AMD EPYC 9654P vs Intel Core i7-1360P Comparison
AMD EPYC 9654P
Core i7-1360P
PERFORMANCE BENCHMARKS
Analysis: AMD EPYC 9654P vs Intel Core i7-1360P
The AMD EPYC 9654P and Intel Core i7-1360P occupy opposite ends of the computing spectrum, yet the benchmark database places them within 0.5% of each other in average score. This comparison is not about equivalence; it is about how a 96-core server monster and a 12-core mobile chip can land in the same statistical neighborhood while delivering radically different experiences.
Head-to-Head Benchmarks
The Cinebench results are not close. They are a complete sweep for the AMD EPYC 9654P, which wins all six head-to-head tests with margins that range from overwhelming to absurd. In Cinebench R23 multi-core, the EPYC scores 98,875 against the i7-1360P’s 11,266.5, a delta of 777.6%. That is not a performance gap; it is a chasm. The EPYC is nearly nine times faster in this workload, which reflects its 96 cores and 192 threads operating on a 360W TDP budget.
Single-core results tell a slightly less lopsided story, though the winner remains the same. In Cinebench R23 single-core, the EPYC posts 13,958 versus 1,829 for the i7-1360P, a 663.1% advantage. Even in Cinebench R15 single-core, where the gap is smallest in percentage terms, the EPYC leads 1,406 to 258, a 445% delta. The i7-1360P’s 5.00 GHz boost clock and Raptor Lake architecture cannot overcome the EPYC’s Zen 4 efficiency at similar single-thread tasks.
The multi-core deltas escalate as the workload scales. Cinebench R20 multi-core shows 41,527 for the EPYC against 6,530 for Intel, a 535.9% difference. Cinebench R15 multi-core is 9,966 versus 1,858, a 436.4% gap. The pattern is consistent: the EPYC’s advantage grows with thread count, which is expected given the 96-core versus 12-core disparity. The i7-1360P has zero wins in the head-to-head set, and its best relative showing is the Cinebench R15 multi-core test, where it still loses by over fourfold.
What makes this matchup strange is the aggregate benchmark picture. The EPYC 9654P has an average benchmark score of 24,465, while the i7-1360P sits at 24,344. The delta between them is a mere 0.5%. This happens because the EPYC’s benchmark list is dominated by Cinebench and Geekbench, while the i7-1360P includes PassMark tests that favor its integrated Iris Xe Graphics and lower latency. The EPYC’s percentile rank is 77, the i7-1360P’s is 76. They are statistical neighbors, yet in the tests they share, the EPYC obliterates the Intel part.
The Verdict
From the data, the EPYC 9654P is the clear performance winner in every benchmark both chips share. If the workload is multi-threaded rendering, compilation, or server-side processing, the EPYC’s 777.6% lead in Cinebench R23 multi-core makes the choice trivial. The i7-1360P is not competitive in any Cinebench test, losing by at least 436.4% even in the most favorable case.
The i7-1360P, however, is a mobile processor with a 28W TDP and integrated graphics. It targets laptops where the EPYC cannot physically exist — the EPYC uses AMD Socket SP5 and a 360W TDP, which requires a server platform. The i7-1360P’s PassMark results, such as 203,746 in data compression and 67,963 in integer math, show it can handle everyday productivity tasks. The EPYC has no integrated graphics and no PassMark scores in the data, so its desktop utility is limited to server or workstation roles.
Choose the EPYC 9654P if your application scales with cores and you have the infrastructure to cool and power a 360W chip. Choose the i7-1360P if you need a laptop CPU with 16 threads, a 5.00 GHz boost, and the flexibility of DDR4 or DDR5 memory. The benchmark data does not support any other conclusion.
Architecture Differences
The EPYC 9654P is built on TSMC’s 5 nm process with Zen 4 architecture, codenamed Genoa. It has 78,840 million transistors spread across a die size of 12x 72 mm², and it uses 64 KB of L1 cache per core, 1 MB of L2 per core, and 384 MB of shared L3 cache. The i7-1360P uses Intel’s 10 nm process with Raptor Lake architecture, codenamed Raptor Lake-P. It has no transistor count or die size listed, but its cache configuration is different: 80 KB of L1 per core, 2 MB of L2 per core, and 18 MB of shared L3.
