AMD Ryzen Threadripper 9970X vs Intel Xeon 6741P Comparison

AMD
AMD

AMD Ryzen Threadripper 9970X

CORE STATE Shimada Peak
CORE SPECS 32 Cores / 64 Threads
CLOCK SPEED 4 Base / 5.4 GHz Turbo
CACHE 128 MB
MAX TDP 350W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
Intel
INTEL

Xeon 6741P

CORE STATE Granite Rapids
CORE SPECS 48 Cores / 96 Threads
CLOCK SPEED 2.5 Base / 3.8 GHz Turbo
CACHE 288 MB (shared)
MAX TDP 300W
ARCHITECTURE Granite Rapids
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

passmark_data_compression
1,757,998
1,816,408
passmark_data_encryption
86,765
89,746
passmark_extended_instructions
142,342
142,682
passmark_find_prime_numbers
615
1,242
passmark_floating_point_math
309,719
358,423
passmark_integer_math
465,378
458,058
passmark_multithread
107,399
100,660
passmark_physics
6,835
13,890
passmark_random_string_sorting
191,445
177,322
passmark_single_thread
4,530
3,195
passmark_singlethread
4,530
3,195
cinebench_cinebench_r15_multicore
N/A
8,624
cinebench_cinebench_r15_singlecore
N/A
1,217
cinebench_cinebench_r20_multicore
N/A
35,935
cinebench_cinebench_r20_singlecore
N/A
5,073
cinebench_cinebench_r23_multicore
N/A
85,561
cinebench_cinebench_r23_singlecore
N/A
12,079

Analysis: AMD Ryzen Threadripper 9970X vs Intel Xeon 6741P

Head-to-Head Benchmarks

The recorded benchmark data splits the two processors almost evenly, with the Intel Xeon 6741P taking 6 of 11 head-to-head tests and the AMD Ryzen Threadripper 9970X winning 5. The margin of victory, however, varies dramatically by workload type, revealing two very different performance profiles.

The largest single win belongs to the Intel Xeon 6741P in the PassMark physics test. The Xeon scores 13,890 against the Threadripper's 6,835, a delta of 50.8% in Intel's favor. This is a massive gap, and it is matched by a nearly identical 50.5% lead in the find prime numbers test, where the Xeon records 1,242 versus 615 for the AMD part. These two workloads, which are heavily dependent on integer execution and memory latency, clearly favor the Intel architecture.

The Xeon also wins the floating point math test with a score of 358,423 against 309,719, a 13.6% advantage. Data compression and encryption also go to Intel, though by much smaller margins: the Xeon scores 1,816,408 in compression versus 1,757,998 for the Threadripper, a 3.2% edge, and 89,746 in encryption versus 86,765, a 3.3% edge. The extended instructions test is nearly a tie, with Intel winning by only 0.2% (142,682 versus 142,342).

The AMD Ryzen Threadripper 9970X counters with its own decisive victories. In single-thread performance, the Threadripper scores 4,530 against the Xeon's 3,195, a 41.8% lead. This is the second-largest margin in the entire comparison. The Threadripper also wins random string sorting by 8% (191,445 versus 177,322), multithread performance by 6.7% (107,399 versus 100,660), and integer math by a narrow 1.6% (465,378 versus 458,058).

The overall average benchmark scores reflect these different strengths. The Threadripper 9970X has an average benchmark score of 279,778, while the Xeon 6741P averages 194,901. This is a 43.5% difference in the AMD part's favor, driven primarily by the huge single-thread gap and the multithread win. However, the nearest rival data places each processor in different competitive contexts. The Threadripper sits within 2.5% of the AMD EPYC 9555P and Intel Xeon 696X, while the Xeon 6741P is within 5.3% of the AMD EPYC 8534P and 3.8% of the Intel Xeon 6740E.

Both processors rank in the 99th percentile among all CPUs in the database. That shared percentile, despite the large average score difference, suggests that the Xeon's strengths in physics and prime number workloads are substantial enough to keep it in the top tier, even though its overall average is lower.

The Verdict

The data paints a clear picture: the AMD Ryzen Threadripper 9970X is the better all-around performer, with a 43.5% higher average benchmark score and a decisive 41.8% lead in single-thread performance. It also wins the multithread test by 6.7%, a result that matters for heavily threaded workloads. The Threadripper is the pick for anyone prioritizing general compute, single-thread responsiveness, or mixed workloads that include sorting and integer math.

