AMD EPYC 9335 vs Intel Xeon 696X Comparison

Head-to-Head Benchmarks

The benchmark data presents a decisive picture. Across all 11 recorded head-to-head comparisons, the Intel Xeon 696X wins every single test. There are no benchmark wins recorded for the AMD EPYC 9335 in this comparison set. This is not a matter of narrow margins; the deltas are substantial in nearly every category.

The largest single advantage appears in the PassMark find prime numbers test. The Intel Xeon 696X scores 853, while the AMD EPYC 9335 scores 340. That is a 150.9% advantage for the Intel part, more than double the AMD result. This type of workload, which is heavily dependent on integer throughput and cache behavior, shows the Xeon 696X at its most dominant.

Floating point math also favors Intel heavily. The Xeon 696X records 450164 in PassMark floating point math, against 228123 for the EPYC 9335. This is a 97.3% delta, meaning the Intel processor delivers nearly twice the floating point throughput in this test. For workloads that rely on scientific computing, simulation, or other FP-heavy tasks, the recorded data indicates a large performance gap.

Data compression shows an 88.3% advantage for the Intel Xeon 696X. The Intel part scores 2264907 in PassMark data compression, while the AMD EPYC 9335 scores 1203096. This workload benefits from high core counts and large caches, both of which the Xeon 696X possesses in greater quantity.

Encryption performance also leans strongly toward Intel. The Xeon 696X scores 112529 in PassMark data encryption, compared to 63159 for the EPYC 9335. That is a 78.2% delta. The physics test shows a similar pattern: 3382 for the Intel part versus 1905 for the AMD part, a 77.5% difference.

The extended instructions benchmark gives the Xeon 696X a 63.6% lead, with scores of 172975 against 105706. Integer math follows at 65.2%, with the Intel part scoring 572072 and the AMD part scoring 346291. The multithread test shows a 59.5% advantage for Intel, with 104974 versus 65811. Random string sorting, a test sensitive to memory latency and cache capacity, favors the Xeon 696X by 54.7%, with scores of 180392 and 116608 respectively.

Even the single-thread test, where one might expect a closer contest given the higher base clock of the AMD part, goes to Intel. The Xeon 696X scores 3742 in PassMark single thread, while the EPYC 9335 scores 2732. That is a 37% advantage. The recorded data shows Intel leading in every measured category, with the smallest margin still being a substantial 37% in single-thread performance.

Architecture Differences

The two processors come from different design philosophies. The Intel Xeon 696X uses the Granite Rapids architecture, built on Intel's 5 nm process node. It belongs to the Xeon 600 (Granite Rapids-WS) generation and uses Intel Socket 4710. The AMD EPYC 9335 uses the Zen 5 architecture, codenamed Turin, built on TSMC's 4 nm process node. It belongs to the EPYC 9005 series and uses AMD Socket SP5.

Core counts differ significantly. The Intel Xeon 696X has 64 cores and 128 threads. The AMD EPYC 9335 has 32 cores and 64 threads. This is a 2:1 ratio in both cores and threads, which explains much of the multithreaded performance gap observed in the benchmarks.

Cache hierarchies also differ in both size and organization. The Intel Xeon 696X provides 112 KB of L1 cache per core, 2 MB of L2 cache per core, and 336 MB of shared L3 cache. The AMD EPYC 9335 provides 80 KB of L1 cache per core, 1 MB of L2 cache per core, and 128 MB of shared L3 cache. The Intel part has a much larger L3 pool, more than 2.5 times the AMD part's shared cache.

Die configuration is another differentiator. The Intel Xeon 696X uses a die size of 2x 598 mm². The AMD EPYC 9335 uses 4x 70.6 mm² chiplets and lists 33,260 million transistors. The Intel part does not have a transistor count recorded in the database, but the physical layout shows a very different approach: two large Intel dies versus four smaller AMD chiplets.

