AMD Ryzen 5 PRO 5655G vs Intel Core i9-12900H Comparison

The AMD Ryzen 5 PRO 5655G and Intel Core i9-12900H sit at nearly identical average benchmark scores—28,032 for the AMD part and 28,003 for the Intel part—a difference of just 0.1%. This places both in the 80th percentile of all CPUs, yet the head-to-head data reveals a dramatic split in workload-specific performance. The Intel chip wins 15 of 17 direct comparisons, while the AMD chip takes two decisive victories in the newer Cinebench R23 tests, including a remarkable 30.7% single-core lead. This is not a story of overall parity but of two very different performance profiles.

Head-to-Head Benchmarks

The Intel Core i9-12900H dominates the multi-core and math-heavy workloads. In Cinebench R15 multicore, Intel scores 2,576.5 against AMD’s 1,738, a 32.5% advantage. The gap narrows in Cinebench R20 multicore but remains substantial: 9,501 versus 7,244, or 23.8% ahead. PassMark tests reinforce this pattern—Intel leads by 43.1% in floating point math (67,940 vs 38,649), by 28% in integer math (93,828 vs 67,561), and by 30% in multithread score (27,340 vs 19,129). The most extreme disparity appears in PassMark physics, where Intel posts 1,753 against AMD’s 686—a 60.9% gap that suggests the Intel architecture handles simulated physics calculations far more efficiently.

The Intel chip also wins every single-thread PassMark test, taking a 12.8% lead in both single_thread and singlethread scores (3,717 vs 3,241). In Cinebench R15 singlecore, Intel leads 266 to 245, a 7.9% margin, and in R20 singlecore the lead grows to 23.8% (1,341 vs 1,022). Data compression and encryption also favor Intel, with 19.3% and 17.8% advantages respectively. Even in extended instructions, where the margin is smallest, Intel still wins by 4.9% (18,875 vs 17,944). Prime number finding shows Intel at 107 versus AMD’s 49—a 54.2% lead that highlights differences in integer-heavy algorithmic work.

The AMD Ryzen 5 PRO 5655G’s two wins are concentrated in Cinebench R23, the most recent rendering benchmark in the set. Its multicore score of 17,249 beats Intel’s 15,946.5 by 8.2%. More striking is the singlecore result: AMD scores 2,435, which is 30.7% higher than Intel’s 1,863.5. This suggests that while Intel’s hybrid architecture excels in many legacy and synthetic workloads, AMD’s Zen 3 cores maintain a significant advantage in the specific rendering tasks measured by Cinebench R23. The data implies that the R23 test may be more sensitive to per-core efficiency rather than raw thread count, which would explain why AMD’s six cores outperform Intel’s fourteen in that particular benchmark.

FAQ

Q: Which CPU has the higher overall average benchmark score?

A: The AMD Ryzen 5 PRO 5655G has a marginally higher average benchmark score of 28,032, compared to the Intel Core i9-12900H’s 28,003. The difference is 0.1%, placing both at the 80th percentile of all CPUs.

Q: How large is Intel’s lead in the PassMark multithread test?

A: Intel leads by 30% in PassMark multithread, scoring 27,340 against AMD’s 19,129. This is one of the larger gaps in the entire benchmark set, alongside the 43.1% lead in floating point math and the 60.9% lead in physics.

Q: Where does the AMD chip achieve its biggest win?

A: AMD’s largest victory is in Cinebench R23 singlecore, where it scores 2,435 versus Intel’s 1,863.5—a 30.7% advantage. This is also the single biggest percentage-based win for either chip across all benchmarks.

Q: Does Intel win every single-thread test?

A: No. Intel wins all three single-thread PassMark and Cinebench R15/R20 tests, but AMD wins Cinebench R23 singlecore. In that specific test, AMD’s 2,435 beats Intel’s 1,863.5 by 30.7%, reversing the trend seen elsewhere.

Q: What is the closest benchmark result between the two?

A: The closest result is in PassMark extended instructions, where Intel leads by only 4.9% (18,875 vs 17,944). This is the only head-to-head test with a margin under 7.9%.

Q: How do the two chips compare in data compression?

A: Intel is ahead by 19.3% in PassMark data compression, scoring 316,714 against AMD’s 255,608. Intel also leads data encryption by 17.8% (18,548 vs 15,253).

