AMD Ryzen Threadripper PRO 3995WX vs Intel Xeon 6745P Comparison

The AMD Ryzen Threadripper PRO 3995WX and Intel Xeon 6745P are both 98th-percentile processors, but they achieve that status through fundamentally different designs. The AMD part fields 64 cores and 128 threads, while the Intel chip counters with 32 cores and 64 threads. The benchmark data reveals a clear split: the Intel Xeon 6745P wins 11 of the 16 head-to-head tests, yet the AMD Ryzen Threadripper PRO 3995WX secures the most decisive victories in specific workloads.

Head-to-Head Benchmarks

The most striking result is in Passmark integer math, where the AMD Ryzen Threadripper PRO 3995WX scores 490116 against the Intel Xeon 6745P's 336926, a 45.5% advantage. This is the largest margin in either direction across all tests. Data compression follows a similar pattern, with the AMD part scoring 1841292 versus 1352801, a 36.1% lead. Random string sorting shows another substantial AMD win at 188574 against 133528, a 41.2% delta.

The AMD processor also dominates data encryption, scoring 124433 compared to the Intel chip's 66665, an 86.7% advantage. Floating point math is closer but still favors AMD: 277715 versus 267438, a 3.8% edge.

The Intel Xeon 6745P counters with consistent but smaller victories across most other tests. In Cinebench R23 multicore, the Intel part scores 71578 against AMD's 71338, a slim 0.3% margin. The pattern repeats in Cinebench R15 multicore (7214 versus 7190, a 0.3% delta) and Cinebench R20 multicore (30062 versus 29961, a 0.3% delta). Single-core Cinebench results show similar narrow Intel leads: R15 at 1018 versus 1014 (0.4%), R20 at 4244 versus 4229 (0.4%).

The largest Intel win comes in Passmark single-thread performance, where the Xeon 6745P scores 3450 against AMD's 2595, a 24.8% advantage. Passmark physics also favors Intel by a double-digit margin: 6144 versus 5498, a 10.5% delta. Find prime numbers shows Intel ahead at 681 versus 566, a 16.9% edge. Extended instructions tests are nearly tied, with Intel scoring 108326 against AMD's 106423, a 1.8% difference. Passmark multithread is essentially a draw: 84210 versus 83928, a 0.3% Intel lead.

Architecture Differences

The foundational difference is core count versus per-core capability. AMD's Ryzen Threadripper PRO 3995WX uses 64 cores based on the Zen 2 architecture, manufactured on a 7 nm process by TSMC. Intel's Xeon 6745P uses 32 cores based on Granite Rapids architecture, built on Intel's 5 nm process. This means the AMD chip has twice the core count but relies on older, larger process technology.

Cache configurations also differ significantly. AMD provides 64 KB of L1 cache per core and 512 KB of L2 per core, with a massive 256 MB of L3 cache. Intel provides 112 KB of L1 per core and 2 MB of L2 per core, with 336 MB of shared L3 cache. The Intel chip's larger per-core caches help explain its single-thread advantage, while AMD's total L3 capacity (256 MB) remains substantial.

Memory support diverges completely. AMD uses DDR4 with an eight-channel bus providing 204.8 GB/s of bandwidth. Intel uses DDR5 with an eight-channel bus delivering 409.6 GB/s, exactly double the bandwidth. Both support ECC memory.

PCIe connectivity differs in both generation and lane count. AMD offers Gen 4 with 128 lanes (CPU only), while Intel offers Gen 5 with 88 lanes (CPU only). The Intel implementation provides a newer standard but fewer lanes. Both processors have locked multipliers.

Physical specifications also vary. AMD's die size is listed as 8x 74 mm² with 30,400 million transistors. Intel's die size is 2x 598 mm² with no transistor count listed. The AMD part uses Socket WRX8, while Intel uses Socket 4710. The Intel chip carries a higher TDP of 300 watts compared to AMD's 280 watts.

FAQ

Q: Which processor has more cores?

A: The AMD Ryzen Threadripper PRO 3995WX has 64 cores and 128 threads, exactly double the Intel Xeon 6745P's 32 cores and 64 threads.

Q: Why does the Intel chip win most benchmark tests despite having fewer cores?

A: The Intel Xeon 6745P has a higher base clock (3.10 GHz versus 2.70 GHz), a higher boost clock (4.30 GHz versus 4.20 GHz), newer DDR5 memory with double the bandwidth (409.6 GB/s versus 204.8 GB/s), and larger per-core caches (112 KB L1 and 2 MB L2 versus 64 KB L1 and 512 KB L2).

