AMD A10-9700 vs Intel Pentium Gold G5620 Comparison

The Intel Pentium Gold G5620 and AMD A10-9700 are both end-of-life desktop processors, but they represent fundamentally different design philosophies. The data shows a clear overall winner, yet the story is more nuanced than a simple scoreboard. This analysis breaks down the benchmark results, architectural choices, and key specifications to determine which chip suits which workload.

Head-to-Head Benchmarks

The benchmark results are remarkably consistent, with the Intel Pentium Gold G5620 winning all five head-to-head comparisons. The most striking pattern is the uniformity of the victory margin. Across Cinebench R15, R20, and R23, the Intel chip’s lead hovers almost exactly at 20% in every test. In Cinebench R15 multicore, the Pentium Gold scores 364 against the A10-9700’s 302, a 20.5% advantage. That lead stretches slightly to 20.3% in Cinebench R20 multicore, with scores of 1517 and 1261 respectively.

The single-core results tell the same story. In Cinebench R20 single-core, the Intel part posts 214 versus 178, a 20.2% gap. Cinebench R23 single-core shows 510 against 424, again a 20.3% delta. The consistency of this margin across different benchmark generations suggests a fundamental per-clock performance advantage rather than a workload-specific quirk.

Interestingly, the A10-9700’s four physical cores do not translate into a multicore win against the Pentium Gold’s two cores with four threads. The Intel chip’s higher base clock of 4.00 GHz, compared to the AMD’s 3.50 GHz base and 3.80 GHz boost, appears to be the decisive factor. The Pentium Gold’s 20.3% lead in Cinebench R23 multicore (3613 vs 3004) demonstrates that raw thread count matters less than architectural efficiency in these workloads.

The overall average benchmark score reinforces this picture. The Pentium Gold G5620 averages 1045 points, placing it in the 29th percentile of all CPUs. The A10-9700 averages 1034 points, sitting just one percentile lower at 28th. That 11-point average difference is minimal, but the head-to-head deltas show where the real separation lies. The Intel chip’s nearest rivals include the AMD PRO A10-9700 at a 0% delta and the Intel Core i5-6350HQ at 0%, indicating it trades blows with those parts. The A10-9700’s closest competitor is the AMD A10-7890K at -0.1%, showing it sits in a slightly weaker position relative to its peer group.

Architecture Differences

The two processors come from completely different eras and design schools. The Intel Pentium Gold G5620 uses the Coffee Lake architecture built on a 14 nm process, while the AMD A10-9700 relies on the older Excavator architecture on a 28 nm node. This process gap explains much of the performance disparity—Intel’s smaller transistor size allows for higher clock speeds at lower power consumption.

The transistor counts tell a dramatic story. The AMD A10-9700 packs 3,100 million transistors into a 250 mm² die, while the Intel chip uses a much smaller 126 mm² die. The Intel part’s transistor count is not listed, but the die size difference alone suggests a far denser, more efficient design. The A10-9700’s larger die and older process explain its 65W TDP compared to the Pentium Gold’s 54W.

Cache configurations differ significantly. The Intel chip features 64 KB of L1 per core and 256 KB of L2 per core, plus 4 MB of shared L3 cache. The AMD part has 320 KB of total L1 and 2 MB of L2, but no L3 cache at all. This lack of L3 cache likely hurts the A10-9700 in repeated workloads and multi-threaded scenarios where shared data access matters.

The memory subsystems are more closely matched. Both support DDR4 memory over a dual-channel bus with identical 38.4 GB/s bandwidth. However, the Intel chip supports ECC memory while the AMD part does not. Both use PCIe Gen 3, but Intel provides 16 lanes (CPU only) versus AMD’s 8 lanes (CPU only). This halves the available bandwidth for discrete GPUs or NVMe drives on the AMD platform.

Integrated graphics differ as well. Intel pairs the Pentium Gold with UHD Graphics 630, while AMD includes Radeon R7 graphics. The benchmark data does not include iGPU tests, so a direct performance comparison is not possible from this data. The A10-9700’s Bristol Ridge codename and Excavator architecture date back to AMD’s pre-Zen designs, while Intel’s Coffee Lake is a much more recent implementation.

FAQ

Q: Which processor has the higher average benchmark score?

