AMD EPYC 73F3 vs Intel Xeon Gold 6336Y Comparison
AMD EPYC 73F3
Xeon Gold 6336Y
PERFORMANCE BENCHMARKS
Analysis: AMD EPYC 73F3 vs Intel Xeon Gold 6336Y
FAQ
Q: Which processor wins the majority of the Cinebench benchmark comparisons?
A: The AMD EPYC 73F3 wins all six head-to-head benchmark comparisons against the Intel Xeon Gold 6336Y, with a consistent delta of 1.3% across every test.
Q: How large is the performance gap in multi-core workloads?
A: In Cinebench R23 multi-core, the AMD EPYC 73F3 scores 39,187 versus 38,689 for the Intel Xeon Gold 6336Y, a 1.3% advantage. The R20 multi-core results show a similar pattern: 16,458 for AMD versus 16,249 for Intel.
Q: Does the Intel processor have any advantage in single-core tests?
A: No. The AMD EPYC 73F3 leads in every single-core benchmark as well. In Cinebench R23 single-core, AMD scores 5,532 against Intel's 5,462, again a 1.3% delta.
Q: What are the core and thread counts for each processor?
A: The AMD EPYC 73F3 has 16 cores and 32 threads, while the Intel Xeon Gold 6336Y has 24 cores and 48 threads. Despite having fewer cores, the AMD part still edges out Intel in multi-core tests.
Q: How do the two processors compare in overall benchmark averages?
A: The AMD EPYC 73F3 has an average benchmark score of 11,334, placing it at the 67th percentile among all CPUs. The Intel Xeon Gold 6336Y averages 11,191, at the 66th percentile.
Q: Are both processors currently in production?
A: Yes, both the AMD EPYC 73F3 and the Intel Xeon Gold 6336Y have an "Active" production status, indicating they remain available for server and workstation deployments.
Architecture Differences
The AMD EPYC 73F3 is built on TSMC's 7 nm process node using the Zen 3 architecture, codenamed Milan. This is a chiplet-based design with a die size of 8x 81 mm² and 33,200 million transistors. In contrast, the Intel Xeon Gold 6336Y uses Intel's 10 nm process with the Ice Lake architecture, specifically Ice Lake-SP. Intel's transistor count and die size are not listed in the data, but the process node difference is significant: AMD's 7 nm TSMC fabrication versus Intel's 10 nm process.
Cache hierarchies differ markedly between the two. Both have 64 KB of L1 cache per core, but the L2 cache per core is 512 KB on the AMD part versus 1 MB per core on the Intel part. The L3 cache is where the divergence is dramatic: the AMD EPYC 73F3 has 256 MB of shared L3 cache, while the Intel Xeon Gold 6336Y has only 36 MB of shared L3. This sevenfold difference in L3 capacity is a defining architectural characteristic.
Both processors support DDR4 memory with an eight-channel memory bus and 204.8 GB/s of memory bandwidth, and both support ECC memory. PCIe lane allocation differs: the AMD EPYC 73F3 provides 128 lanes of PCIe Gen 4 (CPU only), while the Intel Xeon Gold 6336Y provides 64 lanes of PCIe Gen 4 (CPU only). The AMD part doubles the PCIe lane count, which is relevant for high-density I/O configurations.
Clock speeds also diverge. The AMD EPYC 73F3 has a base clock of 3.50 GHz and a boost clock of 4.00 GHz, while the Intel Xeon Gold 6336Y operates at 2.40 GHz base and 3.60 GHz boost. The AMD part has higher clocks in both states. Thermal design power differs as well: the AMD EPYC 73F3 is rated at 240 W TDP, whereas the Intel Xeon Gold 6336Y is rated at 185 W TDP.
The sockets are incompatible: AMD uses Socket SP3, while Intel uses Socket 4189. Neither processor has an unlocked multiplier. The AMD EPYC 73F3 has a launch MSRP of $3521, while no launch MSRP is listed for the Intel Xeon Gold 6336Y.
Head-to-Head Benchmarks
Across all six Cinebench tests, the AMD EPYC 73F3 wins by a uniform 1.3% margin. This consistency is noteworthy because the benchmark suite spans three generations of Cinebench (R15, R20, R23) and both single-core and multi-core workloads.
In Cinebench R15 multi-core, the AMD EPYC 73F3 scores 3,949 versus 3,899 for the Intel Xeon Gold 6336Y. The single-core R15 result is 557 for AMD and 550 for Intel. Both deltas are 1.3%. Moving to R20, the multi-core score for AMD is 16,458 while Intel reaches 16,249; single-core is 2,323 versus 2,294. Again, 1.3% separates them.
The R23 results show the same pattern. AMD's multi-core score of 39,187 tops Intel's 38,689, and AMD's single-core score of 5,532 beats Intel's 5,462. In every case, the delta is exactly 1.3%. This uniformity suggests the performance difference is consistent across workload types and scaling levels, rather than being workload-dependent.