Q: Which processor is better for data encryption workloads?

A: The AMD Ryzen Threadripper PRO 3995WX is vastly better, scoring 124433 in Passmark data encryption compared to Intel's 66665, an 86.7% advantage.

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

A: The Intel Xeon 6745P provides 409.6 GB/s of memory bandwidth using DDR5, which is exactly double the 204.8 GB/s available from the AMD Ryzen Threadripper PRO 3995WX using DDR4.

Q: How do the release dates compare?

A: The AMD Ryzen Threadripper PRO 3995WX was released on 2020-07-13, while the Intel Xeon 6745P was released on 2025-02-23.

Q: Which processor has more PCIe lanes?

A: The AMD Ryzen Threadripper PRO 3995WX provides 128 Gen 4 lanes (CPU only), while the Intel Xeon 6745P provides 88 Gen 5 lanes (CPU only). The AMD part has more lanes, but the Intel part uses a newer generation.

The Verdict

The data indicates that the Intel Xeon 6745P is the better all-around performer, winning 11 of 16 head-to-head benchmarks. Its advantages in single-thread performance (24.8% in Passmark single-thread) and memory bandwidth (double the AMD part) make it the stronger choice for workloads that depend on per-core speed or memory throughput. The consistent, if narrow, Cinebench wins across all versions suggest a slight edge in rendering tasks.

However, the AMD Ryzen Threadripper PRO 3995WX wins the tests that matter most for parallel integer workloads. Its 45.5% lead in integer math, 41.2% lead in random string sorting, and 86.7% lead in encryption indicate that applications that can fully utilize 64 cores will see dramatic performance gains over the Intel chip. The AMD part also wins floating point math by 3.8%.

The choice depends on workload characteristics. For broadly varied server or workstation tasks, the Intel Xeon 6745P's balanced performance and higher single-thread score make it the safe pick. For specialized workloads that heavily use integer operations, compression, sorting, or encryption, the AMD Ryzen Threadripper PRO 3995WX offers performance advantages that the Intel chip cannot match despite its newer architecture.

Specification Differences

The two processors differ in every major specification category. Core count is 64 for AMD versus 32 for Intel. Threads are 128 versus 64. Base clock is 2.70 GHz for AMD versus 3.10 GHz for Intel. Boost clock is 4.20 GHz versus 4.30 GHz. TDP is 280 watts versus 300 watts. Sockets are AMD Socket WRX8 versus Intel Socket 4710.

Architecture is Zen 2 (Castle Peak) for AMD versus Granite Rapids for Intel. Process node is 7 nm from TSMC versus 5 nm from Intel. Die size is 8x 74 mm² versus 2x 598 mm². Transistors are 30,400 million for AMD, with none listed for Intel.

Cache differs per core: AMD has 64 KB L1 and 512 KB L2, while Intel has 112 KB L1 and 2 MB L2. Total L3 cache is 256 MB for AMD versus 336 MB shared for Intel. Memory support is DDR4 for AMD versus DDR5 for Intel, both with eight-channel buses but different bandwidths (204.8 GB/s versus 409.6 GB/s). PCIe is Gen 4 with 128 lanes for AMD versus Gen 5 with 88 lanes for Intel. Integrated graphics are absent for AMD and listed as N/A for Intel. Release dates are 2020-07-13 versus 2025-02-23. Launch MSRP is $5489 for AMD and $5250 for Intel.

Where Each One Wins

The Intel Xeon 6745P wins in single-threaded and lightly threaded workloads. Its 24.8% advantage in Passmark single-thread performance and 10.5% lead in physics tests point to applications that rely on one or a few fast cores. The higher base and boost clocks, combined with larger per-core caches, support this strength. The Intel part also wins all Cinebench tests, though by margins under 0.5%, suggesting parity in typical rendering workloads.

The AMD Ryzen Threadripper PRO 3995WX wins in massively parallel integer workloads. Data compression (36.1% ahead), integer math (45.5% ahead), random string sorting (41.2% ahead), and encryption (86.7% ahead) all show the benefit of 64 cores versus 32. The floating point math win, while smaller at 3.8%, still demonstrates an edge in that domain. Workloads that involve heavy data manipulation, cryptographic operations, or sorting tasks will see substantial gains on the AMD platform.

The specifications reinforce this split. Intel's newer architecture, higher clocks, DDR5 memory, and double bandwidth favor tasks sensitive to memory latency and single-core speed. AMD's double core count and massive L3 cache favor tasks that scale with parallel execution. The choice is between a balanced modern processor and a high-core-count specialist.