AMD EPYC 9654 vs Intel Core i5-12600H Comparison

Head-to-Head Benchmarks

The recorded data shows a complete sweep for the AMD EPYC 9654 across every Cinebench test in the comparison. The smallest margin is a 406.6% advantage in Cinebench R15 multi-core, where the EPYC scores 10248 against the Core i5-12600H's 2023. The single-core results are equally decisive, with the EPYC 9654 leading by 407.4% in Cinebench R15 single-core (1446 vs 285). This is not a close contest by any measure; the EPYC 9654 outperforms the Core i5-12600H by more than five times in every benchmark recorded.

The Cinebench R20 results follow the same pattern. In multi-core, the EPYC 9654 produces 42703 points versus 8430 for the Intel part, a 406.6% difference. Single-core in R20 shows 6028 against 1189, a 407% gap. The R23 results close out the data with 101675 versus 20072 in multi-core (406.6% delta) and 14354 versus 2833 in single-core (406.7% delta). Across all six head-to-head tests, the EPYC 9654 wins 6 and the Core i5-12600H wins 0.

The consistency of the delta percentages, all hovering near 406-407%, indicates that the performance gap scales uniformly across workloads. The EPYC 9654's advantage is not workload-specific; it is fundamental to the design difference between the two processors. The Core i5-12600H's best relative showing is its single-core results, but even there it trails by over 400%. The data suggests that no single-threaded test in this set can close the gap.

Average benchmark scores place the EPYC 9654 at 29409, which sits within a tight cluster of rivals: the AMD Ryzen 7 3800X scores 29447 (0.1% ahead), the AMD Ryzen AI 7 345 scores 29461 (0.2% ahead), the Intel Core i5-13500HX scores 29270 (0.5% behind), and the Intel Xeon Phi 7210 scores 29245 (0.6% behind). The Core i5-12600H's average of 28882 places it near the Intel Core i9-13900H (28886, 0% delta), the AMD Ryzen 5 5600XT (28940, 0.2% behind), the Intel Core i7-12650H (28815, 0.2% ahead), and the Intel Core Ultra 5 135H (29093, 0.7% behind). Both processors land in the 81st percentile of all CPUs, meaning they occupy similar positions relative to the full database, but their head-to-head results show no overlap in performance.

Architecture Differences

The AMD EPYC 9654 is built on the Zen 4 architecture with the Genoa codename, part of the EPYC 9004 series. It uses a 5 nm process from TSMC and packs 96 cores with 192 threads. The Core i5-12600H uses Intel's Alder Lake-H architecture, built on a 10 nm process at Intel, with 12 cores and 16 threads. The core count difference is a factor of eight, and the thread count difference is a factor of twelve. This alone explains the multi-core margins.

Cache hierarchies differ substantially. The EPYC 9654 has 64 KB of L1 per core, 1 MB of L2 per core, and 384 MB of shared L3. The Core i5-12600H has 80 KB of L1 per core, 1.25 MB of L2 per core, and 18 MB of shared L3. The EPYC's total L3 is more than twenty times larger. The Intel part has a slightly larger per-core L1 and L2, but the EPYC's massive shared pool gives it a clear advantage in cache-sensitive server workloads.

Clock speeds tell a different story. The Core i5-12600H has a base clock of 2.70 GHz and a boost clock of 4.50 GHz. The EPYC 9654 has a base clock of 2.40 GHz and a boost clock of 3.70 GHz. The Intel chip boosts 0.80 GHz higher, yet still loses single-core tests by over 400%. This indicates that the EPYC's per-core performance, despite lower clocks, is dramatically higher, likely due to architectural efficiency and the much larger cache. The benchmark data does not support a single-core victory for the Intel part at any clock speed.

Memory and I/O configurations reinforce the server versus mobile split. The EPYC 9654 supports DDR5 with a twelve-channel memory bus and a recorded memory bandwidth of 460.8 GB/s. The Core i5-12600H supports DDR4 and DDR5 over a dual-channel bus, with no bandwidth figure recorded. The EPYC also offers PCIe Gen 5 with 128 lanes (CPU only), while the Intel part offers PCIe Gen 4 with 20 lanes (CPU only). The EPYC supports ECC memory; the Intel part does not. The socket types reflect their markets: the EPYC uses AMD Socket SP5, while the Core i5-12600H uses Intel BGA 1744.

The EPYC 9654 has a TDP of 360 watts, a die made of 12x 72 mm² chiplets, and 78,840 million transistors. The Core i5-12600H has a TDP of 45 watts and a die size of 217 mm². The EPYC's transistor count is enormous, though no figure is recorded for the Intel part. The Core i5-12600H includes integrated graphics (Iris Xe 80EU); the EPYC 9654 has none. The EPYC's launch MSRP is $11805, while no launch MSRP is recorded for the Intel part.