Memory support differs in channel count and bandwidth. The Intel Xeon 696X supports DDR5 memory over an eight-channel bus, with a recorded memory bandwidth of 409.6 GB/s. The AMD EPYC 9335 also supports DDR5 but over a twelve-channel bus, with a recorded memory bandwidth of 576.0 GB/s. The AMD part has a higher peak memory bandwidth, though this did not translate into a benchmark win in any of the recorded tests.

Both processors support ECC memory and offer PCIe Gen 5 with 128 lanes (CPU only). Neither has integrated graphics. Both are classified as Server/Workstation parts and both have Active production status. The Intel Xeon 696X has an unlocked multiplier, while the AMD EPYC 9335 does not.

The release dates differ. The AMD EPYC 9335 was released on 2024-10-09, while the Intel Xeon 696X has a release date of 2026-02-01. The Intel part is the newer of the two in the database.

The Verdict

The benchmark results are unambiguous. The Intel Xeon 696X wins all 11 head-to-head tests against the AMD EPYC 9335. The margins range from 37% in single-thread performance to 150.9% in prime number finding. The average benchmark score tells a similar story: the Intel part averages 286102 across all recorded benchmarks, while the AMD part averages 194228.

It is worth noting where each processor sits among its own nearest rivals, independent of this head-to-head comparison. The Intel Xeon 696X has an average score of 286102, placing it 0.2% above the AMD EPYC 9565 (285471) and 0.3% below the AMD EPYC 9555P (287066). It is 2% ahead of the Intel Xeon 6780E (280438) and 2.3% ahead of the AMD Ryzen Threadripper 9970X (279778). This places it in the 99th percentile of all CPUs in the database.

The AMD EPYC 9335, by contrast, averages 194228. It sits 0.3% below the Intel Xeon 6741P (194901), 0.4% above the Intel Xeon 678X (193477), 3.5% above the Intel Xeon 6740E (187718), and 4.9% above the AMD EPYC 8534P (185092). It also sits in the 99th percentile of all CPUs, but its peer group is at a lower absolute performance level.

The data suggests a clear hierarchy. The Intel Xeon 696X competes with the top-tier EPYC 9005 series parts and high-end Threadripper processors. The AMD EPYC 9335 competes with mid-range Xeon parts and lower EPYC models. In a direct comparison, the Intel part is the stronger performer by a wide margin.

Specification Differences

The two processors differ in nearly every major specification category. The core count difference is the most obvious: 64 cores and 128 threads for the Intel Xeon 696X, versus 32 cores and 64 threads for the AMD EPYC 9335.

Clock speeds also differ. The Intel Xeon 696X has a base clock of 2.40 GHz and a boost clock of 4.80 GHz. The AMD EPYC 9335 has a higher base clock of 3.00 GHz but a lower boost clock of 4.40 GHz. The Intel part reaches higher peak frequencies, but the AMD part has a higher floor frequency.

Thermal design power is significantly different. The Intel Xeon 696X has a TDP of 350 watts. The AMD EPYC 9335 has a TDP of 210 watts. The Intel part draws considerably more power, which is consistent with its higher core count and larger dies.

Cache structures differ in all three levels. The Intel Xeon 696X has 112 KB of L1 per core, 2 MB of L2 per core, and 336 MB of shared L3. The AMD EPYC 9335 has 80 KB of L1 per core, 1 MB of L2 per core, and 128 MB of shared L3.

Memory bandwidth is higher on the AMD part. The EPYC 9335 offers 576.0 GB/s over a twelve-channel bus. The Xeon 696X offers 409.6 GB/s over an eight-channel bus. Both support DDR5 and ECC.

The process nodes differ by one step. Intel uses a 5 nm process with its own foundry. AMD uses a 4 nm process with TSMC as the foundry. The die configurations are also different: 2x 598 mm² for Intel, 4x 70.6 mm² for AMD, with the AMD part listing 33,260 million transistors.