Notably, the Intel Xeon Gold 6336Y has 50% more cores (24 versus 16) and 50% more threads (48 versus 32), yet it still trails in multi-core tests. This indicates that the AMD EPYC 73F3's higher clock speeds (3.50 GHz base and 4.00 GHz boost versus 2.40 GHz and 3.60 GHz) and larger L3 cache (256 MB versus 36 MB) more than compensate for the core-count disadvantage in these Cinebench workloads.
The average benchmark scores align with the head-to-head results. The AMD EPYC 73F3 averages 11,334, which is 1.3% higher than the Intel Xeon Gold 6336Y's average of 11,191. The nearest rivals for the AMD part include the AMD EPYC 7402 with an average score of 11,312 (0.2% delta) and the AMD EPYC 7452 at 11,279 (0.5% delta). For the Intel part, the nearest rivals are the Intel Core i3-1305U at 11,200 (0.1% delta) and the AMD Ryzen 5 3500U at 11,176 (0.1% delta).
Specification Differences
| Specification | AMD EPYC 73F3 | Intel Xeon Gold 6336Y |
|---|---|---|
| Cores | 16 | 24 |
| Threads | 32 | 48 |
| Base Clock | 3.50 GHz | 2.40 GHz |
| Boost Clock | 4.00 GHz | 3.60 GHz |
| TDP | 240 W | 185 W |
| Socket | AMD Socket SP3 | Intel Socket 4189 |
| Architecture | Zen 3 | Ice Lake |
| Codename | Milan | Ice Lake-SP |
| Process Node | 7 nm | 10 nm |
| Foundry | TSMC | Intel |
| Transistors | 33,200 million | Not listed |
| Die Size | 8x 81 mm² | Not listed |
| L2 Cache | 512 KB (per core) | 1 MB (per core) |
| L3 Cache | 256 MB (shared) | 36 MB (shared) |
| PCIe | Gen 4, 128 Lanes (CPU only) | Gen 4, 64 Lanes (CPU only) |
| Launch MSRP | $3521 | Not listed |
| Part Number | 100-000000321100-100000321WOF | Not listed |
| Release Date | 2021-03-14 | 2021-04-05 |
The two processors share several specifications: both use DDR4 memory, both have an eight-channel memory bus, both offer 204.8 GB/s memory bandwidth, both support ECC memory, both lack integrated graphics, both target the server/workstation market segment, and both are currently in production. Neither has an unlocked multiplier.
Where Each One Wins
The AMD EPYC 73F3 wins all six benchmark comparisons, so the use-case split is heavily skewed toward AMD. In multi-threaded rendering workloads, the AMD part demonstrates that its combination of higher clock speeds and massive L3 cache delivers superior performance despite having fewer cores. Cinebench R23 multi-core results show the AMD EPYC 73F3 at 39,187 versus 38,689 for Intel, a margin that would translate to faster render times in CPU-bound production pipelines.
For single-threaded applications, the AMD EPYC 73F3 also takes the lead. Its 4.00 GHz boost clock versus Intel's 3.60 GHz boost clock contributes to the 1.3% advantage in every single-core test. This matters for lightly threaded workloads such as database queries, legacy application execution, or any software that scales poorly across cores.
The Intel Xeon Gold 6336Y does not win any benchmark in this comparison, but its strengths lie in its specification sheet. With 24 cores and 48 threads versus 16 cores and 32 threads, the Intel part offers more parallel compute resources. Its 185 W TDP is lower than AMD's 240 W TDP, which may be relevant for power-constrained deployments. The larger L2 cache per core (1 MB versus 512 KB) could benefit certain access patterns, though the benchmark data does not show an advantage in the tested Cinebench workloads.
The AMD EPYC 73F3's 128 PCIe Gen 4 lanes double Intel's 64 lanes, making it the stronger choice for systems requiring extensive I/O expansion, such as GPU clusters or NVMe storage arrays. The Intel part's 64 lanes are still substantial but more limiting for high-density configurations.
In terms of market positioning, the AMD EPYC 73F3 sits at the 67th percentile among all CPUs, while the Intel Xeon Gold 6336Y is at the 66th percentile. The AMD part's nearest rivals include the AMD EPYC 7402 and 7452, both within 0.5% of its average score. The Intel part's nearest rivals include the Intel Core i3-1305U and AMD Ryzen 5 3500U, both within 0.1% of its average score. These rival relationships indicate that both processors compete in a dense field of similarly performing parts.
For workloads that are sensitive to clock speed and cache capacity, the AMD EPYC 73F3 is the clear choice based on the data. For workloads that require maximum core counts and lower power draw, the Intel Xeon Gold 6336Y offers a different trade-off, though the benchmark results suggest that in Cinebench at least, the AMD part's efficiency per core outperforms Intel's core-count advantage.