AMD Ryzen Threadripper PRO 9995WX vs Intel Xeon 6774P Comparison

Two recent server and workstation flagships, the AMD Ryzen Threadripper PRO 9995WX and the Intel Xeon 6774P, sit at the top of the performance spectrum, but the recorded data separates them sharply. The AMD part occupies the 100th percentile among all CPUs, while the Intel sits at the 99th. Across 14 head-to-head tests, the AMD wins every single one, with margins ranging from 5.2% to 102.8%. That is the headline. The rest of this analysis examines what the data implies about architecture, benchmark behavior, and use cases.

The Verdict

The benchmark results are unambiguous. The AMD Ryzen Threadripper PRO 9995WX outperforms the Intel Xeon 6774P in all 14 recorded head-to-head tests. The Intel wins zero tests. The AMD's average benchmark score is 412068, which places it between the AMD EPYC 9745 (425973, 3.3% higher) and the AMD EPYC 9655P (397773, 3.6% lower). The Intel's average score is 300372, which is 3.3% below the AMD EPYC 9734 (310619) and 4.6% above the AMD EPYC 9555P (287066). In other words, the AMD operates in a higher performance tier than the Intel.

Who should pick which? For any workload represented by these tests, the data says the AMD. The Intel does have a lower launch MSRP, which is recorded as $6760 versus the AMD's $11700, but that is a specification detail, not a benchmark result. The AMD wins all 14 measured categories, including single-thread, multi-thread, integer math, encryption, compression, and physics. The Intel's only recorded advantages appear in the specification sheet, not in any measured test. For maximum compute throughput, the AMD is the data-backed choice.

Architecture Differences

The two processors come from different design philosophies. The AMD uses the Zen 5 architecture, codename Shimada Peak, built on a 4 nm TSMC process. The Intel uses the Granite Rapids architecture, also codename Granite Rapids, built on a 5 nm Intel process. The AMD integrates 99,780 million transistors across a die configuration of 12x 70.6 mm², while the Intel uses a 2x 598 mm² die arrangement and has no recorded transistor count in the database.

Core counts differ sharply. The AMD has 96 cores and 192 threads, while the Intel has 64 cores and 128 threads. Cache layouts also differ. The AMD provides 64 KB of L1 per core, 1 MB of L2 per core, and a 384 MB L3 pool. The Intel provides 112 KB of L1 per core, 2 MB of L2 per core, and a 336 MB shared L3. The AMD has the larger total L3, while the Intel has larger per-core L1 and L2 allocations.

Both parts run a base clock of 2.50 GHz, but the AMD boosts to 5.40 GHz while the Intel boosts to 3.90 GHz. The AMD has an unlocked multiplier, the Intel is locked. Both use DDR5 memory with an eight-channel bus and a recorded bandwidth of 409.6 GB/s, and both support ECC memory. The AMD supports PCIe Gen 5 with 128 lanes (CPU only), while the Intel supports 136 lanes (CPU only). Neither has integrated graphics. The AMD's TDP is 350W, and the Intel's TDP is also 350W. The AMD is a server and workstation part, as is the Intel, and both are marked as active in production.

Head-to-Head Benchmarks

The AMD's largest win is in integer math. The AMD scores 1203634, the Intel scores 593440, a 102.8% difference, meaning the AMD more than doubles the Intel's integer throughput. That is the single biggest gap in the dataset. Data encryption follows closely, with the AMD at 206662 versus the Intel's 115025, a 79.7% lead. Data compression goes to the AMD at 3723652 versus 2309868, a 61.2% gap. Random string sorting sees the AMD at 418973 versus 262417, a 59.7% lead. Floating point math shows the AMD at 725066 versus 465314, a 55.8% advantage.

The Cinebench multi-core results are consistent across all three versions. In Cinebench R15 multi-core, the AMD scores 14978 versus 9660; in R20, it is 62412 versus 40251; in R23, 148601 versus 95836. All three show the same 55.1% lead for the AMD. The extended instructions test shows the AMD at 270647 versus 177273, a 52.7% lead. The PassMark multi-thread test gives the AMD 171200 versus 112749, a 51.8% gap.

