AMD Ryzen 9 PRO 9945 vs Intel Xeon 6736P Comparison

Where Each One Wins

The benchmark split is heavily one-sided, but the two wins for the AMD Ryzen 9 PRO 9945 are decisive in a way that matters for everyday responsiveness. The AMD part wins both single-thread tests, scoring 4619 in PassMark single-thread versus 2024 for the Intel Xeon 6736P, a 128.2% advantage. That is not a small gap; it means the AMD chip is more than twice as fast in workloads that rely on one core, such as application launching, light scripting, and general desktop interaction.

The Intel Xeon 6736P, by contrast, wins 9 of the 11 head-to-head benchmarks, and its victories are broad across both integer and floating-point work. The largest margins come in passmark_physics, where Intel scores 6531 versus 2902, a 55.6% lead, and in passmark_random_string_sorting, where Intel posts 103723 against 62458, a 39.8% advantage. Data compression also favors Intel heavily: 796658 versus 579972, a 27.2% gap. Encryption shows a similar pattern at 46236 versus 30450, which is 34.1% ahead.

The interesting case is passmark_multithread, where Intel wins but only by 2.8% (50072 versus 48664). That is the closest result in the entire comparison, and it suggests that in a generic all-core workload, the two processors are nearly equivalent despite Intel having three times the core count. The Intel part pulls ahead more decisively in tests that benefit from memory bandwidth or specific instruction throughput, such as extended instructions (55563 versus 44374, a 20.1% lead) and integer math (206833 versus 171175, a 17.2% lead).

So the use-case split is clear: the AMD Ryzen 9 PRO 9945 is the pick for single-thread latency and lightly threaded workloads, while the Intel Xeon 6736P dominates in sustained multi-core compute, data processing, and physics simulation. For a server handling many parallel requests, Intel is the obvious choice. For a workstation where a developer or engineer sits at the machine and runs mixed tasks, AMD's single-thread advantage will be felt constantly.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen 9 PRO 9945 is built on a 4 nm TSMC process, using the Granite Ridge architecture with Zen 5 cores. It has 12 cores and 24 threads, with a base clock of 3.40 GHz and a boost clock of 5.40 GHz. The chip is a dual-die design with a die size of 2x 70.6 mm², containing 16,630 million transistors. It uses AMD Socket AM5 and supports dual-channel DDR5 memory with a bandwidth of 89.6 GB/s. The integrated Radeon Graphics is present, which means the CPU can drive a display without a discrete GPU.

The Intel Xeon 6736P is a 5 nm part from Granite Rapids, with 36 cores and 72 threads. Its base clock is 2.00 GHz and boost is 4.10 GHz. It uses a single monolithic die of 598 mm², which is more than four times the area of each AMD CCD. Intel does not list a transistor count in the database, but the die size alone indicates a very different physical layout. The Xeon runs on Intel Socket 4710 and supports eight-channel DDR5 memory with a bandwidth of 409.6 GB/s, which is 4.6 times the AMD part's memory bandwidth. It has no integrated graphics.

Cache configurations also differ significantly. The AMD chip has 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. The Intel part has 112 KB of L1 per core, 2 MB of L2 per core, and a large 144 MB of shared L3. That extra L3, combined with the eight-channel memory controller, explains why Intel wins the data-heavy benchmarks by such wide margins. The AMD part's strength is clock speed: its 5.40 GHz boost is substantially higher than Intel's 4.10 GHz, which drives the single-thread wins.

The memory bus width is a major architectural differentiator. Dual-channel versus eight-channel is not a small change; it directly impacts how much data can be fed to the cores. The Xeon's 409.6 GB/s versus 89.6 GB/s makes it a different class of platform for memory-intensive server workloads. PCIe lanes also differ: AMD provides 24 Gen 5 lanes, while Intel offers 88 Gen 5 lanes, which matters for systems with many GPUs or NVMe drives. Both support ECC memory, which is expected for the server/workstation segment both chips target.

The Verdict

The data points to a simple decision rule. If the workload is primarily single-threaded or lightly threaded, the AMD Ryzen 9 PRO 9945 is the better choice. Its 4619 single-thread score is more than double the Intel's 2024, and that advantage will translate into faster response times for interactive work, compilation of small code units, and any task that does not scale across many cores.

If the workload is heavily multi-threaded and memory-bound, the Intel Xeon 6736P is the clear winner. It wins 9 of 11 benchmarks, and its margins in physics (55.6%), random string sorting (39.8%), and data encryption (34.1%) are substantial. The Xeon also has a major platform advantage with eight-channel memory and 88 PCIe lanes, which are not reflected in the core benchmark scores but matter for real-world server builds.

The passmark_multithread result is the most telling statistic: Intel wins by only 2.8%, despite having 24 more cores and 48 more threads. That means the AMD chip's per-core efficiency is much higher, and in a generic multi-threaded workload, the core count advantage of the Intel part is almost neutralized. However, in workloads that specifically benefit from the Xeon's memory bandwidth and large L3 cache, the gap grows dramatically.