Where Each One Wins

The EPYC 9654 wins every Cinebench test in the head-to-head data, so the question is not whether it wins, but where its margins are largest. The biggest single margin is 407.4% in Cinebench R15 single-core, followed closely by 407% in R20 single-core and 406.7% in R23 single-core. Multi-core margins are 406.6% across R15, R20, and R23. The EPYC's single-core lead is slightly larger than its multi-core lead, which is notable for a 96-core server part. This suggests that the Zen 4 architecture delivers exceptional per-thread performance even at lower clock speeds.

The Core i5-12600H has no benchmark wins in this data set. Its recorded PassMark results are not compared head-to-head with the EPYC, but they provide context for its strengths. The Core i5-12600H scores 247404 in data compression, 14799 in data encryption, 14508 in extended instructions, 71 in find prime numbers, 51264 in floating point math, 71168 in integer math, 21128 in multithread, 1262 in physics, 27653 in random string sorting, and 3453 in single-thread. These are standalone figures without EPYC equivalents, so they cannot be used for direct comparison. However, they show that the Core i5-12600H is a capable mobile processor in its own right, with strong integer and floating point scores for a 45-watt part.

The use-case split is therefore clear. The EPYC 9654 is designed for workloads that scale with cores, threads, memory bandwidth, and cache capacity: database serving, virtualization, scientific computing, and high-throughput enterprise applications. The Core i5-12600H is designed for mobile systems where power consumption and integrated graphics matter, but the data shows it cannot match the EPYC in any measured Cinebench metric. The 360-watt TDP of the EPYC versus the 45-watt TDP of the Core i5-12600H highlights the trade-off: the EPYC demands server-class cooling and power delivery, while the Core i5-12600H fits in a laptop chassis.

The Verdict

The data is unambiguous. The AMD EPYC 9654 outperforms the Intel Core i5-12600H by approximately 406-407% in every Cinebench benchmark recorded. The EPYC 9654 wins 6 tests; the Core i5-12600H wins 0. Anyone selecting between these two processors for raw compute throughput should choose the EPYC 9654 without hesitation. Its 96 cores, 192 threads, 384 MB of L3 cache, twelve-channel DDR5 memory, and 128 PCIe Gen 5 lanes make it a server-class part with no competition in this comparison.

The Core i5-12600H's advantages are not visible in the head-to-head data. It has a higher boost clock, a much lower TDP, integrated graphics, and support for both DDR4 and DDR5. It also fits in a mobile form factor with a BGA 1744 socket. But none of these attributes translate into a benchmark win against the EPYC 9654. For a mobile user who needs a capable processor for everyday tasks, the Core i5-12600H may be appropriate, but that decision cannot be made from the recorded benchmark data, which shows no scenario where the Intel part leads.

The percentile rankings are identical at 81 for both CPUs, and their average scores are close (29409 for the EPYC, 28882 for the Core i5-12600H). This illustrates how average scores can obscure extreme pairwise differences. The EPYC 9654's nearest rivals include the Ryzen 7 3800X and Ryzen AI 7 345, both within 0.2%. The Core i5-12600H's nearest rivals include the Core i9-13900H and Core Ultra 5 135H, both within 0.7%. In the broader database, these two processors occupy similar tiers, but head-to-head, the EPYC is in a different class. The verdict from the data: choose the EPYC 9654 for any workload where compute density matters, and choose the Core i5-12600H only if the platform constraints (mobile, low power, integrated graphics) rule out the EPYC entirely.

FAQ

Q: How much faster is the AMD EPYC 9654 than the Intel Core i5-12600H in multi-core tests?

A: The EPYC 9654 leads by 406.6% in Cinebench R15, R20, and R23 multi-core tests. Scores are 10248 vs 2023 in R15, 42703 vs 8430 in R20, and 101675 vs 20072 in R23.

Q: Does the Intel Core i5-12600H win any single-core test?

A: No. The EPYC 9654 wins all three single-core tests, with deltas of 407.4% in R15, 407% in R20, and 406.7% in R23.

Q: What are the core and thread counts for each processor?

A: The AMD EPYC 9654 has 96 cores and 192 threads. The Intel Core i5-12600H has 12 cores and 16 threads.

Q: Which processor supports ECC memory?

A: The AMD EPYC 9654 supports ECC memory. The Intel Core i5-12600H does not.

Q: What is the launch MSRP of the AMD EPYC 9654?

A: The launch MSRP is $11805.

Q: How do the two processors compare in average benchmark score and percentile?

A: The EPYC 9654 has an average benchmark score of 29409 and is in the 81st percentile. The Core i5-12600H has an average of 28882 and is also in the 81st percentile.