AMD Ryzen 5 PRO 5655G vs Intel Core i9-10900K Comparison

The AMD Ryzen 5 PRO 5655G and Intel Core i9-10900K present a clear split: Intel dominates raw multi-threaded and compute-heavy workloads, while AMD wins decisively in encryption and single-threaded PassMark tests. The Intel part takes 14 of 17 head-to-head benchmarks, but the AMD chip’s victories are significant enough to define distinct use cases. Overall average benchmark scores are nearly identical — 28032 for the AMD versus 27872 for the Intel — placing both in the same performance tier despite their architectural differences.

Head-to-Head Benchmarks

The Intel Core i9-10900K is the clear winner in Cinebench across the board. In Cinebench R23 multi-core, Intel scores 18895 against AMD’s 17249, a margin of 8.7%. The same 8.7% delta repeats in R20 multi-core (7935 vs 7244), R15 multi-core (1904 vs 1738), and in every single-core Cinebench test — R23 single-core shows 2667 vs 2435, R20 shows 1120 vs 1022, and R15 shows 268 vs 245. This consistency indicates the Intel chip holds a uniform advantage in this rendering workload, likely driven by its higher core count and boost clock.

The gap widens substantially in PassMark’s integer-heavy tests. Intel wins integer math with 85404 versus 67561, a 20.9% lead. Floating point math shows a 27.2% advantage for Intel (53123 vs 38649), and extended instructions favor Intel by 27.1% (24620 vs 17944). Data compression is the largest single win for Intel: 375336 versus 255608, a 31.9% margin. Random string sorting delivers Intel’s biggest proportional victory at 45.6% (46389 vs 25253), while physics shows a 33.3% lead (1029 vs 686). Find prime numbers goes to Intel by 22.2% (63 vs 49).

The AMD Ryzen 5 PRO 5655G has two notable wins. PassMark data encryption is a landslide: AMD scores 15253 versus Intel’s 7796, a 95.7% advantage — nearly double the throughput. The other win is PassMark single-threaded, where AMD scores 3241 versus Intel’s 3113, a 4.1% edge. Notably, this single-thread result contradicts the Cinebench single-core results, where Intel leads by 8.6-8.7%. The data suggests AMD’s Zen 3 architecture handles certain single-threaded workloads more efficiently, even though its lower boost clock of 4.40 GHz trails Intel’s 5.30 GHz.

Across multi-threaded PassMark tests, Intel wins multithread with 22452 versus 19129 (14.8% lead), but the overall average benchmark scores remain close. The AMD chip sits at the 80th percentile of all CPUs, while Intel sits at the 79th, a negligible difference. The Intel part’s nearest rival is the AMD Ryzen 7 5800X3D with a delta of -0.1%, while AMD’s closest competitor is the Intel Core i9-12900H at +0.1% — both chips trade places within a fraction of a percent of their respective peers.

Architecture Differences

The foundational difference is process node and core count. AMD uses TSMC’s 7 nm process with 6 cores and 12 threads, while Intel uses its own 14 nm process with 10 cores and 20 threads. The Intel chip’s die size is 206 mm², larger than AMD’s 180 mm², but AMD packs 10,700 million transistors versus Intel’s unspecified count. AMD’s Zen 3 architecture (codename Cezanne) is a newer design than Intel’s Comet Lake, which explains how AMD achieves competitive scores with 40% fewer cores.

Cache configurations differ significantly. Both use 64 KB L1 per core, but AMD provides 512 KB L2 per core versus Intel’s 256 KB per core. For L3, AMD has 16 MB total, while Intel offers 20 MB shared. The larger per-core L2 on AMD likely contributes to its encryption and single-thread PassMark wins, while Intel’s larger shared L3 and extra cores help in compression and physics workloads.

Memory support is similar in type — both use DDR4 in dual-channel mode — but AMD’s memory bandwidth is rated higher at 51.2 GB/s versus Intel’s 46.9 GB/s. AMD also supports ECC memory, which Intel does not. Both use PCIe Gen 3 with 16 lanes from the CPU. Integrated graphics differ: AMD includes Radeon Vega 7, while Intel includes UHD Graphics 630. AMD’s TDP is 65 watts versus Intel’s 125 watts, a significant efficiency gap.

Production status separates the two: AMD’s Ryzen 5 PRO 5655G is listed as Active, while Intel’s Core i9-10900K is End-of-life. The release dates also differ by four years — AMD launched in May 2024, Intel in April 2020. The Intel part has an unlocked multiplier, while the AMD part is locked. Intel’s launch MSRP was $547; AMD has no listed launch MSRP.

FAQ

Q: Which CPU is faster in multi-core rendering?