For a single-socket workstation where the user values snappy single-thread performance and does not need massive memory throughput, the AMD Ryzen 9 PRO 9945 is the better fit. For a server or a workstation running sustained parallel compute, database operations, or virtualization with many tenants, the Intel Xeon 6736P is the stronger platform. The AMD part has a 96th percentile ranking among all CPUs, and the Intel part also sits at the 96th percentile, so both are top-tier parts; the choice is about matching the workload to the platform.

FAQ

Q: Which CPU has better single-thread performance?

A: The AMD Ryzen 9 PRO 9945 wins decisively, scoring 4619 in PassMark single-thread versus 2024 for the Intel Xeon 6736P, a 128.2% advantage.

Q: How much faster is the Intel Xeon 6736P in multi-threaded workloads?

A: In PassMark multithread, the Intel part scores 50072 versus 48664 for AMD, a lead of only 2.8%. That is the narrowest margin in the entire comparison.

Q: What is the biggest benchmark gap between the two?

A: The largest gap is in PassMark physics, where Intel scores 6531 versus 2902 for AMD, a 55.6% difference. The second largest is random string sorting at 39.8%.

Q: Does the Intel Xeon 6736P have integrated graphics?

A: No, it does not. The AMD Ryzen 9 PRO 9945 includes Radeon Graphics, while the Intel part lists integrated graphics as N/A.

Q: What is the memory bandwidth difference?

A: The Intel Xeon 6736P supports eight-channel DDR5 with 409.6 GB/s bandwidth, while the AMD Ryzen 9 PRO 9945 uses dual-channel DDR5 with 89.6 GB/s. That is a 4.6 times difference.

Q: How do the core counts compare?

A: The Intel Xeon 6736P has 36 cores and 72 threads, while the AMD Ryzen 9 PRO 9945 has 12 cores and 24 threads. Intel has three times the cores and threads.

Head-to-Head Benchmarks

The PassMark suite reveals a consistent pattern. The AMD Ryzen 9 PRO 9945 wins both single-thread tests with identical scores of 4619, while the Intel Xeon 6736P scores 2024 in both, giving AMD a 128.2% advantage in each. This is the only test family where AMD leads.

Every other benchmark goes to Intel. In data compression, Intel scores 796658 against AMD's 579972, a 27.2% lead. Data encryption shows Intel at 46236 versus 30450, a 34.1% gap. Extended instructions favor Intel at 55563 versus 44374, a 20.1% margin. Find prime numbers is closer: Intel scores 392 versus 342, a 12.8% lead. Floating point math goes to Intel at 149770 versus 107340, a 28.3% gap. Integer math is Intel at 206833 versus 171175, a 17.2% lead.

The multithread test is the outlier: Intel wins with 50072 versus 48664, but the margin is just 2.8%. Physics is a blowout for Intel at 6531 versus 2902, a 55.6% difference. Random string sorting goes to Intel at 103723 versus 62458, a 39.8% gap.

The average benchmark score tells the overall story: AMD averages 96083, while Intel averages 87864. That means the AMD chip has a higher average score across all recorded benchmarks, despite losing 9 of 11 head-to-head tests. This is because the single-thread wins are so large that they pull up AMD's average. The AMD part also ranks at the 96th percentile versus all CPUs, and Intel is also at the 96th percentile, so both are in the top tier of available processors.

For the nearest rivals, the AMD Ryzen 9 PRO 9945 sits close to the AMD EPYC 4565P, which averages 95764, a delta of 0.3%. The Intel Xeon 6736P is near the Intel Xeon 6731P, which averages 87756, a delta of 0.1%. These numbers show that each chip is well positioned within its own product family.

Specification Differences

The two processors differ on nearly every major specification. The AMD Ryzen 9 PRO 9945 uses 12 cores and 24 threads, while the Intel Xeon 6736P has 36 cores and 72 threads. Base clocks are 3.40 GHz for AMD and 2.00 GHz for Intel. Boost clocks are 5.40 GHz for AMD and 4.10 GHz for Intel. The TDP is dramatically different: 65 watts for AMD versus 205 watts for Intel.

The sockets are incompatible: AMD uses Socket AM5, while Intel uses Socket 4710. The process nodes are 4 nm for AMD (TSMC) and 5 nm for Intel (Intel foundry). The AMD die size is listed as 2x 70.6 mm², while Intel's is 598 mm². AMD has 16,630 million transistors, while Intel does not list a transistor count.

Cache layouts differ per core and in total. AMD has 80 KB L1 per core, 1 MB L2 per core, and 64 MB L3. Intel has 112 KB L1 per core, 2 MB L2 per core, and 144 MB shared L3. Memory support is DDR5 for both, but AMD uses dual-channel with 89.6 GB/s, while Intel uses eight-channel with 409.6 GB/s. PCIe is Gen 5 for both, but AMD offers 24 lanes and Intel offers 88 lanes.

Integrated graphics: AMD has Radeon Graphics, Intel has N/A. The market segment is server/workstation for both. Production status is active for both. The AMD release date is 2025-09-15, while Intel's is 2025-02-23. AMD has no launch MSRP listed in the database; Intel has a launch MSRP of $3351. Both have locked multipliers. The AMD part number is 100-000001407, and Intel's is SRVNW.