The memory architectures diverge sharply. The EPYC supports DDR5 only, with a twelve-channel memory bus and 460.8 GB/s of bandwidth. The i7-1360P supports both DDR4 and DDR5, but only on a dual-channel bus, with no bandwidth figure provided. ECC memory is supported on the EPYC but not on the i7-1360P. PCIe lanes differ as well: the EPYC offers Gen 5 with 128 lanes, while the i7-1360P offers Gen 4 with 20 lanes.
Integrated graphics is a major differentiator. The i7-1360P includes Iris Xe Graphics 96EU, making it a complete mobile package. The EPYC has no integrated graphics, which is standard for server processors. The EPYC’s market segment is Server/Workstation, while the i7-1360P is Mobile. The EPYC is unlocked? No — both chips have locked multipliers, so neither is overclockable.
Specification Differences
The core and thread counts are the headline difference: 96 cores and 192 threads for the EPYC versus 12 cores and 16 threads for the i7-1360P. Base clocks are similar — 2.40 GHz for AMD, 2.20 GHz for Intel — but boost clocks favor Intel at 5.00 GHz versus 3.70 GHz. TDP is the most extreme gap: 360W for the EPYC, 28W for the i7-1360P.
Sockets are incompatible: AMD Socket SP5 versus Intel BGA 1744. Process nodes differ: 5 nm TSMC versus 10 nm Intel. Cache totals are wildly different: the EPYC has 384 MB of L3, the i7-1360P has 18 MB. Memory support differs as noted: DDR5-only twelve-channel versus DDR4/DDR5 dual-channel. PCIe generation and lane counts differ: Gen 5 128 lanes versus Gen 4 20 lanes. ECC memory is present on the EPYC, absent on the i7-1360P. Integrated graphics exist only on the Intel part. Release dates differ: November 2022 for AMD, January 2023 for Intel. The launch MSRP is $10625 for the EPYC and $480 for the i7-1360P.
FAQ
Q: Which CPU is faster in Cinebench R23 multi-core?
A: The AMD EPYC 9654P scores 98,875, which is 777.6% higher than the Intel Core i7-1360P’s 11,266.5.
Q: Does the Intel Core i7-1360P win any benchmark in the head-to-head set?
A: No. The EPYC 9654P wins all six head-to-head tests, with the smallest margin being 436.4% in Cinebench R15 multi-core.
Q: Why are their average benchmark scores so close?
A: The EPYC 9654P averages 24,465, and the i7-1360P averages 24,344, a 0.5% difference. This occurs because the i7-1360P has PassMark scores in areas like data compression and encryption, while the EPYC’s benchmark list is limited to Cinebench and Geekbench.
Q: Can the EPYC 9654P be used in a laptop?
A: No. The EPYC 9654P uses AMD Socket SP5, has a 360W TDP, and is classified as a Server/Workstation chip. The i7-1360P uses Intel BGA 1744 with a 28W TDP and is a Mobile part.
Q: What memory types does each CPU support?
A: The EPYC 9654P supports DDR5 only with a twelve-channel bus. The i7-1360P supports both DDR4 and DDR5 with a dual-channel bus.
Q: Which CPU has integrated graphics?
A: Only the Intel Core i7-1360P, which includes Iris Xe Graphics 96EU. The AMD EPYC 9654P has no integrated graphics.
Where Each One Wins
The EPYC 9654P wins every multi-threaded and single-threaded Cinebench test by massive margins. It is the choice for server workloads, scientific computing, and any application that can use 96 cores and 192 threads. Its 460.8 GB/s memory bandwidth and 128 PCIe Gen 5 lanes support heavy data movement. The 384 MB of L3 cache is a server-grade asset for large datasets.
The i7-1360P wins in the fields where the EPYC has no data: PassMark tests. It scores 203,746 in data compression, 12,463 in data encryption, and 11,581 in extended instructions. It also has integrated graphics, a 5.00 GHz boost clock, and a 28W TDP, making it suitable for thin-and-light laptops. The i7-1360P supports both DDR4 and DDR5, giving OEMs flexibility, and its Gen 4 PCIe lanes are adequate for mobile peripherals. The EPYC cannot run on battery power, fit in a laptop chassis, or drive a display without a discrete GPU. The i7-1360P does all three. The data shows a clear split: raw compute belongs to AMD, mobile practicality belongs to Intel.