Where Each One Wins

The Intel Core i9-12900H is the clear choice for multi-threaded productivity and math-intensive tasks. Its wins span 15 benchmarks, including all PassMark tests and the older Cinebench R15/R20 multi-core variants. The data shows Intel is particularly strong in physics simulation (60.9% lead), prime number finding (54.2% lead), and floating point math (43.1% lead). For workloads involving data compression, encryption, random string sorting, or integer math, Intel’s 14 cores and 20 threads provide a consistent edge. The 3DMark results, while not directly compared head-to-head in the fact pack, show Intel’s scaling from 999 single-thread to 6,927 max-threads, indicating robust thread scaling behavior.

The AMD Ryzen 5 PRO 5655G wins only in Cinebench R23, both multi-core and single-core. This makes it the preferred option for rendering workloads that use the R23 engine, particularly single-threaded rendering where its 30.7% lead is decisive. The 8.2% multi-core win in R23 is notable because it suggests that AMD’s six cores with 12 threads can outperform Intel’s fourteen cores in this specific test, likely due to higher per-core efficiency or better clock behavior under the R23 load. Users running modern rendering pipelines that rely on R23-style workloads would see better results on the AMD chip.

Specification Differences

The two processors differ significantly in core configuration and physical attributes. The AMD Ryzen 5 PRO 5655G has 6 cores and 12 threads, while the Intel Core i9-12900H has 14 cores and 20 threads. Base clocks are 3.90 GHz for AMD versus 2.50 GHz for Intel, but Intel boosts to 5.00 GHz compared to AMD’s 4.40 GHz. Thermal design power differs: AMD is rated at 65W, Intel at 45W. The AMD chip uses the AMD Socket AM4, while Intel uses Intel BGA 1744. Memory support diverges—AMD supports only DDR4, whereas Intel supports both DDR4 and DDR5. AMD’s memory bandwidth is listed at 51.2 GB/s; Intel’s is not specified. ECC memory is supported by AMD but not by Intel. PCIe generations differ as well: AMD offers Gen 3 with 16 lanes (CPU only), while Intel offers Gen 4 with 20 lanes. The integrated graphics also differ: AMD uses Radeon Vega 7, Intel uses Iris Xe 96EU. AMD’s part number is 100-000001513, Intel’s is SRLD4. Neither chip has an unlocked multiplier.

Architecture Differences

The architectural divide is stark. AMD builds on a 7 nm process from TSMC, with a die size of 180 mm² and 10,700 million transistors. Intel uses a 10 nm process from its own foundry, with a die size of 217 mm² and no transistor count listed. AMD’s architecture is Zen 3, codenamed Cezanne, part of the 5000 series. Intel’s architecture is Alder Lake, codenamed Alder Lake-H, part of the Core 12th Gen. Cache hierarchies differ: AMD has 64 KB L1 per core, 512 KB L2 per core, and 16 MB L3. Intel has 80 KB L1 per core, 1.25 MB L2 per core, and 24 MB L3 shared. The release dates reveal a generation gap—AMD launched on 2024-05-06, while Intel launched on 2022-01-03. AMD’s market segment is Desktop, whereas Intel’s is Mobile. Both are in active production. The transistor count difference (10,700 million for AMD, unspecified for Intel) and die size difference (180 mm² vs 217 mm²) suggest AMD packs more transistors into a smaller area, which may explain its efficiency in single-thread R23 performance.

The Verdict

The data points to a clear split: choose the Intel Core i9-12900H for most compute-heavy tasks, and the AMD Ryzen 5 PRO 5655G for Cinebench R23 rendering workloads. Intel’s 15-2 win record in head-to-head tests is decisive, with leads exceeding 30% in physics, prime numbers, and floating point math. The Intel chip’s higher thread count and boost clock of 5.00 GHz give it a broad edge across PassMark’s diverse test suite. Users running simulations, data processing, encryption, or general multi-threaded software will see measurably better performance on Intel.

However, the AMD chip’s 30.7% single-core lead in Cinebench R23 is not a trivial anomaly. It suggests that for applications built around the R23 rendering engine—which includes many modern 3D modeling and animation tools—AMD’s Zen 3 architecture delivers superior per-core performance. The 8.2% multi-core win in R23 also indicates that Intel’s thread advantage does not always translate to real-world rendering gains. If a user’s primary workload is R23-based rendering, the AMD Ryzen 5 PRO 5655G is the better pick despite losing the broader benchmark war. For everything else in this comparison, the Intel Core i9-12900H is the data-backed choice.