Single-thread results also favor the AMD. The passmark_single_thread and passmark_singlethread tests both show 4542 for the AMD and 3047 for the Intel, a 49.1% lead. The smallest margins are in the find prime numbers test, where the AMD scores 1476 versus 1264, a 16.8% lead, and in physics, where the AMD scores 16860 versus 16023, a 5.2% lead. Even the smallest margin is a win for the AMD.

Specification Differences

The fields where the two processors differ are as follows:

• Cores: 96 (AMD) versus 64 (Intel)
• Threads: 192 versus 128
• Boost clock: 5.40 GHz versus 3.90 GHz
• Socket: AMD Socket sTR5 versus Intel Socket 4710
• Architecture: Zen 5 versus Granite Rapids
• Codename: Shimada Peak versus Granite Rapids
• Process node: 4 nm versus 5 nm
• Foundry: TSMC versus Intel
• Transistor count: 99,780 million versus not recorded
• Die size: 12x 70.6 mm² versus 2x 598 mm²
• L1 cache: 64 KB per core versus 112 KB per core
• L2 cache: 1 MB per core versus 2 MB per core
• L3 cache: 384 MB versus 336 MB (shared)
• PCIe lanes: 128 versus 136
• Multiplier: unlocked versus locked
• Launch MSRP: $11700 versus $6760
• Release date: 2025-07-22 versus 2025-05-21

Fields that are identical include the base clock (2.50 GHz), TDP (350W), memory type (DDR5), memory bus (eight-channel), memory bandwidth (409.6 GB/s), ECC support (both true), integrated graphics (both N/A), market segment (server/workstation), and production status (both active).

FAQ

Q: Which processor has more cores?

A: The AMD Ryzen Threadripper PRO 9995WX has 96 cores, while the Intel Xeon 6774P has 64 cores.

Q: What is the largest performance gap in the head-to-head data?

A: The largest gap is in integer math, where the AMD scores 1203634 versus the Intel's 593440, a 102.8% difference. That is more than double the Intel's result.

Q: Which processor has the higher boost clock?

A: The AMD boosts to 5.40 GHz, while the Intel boosts to 3.90 GHz. The base clock is identical at 2.50 GHz.

Q: Which processor has more PCIe lanes?

A: The Intel Xeon 6774P has 136 PCIe Gen 5 lanes (CPU only), while the AMD has 128 lanes (CPU only).

Q: Are both processors unlocked for overclocking?

A: No. The AMD has an unlocked multiplier, while the Intel's multiplier is locked.

Q: What is the smallest margin in the benchmark data?

A: The smallest margin is in the physics test, where the AMD scores 16860 versus the Intel's 16023, a 5.2% lead.

Where Each One Wins

The data shows a complete sweep for the AMD. It wins all 14 recorded head-to-head tests, so the "where each one wins" split is one-sided in terms of measured performance. The AMD is the winner in every category: single-thread, multi-thread, integer math, floating point, encryption, compression, extended instructions, prime finding, physics, and random string sorting. The AMD's dominance is most pronounced in integer math (102.8%), encryption (79.7%), and compression (61.2%), which are common in data-heavy server and workstation workloads.

The Intel, meanwhile, does not win a single benchmark in this dataset. Its advantages are purely on the specification sheet: it has more PCIe lanes (136 versus 128), larger per-core L1 and L2 caches (112 KB versus 64 KB L1, 2 MB versus 1 MB L2), and a lower launch MSRP ($6760 versus $11700). However, those specification advantages do not translate into any measured performance win in the database. The Intel's closest recorded result is in the physics test, where it trails the AMD by only 5.2%, but it still loses. For any workload represented by the recorded benchmarks, the data says the AMD Ryzen Threadripper PRO 9995WX is the processor to choose, while the Intel Xeon 6774P remains a strong but secondary option.