The sockets are incompatible: Intel Socket 4710 for the Xeon 696X, AMD Socket SP5 for the EPYC 9335. The multiplier is unlocked on the Intel part but locked on the AMD part. The launch MSRP for the Intel Xeon 696X is $5599. The launch MSRP for the AMD EPYC 9335 is $3178.

FAQ

Q: Which processor wins more benchmark tests in this comparison?

A: The Intel Xeon 696X wins all 11 head-to-head benchmark tests recorded against the AMD EPYC 9335. The AMD EPYC 9335 has no wins in this comparison set.

Q: What is the largest performance gap between the two?

A: The largest gap is in PassMark find prime numbers, where the Intel Xeon 696X scores 853 against 340 for the AMD EPYC 9335, a delta of 150.9%.

Q: How do the core counts compare?

A: The Intel Xeon 696X has 64 cores and 128 threads. The AMD EPYC 9335 has 32 cores and 64 threads. The Intel part has exactly twice the core and thread count.

Q: Which processor has higher memory bandwidth?

A: The AMD EPYC 9335 has higher recorded memory bandwidth at 576.0 GB/s over a twelve-channel DDR5 bus. The Intel Xeon 696X records 409.6 GB/s over an eight-channel DDR5 bus.

Q: What is the single-thread performance difference?

A: The Intel Xeon 696X scores 3742 in PassMark single thread, while the AMD EPYC 9335 scores 2732. This is a 37% advantage for the Intel part.

Q: Are both processors in the same performance percentile?

A: Yes, both are in the 99th percentile of all CPUs in the database, but their average benchmark scores differ substantially: 286102 for the Intel Xeon 696X and 194228 for the AMD EPYC 9335.

Where Each One Wins

The Intel Xeon 696X wins every recorded benchmark category, so the use-case split is defined by the magnitude of its advantages rather than by any AMD wins. Still, the data supports different conclusions for different workload types.

For heavily multithreaded workloads, the Intel Xeon 696X is the clear choice. Its 64 cores and 128 threads, combined with 336 MB of shared L3 cache, produce dominant results in multithread tests (104974 versus 65811, a 59.5% advantage) and in data compression (88.3% ahead). Workloads that scale with core count and cache capacity will see the largest benefits from the Intel part.

For floating point and scientific workloads, the Xeon 696X again leads by a wide margin. The floating point math score of 450164 is 97.3% higher than the EPYC 9335's 228123. The find prime numbers test, which is also compute-intensive, shows the largest delta of all at 150.9%. Any workload dominated by mathematical computation will favor the Intel part strongly.

For integer-heavy workloads, the Xeon 696X leads by 65.2% in integer math and 78.2% in encryption. These tasks benefit from the combination of more cores and the larger cache hierarchy. The extended instructions test also favors Intel at 63.6%, indicating strong SIMD and vector processing capability.

For single-threaded or lightly threaded workloads, the gap narrows but still favors Intel. The single-thread test shows a 37% advantage for the Xeon 696X, despite the EPYC 9335 having a higher base clock of 3.00 GHz versus 2.40 GHz. The Intel part's higher boost clock of 4.80 GHz likely contributes to this result.

The AMD EPYC 9335, while losing every head-to-head test, still holds advantages in certain specification areas that may matter for system design. It has lower power consumption at 210 watts TDP versus 350 watts, higher memory bandwidth at 576.0 GB/s versus 409.6 GB/s, and a smaller physical footprint with 4x 70.6 mm² chiplets versus 2x 598 mm² dies. For systems where power draw, memory bandwidth, or physical layout are the primary constraints, the EPYC 9335 offers a different trade-off profile.

The AMD part also has a lower launch MSRP of $3178 compared to $5599 for the Intel part, though the benchmark data shows the Intel part delivering substantially higher performance across all measured categories. The choice between these two processors depends on whether absolute performance or system-level constraints take priority. The recorded data, however, is unambiguous about which one is faster in every measured workload.