A: Intel wins every Cinebench multi-core test by exactly 8.7%. R23 multi-core: 18895 vs 17249; R20: 7935 vs 7244; R15: 1904 vs 1738.

Q: Why does AMD win data encryption so decisively?

A: The AMD Ryzen 5 PRO 5655G scores 15253 in PassMark data encryption versus Intel’s 7796, a 95.7% advantage. This likely stems from Zen 3’s newer cryptographic instruction handling and larger per-core L2 cache.

Q: Is the Intel chip better for single-threaded performance?

A: It depends on the benchmark. Intel leads Cinebench R23 single-core by 8.7% (2667 vs 2435), but AMD wins PassMark single-threaded by 4.1% (3241 vs 3113). The results are workload-specific.

Q: What is the overall benchmark standing of each CPU?

A: AMD has an average benchmark score of 28032 and sits at the 80th percentile of all CPUs. Intel has an average score of 27872 and sits at the 79th percentile. They are effectively tied overall.

Q: Does the Intel chip have more cores and threads?

A: Yes. Intel has 10 cores and 20 threads, while AMD has 6 cores and 12 threads. Intel also has a higher boost clock at 5.30 GHz versus AMD’s 4.40 GHz.

Q: Which CPU supports ECC memory?

A: Only AMD supports ECC memory. Intel does not list ECC support. Both support DDR4 in dual-channel mode, but AMD’s memory bandwidth is 51.2 GB/s versus Intel’s 46.9 GB/s.

Specification Differences

| Specification | AMD Ryzen 5 PRO 5655G | Intel Core i9-10900K |

|---|---|---|

| Cores | 6 | 10 |

| Threads | 12 | 20 |

| Base Clock | 3.90 GHz | 3.70 GHz |

| Boost Clock | 4.40 GHz | 5.30 GHz |

| TDP | 65 W | 125 W |

| Process Node | 7 nm (TSMC) | 14 nm (Intel) |

| Die Size | 180 mm² | 206 mm² |

| Transistors | 10,700 million | Not listed |

| L2 Cache | 512 KB (per core) | 256 KB (per core) |

| L3 Cache | 16 MB | 20 MB (shared) |

| Memory Bandwidth | 51.2 GB/s | 46.9 GB/s |

| ECC Memory | Yes | No |

| Integrated Graphics | Radeon Vega 7 | UHD Graphics 630 |

| Production Status | Active | End-of-life |

| Release Date | 2024-05-06 | 2020-04-29 |

| Multiplier Unlocked | No | Yes |

| Launch MSRP | Not listed | $547 |

The Verdict

The data points to a straightforward conclusion: the Intel Core i9-10900K is the stronger multi-threaded workhorse, winning 14 of 17 head-to-head benchmarks with margins ranging from 8.7% to 45.6%. It leads in every Cinebench test, all major PassMark compute workloads (integer math, floating point, compression, physics, prime numbers, extended instructions), and multithread. Its 10 cores and 20 threads give it a structural advantage that AMD’s 6 cores cannot overcome in most tasks.

However, the AMD Ryzen 5 PRO 5655G is not a loser — it wins the two most specialized tests. Data encryption at 95.7% ahead is a massive edge, and its 4.1% PassMark single-thread win shows Zen 3’s efficiency. With a 65-watt TDP versus Intel’s 125 watts, AMD delivers comparable average scores (28032 vs 27872) at nearly half the power envelope. The Intel chip is also end-of-life, while AMD remains active.

For buyers prioritizing raw compute throughput, rendering, compression, and physics — the Intel part is the clear choice. For those needing encryption performance, lower power consumption, ECC memory support, or a current production CPU, the AMD chip is the better fit. The overall average scores are so close that neither chip dominates the other in general-purpose use; the decision hinges entirely on workload.

Where Each One Wins

Pick the AMD Ryzen 5 PRO 5655G if: your workload involves data encryption, where it delivers 95.7% higher PassMark throughput. Its 4.1% PassMark single-thread lead also makes it suitable for lightly threaded applications that respond to that specific benchmark profile. The 65-watt TDP and ECC memory support are operational advantages for power-constrained or reliability-focused systems. The chip remains in active production, ensuring ongoing availability.

Pick the Intel Core i9-10900K if: you need maximum multi-threaded performance in Cinebench (8.7% ahead in all versions), integer math (20.9% ahead), floating point (27.2% ahead), or data compression (31.9% ahead). Its physics score is 33.3% higher, and random string sorting leads by 45.6%. The 10-core, 20-thread configuration with a 5.30 GHz boost clock makes it the superior choice for rendering, simulation, and throughput-heavy tasks. The unlocked multiplier also allows overclocking, which the locked AMD chip does not permit.