus 3.80 GHz for the Ryzen.

Cache hierarchies are built for different purposes. The Xeon has 112 KB of L1 cache per core and 2 MB of L2 per core, with a massive 72 MB of shared L3 cache. The Ryzen has 80 KB of L1 per core, 1 MB of L2 per core, and 32 MB of shared L3. The Xeon's larger L3 cache supports its memory-intensive server workloads.

Memory subsystems diverge sharply. The Xeon supports DDR5 over an eight-channel bus, delivering 409.6 GB/s of bandwidth. The Ryzen also supports DDR5 but over a dual-channel bus, capping at 89.6 GB/s. This 4.6x difference in memory bandwidth is a critical factor in the Xeon's dominance in data-heavy benchmarks. Both processors support ECC memory, and both use PCIe Gen 5, though the Xeon offers 88 lanes versus 24 lanes for the Ryzen.

The Ryzen includes integrated Radeon Graphics, while the Xeon has no integrated graphics. The Ryzen's transistor count is listed at 8,315 million on a 70.6 mm² die, while the Xeon's transistor count and die size are not specified in the data. The Xeon uses Intel Socket 4710, while the Ryzen uses AMD Socket AM5. Both are actively in production, with the Xeon released on 2025-02-23 and the Ryzen on 2025-09-15.

FAQ

Q: Which processor has a higher single-thread performance?

A: The AMD Ryzen 7 PRO 9745 wins both PassMark single-thread benchmarks with a score of 4624, compared to 3279 for the Intel Xeon 6724P, representing a 29.1% advantage.

Q: How much faster is the Intel Xeon 6724P in multi-threaded workloads?

A: The Xeon scores 51345 in PassMark multi-thread versus 38201 for the Ryzen, a 34.4% lead. In the physics test, the lead is even larger at 147% (5004 versus 2026).

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

A: The Intel Xeon 6724P supports eight-channel DDR5 with 409.6 GB/s bandwidth, while the AMD Ryzen 7 PRO 9745 supports dual-channel DDR5 with 89.6 GB/s. The Xeon offers roughly 4.6 times the memory bandwidth.

Q: Which processor has more cores and threads?

A: The Intel Xeon 6724P has 16 cores and 32 threads. The AMD Ryzen 7 PRO 9745 has 8 cores and 16 threads, exactly half.

Q: What are the TDP ratings for each processor?

A: The Intel Xeon 6724P has a TDP of 210 W, while the AMD Ryzen 7 PRO 9745 has a TDP of 65 W. The Ryzen consumes significantly less power.

Q: Do both processors support ECC memory?

A: Yes, both the Intel Xeon 6724P and the AMD Ryzen 7 PRO 9745 have ECC memory support enabled.

Where Each One Wins

The Intel Xeon 6724P wins in all multi-threaded and compute-heavy categories. It is the clear choice for data compression workloads, scoring 627185 versus 457211. It wins data encryption by 50.3%, extended instructions by 47%, and floating-point math by 64.2%. The Xeon's 147% lead in physics simulations makes it the obvious pick for scientific and engineering simulation workloads. Its 97.9% advantage in prime number finding suggests it excels at integer-based mathematical computations. The Xeon's multithread score of 51345 versus 38201 confirms its superiority in any workload that can use more than 16 threads.

The AMD Ryzen 7 PRO 9745 wins exclusively in single-threaded performance. Its 4624 score versus 3279 for the Xeon means it will outperform in applications that are latency-sensitive and cannot parallelize across many cores. This includes many legacy enterprise applications, certain database operations, and interactive desktop use. The Ryzen's lower TDP of 65 W also makes it the better choice for fanless or low-power server designs, though the benchmark data does not directly measure thermal efficiency.

Given the 9-to-2 split in wins, the Xeon 6724P is the dominant processor for server workloads that demand throughput. The Ryzen 7 PRO 9745 is the specialist for single-threaded responsiveness, offering a 29.1% edge that could be decisive in specific applications. The choice between them should be driven entirely by whether the primary workload is parallel or serial in nature.