AMD Ryzen Threadripper PRO 5945WX vs Intel Xeon Gold 6346 Comparison
AMD Ryzen Threadripper PRO 5945WX
Xeon Gold 6346
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen Threadripper PRO 5945WX vs Intel Xeon Gold 6346
The AMD Ryzen Threadripper PRO 5945WX and Intel Xeon Gold 6346 are both active, high-end server/workstation parts, but the benchmark data shows a clear, consistent winner. Across every head-to-head Cinebench test, the AMD processor takes the lead, with margins hovering around 8% in both single-core and multi-core workloads. The Xeon Gold 6346 counters with a higher core count (16 vs. 12) and a lower 205W TDP, but this does not translate into a performance advantage in the measured tests. The Threadripper PRO 5945WX is the pick for raw compute throughput, while the Xeon Gold 6346 is the choice for those prioritizing core count per socket and lower thermal design power.
The Verdict
The data is unambiguous: the AMD Ryzen Threadripper PRO 5945WX wins every single benchmark in the head-to-head comparison, securing all 6 wins. Its average benchmark score of 9348 places it ahead of the Xeon Gold 6346's 9155, and its nearest rival, the Intel Xeon Platinum 8180, is only 0.6% behind. The Xeon Gold 6346, conversely, sits 0.9% behind its closest competitor, the Intel Xeon Gold 5320. If you are building a workstation where Cinebench-style rendering performance is the primary metric, the Threadripper PRO 5945WX is the superior choice. Its 8.1% lead in multi-core tests like Cinebench R23 (34204 vs. 31650) is substantial enough to notice in long render times. However, the Xeon Gold 6346 is not without merit: it offers 16 cores and 32 threads versus the AMD's 12 cores and 24 threads, which may be relevant for specific virtualization or heavily threaded workloads not captured by these benchmarks. The Xeon also consumes less power (205W vs. 280W TDP), making it a more thermally manageable option in dense server chassis. The choice hinges on whether you prioritize the demonstrated single-thread and multi-thread speed of the AMD part or the higher core count and lower power envelope of the Intel part.
Where Each One Wins
The AMD Ryzen Threadripper PRO 5945WX is the clear winner in all measured CPU-bound benchmarks. Its victories span both single-core and multi-core tests, indicating a holistic architecture advantage. In Cinebench R23 multi-core, it posts 34204 points, an 8.1% improvement over the Xeon's 31650. This pattern repeats in R20 (14365 vs. 13293) and R15 (3447 vs. 3190). The single-core results are equally decisive, with the AMD part leading by 8% in R20 (2027 vs. 1876) and 8.1% in R23 (4828 vs. 4468). This suggests the Zen 3 architecture's higher 4.50 GHz boost clock and superior instructions-per-clock are delivering tangible results, even against a processor with 4 more physical cores. The Xeon Gold 6346's only theoretical advantage lies in its core count and lower TDP. It has 16 cores and 32 threads, offering more parallel hardware threads for workloads that scale perfectly with core count, but the benchmark results show that its 3.60 GHz boost clock and Ice Lake architecture cannot overcome the AMD's efficiency. The Xeon's 205W TDP is a significant advantage for system builders concerned about cooling and power delivery, but it does not show up in the performance data.
Architecture Differences
The two processors come from different architectural lineages. The AMD Ryzen Threadripper PRO 5945WX is built on the Zen 3 architecture (codename Chagall PRO) using TSMC's 7 nm process node. It packs 12 cores and 24 threads in a 4x 81 mm² die configuration, totaling 16,600 million transistors. The Intel Xeon Gold 6346, in contrast, uses the Ice Lake-SP architecture on Intel's 10 nm process. It is a monolithic 16-core, 32-thread design, though the fact pack does not list a transistor count or die size for it. The AMD part uses an AMD Socket WRX8, while the Xeon uses Intel Socket 4189. Both support DDR4 memory in an eight-channel configuration, providing an identical 204.8 GB/s of memory bandwidth, and both support ECC memory. The PCIe implementation differs: the AMD offers 128 lanes of Gen 4 (CPU only), while the Intel provides 64 lanes of Gen 4 (CPU only). This is a substantial difference in expandability for multi-GPU or high-speed storage configurations. The cache hierarchies also diverge. The AMD has 64 MB of L3 cache, while the Intel has 36 MB of shared L3 cache. Both have 64 KB of L1 per core, but the AMD has 512 KB of L2 per core versus the Intel's 1 MB per core. Neither processor features integrated graphics, and both have locked multipliers.