The Intel Xeon 6741P, despite its lower average score, is not a weak processor. It dominates in physics and prime number finding with margins above 50%, and it holds smaller wins in floating point math, compression, encryption, and extended instructions. These are not niche results: physics simulations and prime number calculations are common in scientific computing, and the compression and encryption wins are relevant for database and storage workloads. The Xeon is the pick for those specific workloads, and its 48 cores and 96 threads versus the Threadripper's 32 cores and 64 threads provide a raw thread count advantage that shows up in its physics and prime number results.

From the nearest rival data, the Threadripper 9970X competes at a level 0.2% below the Intel Xeon 6780E and 2.5% below the AMD EPYC 9555P. The Xeon 6741P, meanwhile, sits 0.3% above the AMD EPYC 9335 and 5.3% above the AMD EPYC 8534P. These deltas place the Threadripper in a higher overall performance tier, while the Xeon is positioned among mid-range server parts.

For a desktop workstation user who needs one CPU for many task types, the Threadripper's average score and single-thread lead make it the logical choice. For a server operator running a fixed workload dominated by physics or prime number calculations, the Xeon's 50% margins in those tests are too large to ignore.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen Threadripper 9970X is built on the Zen 5 architecture with the codename Shimada Peak, manufactured by TSMC on a 4 nm process. It uses 32 cores and 64 threads, with a base clock of 4.00 GHz and a boost clock of 5.40 GHz. The thermal design power is 350 watts. The chip is constructed from four chiplets, each with a die size of 70.6 mm², totaling 33,260 million transistors. Cache is organized per core: 64 KB of L1 and 1 MB of L2 per core, with a shared 128 MB of L3 cache.

The Intel Xeon 6741P uses the Granite Rapids architecture, manufactured by Intel on a 5 nm process. It has 48 cores and 96 threads, with a base clock of 2.50 GHz and a boost clock of 3.80 GHz. The thermal design power is 300 watts. The die size is significantly larger at 2x 598 mm², and the cache structure differs: 112 KB of L1 per core, 2 MB of L2 per core, and a shared 288 MB of L3 cache.

These differences explain the benchmark results. The Threadripper's much higher clock speeds (5.40 GHz boost versus 3.80 GHz) drive its 41.8% single-thread advantage. The Xeon's larger L3 cache (288 MB versus 128 MB) and higher core count (48 versus 32) contribute to its physics and prime number wins, workloads that benefit from large caches and more threads.

Memory configurations also diverge. The Threadripper supports DDR5 memory over a quad-channel bus with a bandwidth of 204.8 GB/s. The Xeon supports DDR5 over an eight-channel bus with a bandwidth of 409.6 GB/s, exactly double the Threadripper's memory bandwidth. This is a critical architectural difference: the Xeon's physics and prime number results likely benefit from the higher memory bandwidth, while the Threadripper's clock speed advantage compensates in single-thread tests.

PCIe connectivity differs as well. The Threadripper provides Gen 5 with 80 lanes from the CPU, while the Xeon provides Gen 5 with 136 lanes. The Xeon also has a higher TDP of 300 watts versus the Threadripper's 350 watts, which is notable given the Xeon's much larger die size. Both processors support ECC memory, and neither has integrated graphics.

The Threadripper has an unlocked multiplier, while the Xeon's multiplier is locked. The Threadripper is designated for the desktop market segment, while the Xeon is a server or workstation part. The Threadripper uses AMD Socket sTR5, and the Xeon uses Intel Socket 4710.

FAQ

Q: Which processor has the higher single-thread performance?

A: The AMD Ryzen Threadripper 9970X scores 4,530 in the PassMark single-thread test, compared to 3,195 for the Intel Xeon 6741P. This is a 41.8% advantage for the AMD part.

Q: How do the core counts compare?

A: The Intel Xeon 6741P has 48 cores and 96 threads, while the AMD Ryzen Threadripper 9970X has 32 cores and 64 threads. The Xeon has 50% more cores and 50% more threads.

Q: What is the largest performance gap in either direction?

A: The Intel Xeon 6741P leads by 50.8% in the physics test (13,890 versus 6,835). The AMD Ryzen Threadripper 9970X leads by 41.8% in single-thread performance (4,530 versus 3,195).

Q: Which processor has more L3 cache?

A: The Intel Xeon 6741P has 288 MB of shared L3 cache, while the AMD Ryzen Threadripper 9970X has 128 MB of L3 cache.

Q: What are the memory bandwidth figures?

A: The Intel Xeon 6741P supports eight-channel DDR5 with a bandwidth of 409.6 GB/s. The AMD Ryzen Threadripper 9970X supports quad-channel DDR5 with a bandwidth of 204.8 GB/s.

Q: How do their average benchmark scores compare?

A: The AMD Ryzen Threadripper 9970X has an average benchmark score of 279,778, while the Intel Xeon 6741P averages 194,901. The AMD part is 43.5% higher on average.

Where Each One Wins

The AMD Ryzen Threadripper 9970X wins in workloads that depend on per-core speed and low-latency execution. Its 41.8% single-thread lead makes it the superior choice for interactive applications, lightly threaded software, and any task where a single core must do heavy lifting. The 8% win in random string sorting indicates strength in data processing that is not easily parallelized. The 6.7% win in multithread performance, despite having fewer cores, shows that its higher clocks can overcome a core count deficit in many parallel workloads. The narrow 1.6% win in integer math is another data point for general compute.

The Intel Xeon 6741P wins in workloads that are bandwidth-hungry and highly parallel. The 50.8% physics advantage and 50.5% prime number advantage are the standout results, pointing to scientific simulation and number theory applications. The 13.6% win in floating point math is significant for engineering and financial modeling. The 3.2% compression win and 3.3% encryption win make it the better choice for storage servers and security applications. The near-tie in extended instructions (0.2% Intel lead) means neither processor has a meaningful edge in SIMD-heavy code.

The memory bandwidth difference, with the Xeon offering 409.6 GB/s versus 204.8 GB/s, underpins many of these Intel wins. The larger 288 MB L3 cache also favors the Xeon in workloads with large working sets. Conversely, the Threadripper's 5.40 GHz boost clock and unlocked multiplier give it an edge in scenarios where clock speed is the limiting factor.

For a workstation that runs a variety of software, the Threadripper's higher average score and single-thread dominance make it the more versatile choice. For a server dedicated to physics simulations, prime number calculations, or high-bandwidth data processing, the Xeon's specific wins justify its selection. The launch MSRP for the Threadripper is $2499, and for the Xeon it is $4421, though the performance data, not the price, should drive the decision.

DETAILED SPECIFICATIONS

SPECIFICATION
Threadripper 9970X
6741P
Core Specs
Cores
32
48 +50.0%
Threads
64
96 +50.0%
Base Clock (GHz)
4
2.5 -37.5%
Boost Clock (GHz)
5.4
3.8 -29.6%
Frequency (GHz)
4
2.5 -37.5%
Turbo Clock (GHz)
5.4
3.8 -29.6%
Multiplier
40
25 -37.5%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
64 KB (per core)
112 KB (per core)
L2 Cache
1 MB (per core)
2 MB (per core)
L3 Cache
128 MB
288 MB (shared)
Power
TDP (W)
350
300 -14.3%
Architecture
Architecture
Zen 5
Granite Rapids
Codename
Shimada Peak
Granite Rapids
Generation
Ryzen Threadripper (Zen 5 (Shimada Peak))
Xeon 6 (Granite Rapids-SP)
Process Size
4 nm
5 nm
Transistors
33,260 million
Die Size
4x 70.6 mm²
2x 598 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR5
Memory Bus
Quad-channel
Eight-channel
Memory Bandwidth
204.8 GB/s
409.6 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket sTR5
Intel Socket 4710
PCIe
Gen 5, 80 Lanes(CPU only)
Gen 5, 136 Lanes(CPU only)
AMD Multi-Die
IO Process Size
6 nm
10 nm
Interconnect
CXL
Gen 2.0, 64 Lanes (Shared with PCI-E)
Other
Market
Desktop
Server/Workstation
Production Status
Active
Active
Launch Price
$2499
$4421
Part Number
100-000001594
SRVEY
Package
FC-LGA4844
FC-LGA18N
Tj Max
95°C
93°C
Bundled Cooler
None
None
View Ryzen Threadripper 9970X Details View Xeon 6741P Details