FAQ
Q: Which processor has the higher boost clock?
A: The AMD Ryzen Threadripper PRO 5945WX has a boost clock of 4.50 GHz, which is higher than the Intel Xeon Gold 6346's 3.60 GHz.
Q: Does the Intel Xeon Gold 6346 have more cores than the AMD?
A: Yes, the Intel Xeon Gold 6346 has 16 cores and 32 threads, while the AMD Ryzen Threadripper PRO 5945WX has 12 cores and 24 threads.
Q: Which CPU is faster in single-core performance?
A: The AMD Ryzen Threadripper PRO 5945WX wins all single-core tests, including an 8.1% lead in Cinebench R23 single-core with a score of 4828 versus 4468.
Q: What is the difference in memory bandwidth between the two?
A: Both processors support eight-channel DDR4 memory and have an identical memory bandwidth of 204.8 GB/s.
Q: Which processor has a lower TDP?
A: The Intel Xeon Gold 6346 has a TDP of 205W, which is lower than the AMD Ryzen Threadripper PRO 5945WX's 280W.
Q: How do they compare in PCIe lanes?
A: The AMD Ryzen Threadripper PRO 5945WX supports 128 PCIe Gen 4 lanes (CPU only), while the Intel Xeon Gold 6346 supports 64 PCIe Gen 4 lanes (CPU only).
Head-to-Head Benchmarks
The benchmark results are remarkably consistent. The AMD Ryzen Threadripper PRO 5945WX wins all six head-to-head Cinebench tests, with the deltaPct almost exactly 8% across the board. The largest margin is an 8.1% win in Cinebench R23 multi-core, where the AMD scores 34204 against the Intel's 31650. This is a 2554-point difference, a significant gap in a demanding render workload. The Cinebench R20 multi-core test tells a similar story: AMD scores 14365, Intel scores 13293, an 8.1% delta. In the older Cinebench R15 multi-core test, the AMD leads 3447 to 3190, again an 8.1% margin. The single-core results are just as one-sided. In Cinebench R23 single-core, the AMD posts 4828, which is 8.1% ahead of the Intel's 4468. The R20 single-core test shows a 2027 to 1876 score in favor of AMD (8% delta), and the R15 single-core test shows a 486 to 450 win for AMD (8% delta). The consistency of the ~8% margin is striking. It suggests that the AMD's architectural efficiency and higher clock speeds provide a uniform performance uplift across different Cinebench versions, irrespective of the workload's thread count. The Intel's 4-core advantage does not help it close the gap; instead, the AMD's superior per-core performance carries the day. While the AMD also shows a Geekbench multi-core score of 13371 and a single-core score of 2059, the Intel does not have Geekbench results in the fact pack, so the comparison is limited to Cinebench. The AMD's average benchmark score of 9348 also exceeds the Intel's 9155, reinforcing the overall trend.
Specification Differences
| Specification | AMD Ryzen Threadripper PRO 5945WX | Intel Xeon Gold 6346 |
| :--- | :--- | :--- |
| Cores | 12 | 16 |
| Threads | 24 | 32 |
| Base Clock | 4.10 GHz | 3.10 GHz |
| Boost Clock | 4.50 GHz | 3.60 GHz |
| TDP | 280 W | 205 W |
| Socket | AMD Socket WRX8 | Intel Socket 4189 |
| Architecture | Zen 3 | Ice Lake |
| Process Node | 7 nm (TSMC) | 10 nm (Intel) |
| L2 Cache | 512 KB (per core) | 1 MB (per core) |
| L3 Cache | 64 MB | 36 MB (shared) |
| PCIe | Gen 4, 128 Lanes (CPU only) | Gen 4, 64 Lanes (CPU only) |
| Foundry | TSMC | Intel |
| Transistors | 16,600 million | Not listed |
| Die Size | 4x 81 mm² | Not listed |
| Release Date | 2022-03-07 | 2021-04-05 |
| Part Number | 100-000000448 | SRKHNCD8068904570201 |