A: The Intel Pentium Gold G5620 averages 1045 points, placing it in the 29th percentile of all CPUs. The AMD A10-9700 averages 1034 points, sitting in the 28th percentile. The 11-point difference is small, but the Intel chip edges ahead.

Q: Does the AMD A10-9700’s four cores beat the Intel’s two cores in multicore tests?

A: No. The Intel Pentium Gold G5620 wins Cinebench R23 multicore with 3613 points against the AMD’s 3004, a 20.3% advantage. The Intel chip’s four threads plus higher base clock of 4.00 GHz overcome the AMD’s physical core count advantage.

Q: What is the single-core performance difference?

A: The Intel chip leads by 20.2% in Cinebench R20 single-core (214 vs 178) and by 20.3% in Cinebench R23 single-core (510 vs 424). This consistent ~20% gap indicates a clear per-clock performance advantage for Intel.

Q: Which processor has better power efficiency?

A: The Intel Pentium Gold G5620 has a 54W TDP versus the AMD A10-9700’s 65W TDP. Despite the lower power draw, the Intel chip delivers roughly 20% higher performance in all tested workloads, making it significantly more efficient.

Q: Do both processors support ECC memory?

A: No. The Intel Pentium Gold G5620 supports ECC memory, while the AMD A10-9700 does not. This could be a deciding factor for users building a low-cost workstation or home server.

Q: How do these chips compare to their nearest rivals?

A: The Intel chip’s nearest rival is the AMD PRO A10-9700 at a 0% delta and the Intel Core i5-6350HQ at 0%. The AMD A10-9700’s nearest rival is the AMD A10-7890K at -0.1% and the Intel Pentium G4560 at 0.3%, showing both chips sit in a similar performance tier overall.

The Verdict

The benchmark data is unambiguous: the Intel Pentium Gold G5620 is the faster processor in every measured category. It wins all five head-to-head benchmarks with margins ranging from 20.2% to 20.5%. The Intel chip achieves this with fewer physical cores (2 vs 4), a smaller die (126 mm² vs 250 mm²), and lower power consumption (54W vs 65W). For users prioritizing raw CPU performance in Cinebench workloads, the Pentium Gold G5620 is the clear choice.

The AMD A10-9700 still has a role, but it is defined by what it offers beyond CPU compute. It provides more physical cores, which could benefit highly threaded applications that scale differently than Cinebench. It also includes Radeon R7 integrated graphics, potentially offering stronger iGPU performance for light gaming or media use—though the data does not test this directly. The AMD chip’s 8 PCIe lanes versus Intel’s 16 may limit expansion, but its AM4 socket offers a different upgrade path.

The average benchmark scores tell a more nuanced story. With 1045 versus 1034 points, the overall performance gap is under 1.1%. The 28th versus 29th percentile ranking places both chips in the same performance tier relative to all CPUs. This suggests that for general desktop use, the difference may be less perceptible than the Cinebench deltas imply. Users who need ECC memory support should choose the Intel part. Users who value the AMD platform’s integrated graphics or need more physical cores for specific workloads may still consider the A10-9700.

The verdict from the data: pick the Intel Pentium Gold G5620 for superior single-core and multicore compute, lower power draw, and ECC support. Pick the AMD A10-9700 if the AM4 platform, Radeon R7 graphics, or the extra two physical cores are essential to your specific use case. For pure processing throughput, the Intel chip is the winner by a consistent 20%.

Specification Differences

| Specification | Intel Pentium Gold G5620 | AMD A10-9700 |

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

| Cores | 2 | 4 |

| Threads | 4 | 4 |

| Base Clock | 4.00 GHz | 3.50 GHz |

| Boost Clock | None | 3.80 GHz |

| TDP | 54W | 65W |

| Socket | Intel Socket 1151 | AMD Socket AM4 |

| Architecture | Coffee Lake | Excavator |

| Codename | Coffee Lake | Bristol Ridge |

| Process Node | 14 nm | 28 nm |

| Foundry | Intel | GlobalFoundries |

| Die Size | 126 mm² | 250 mm² |

| Transistors | Not listed | 3,100 million |

| L1 Cache | 64 KB (per core) | 320 KB |

| L2 Cache | 256 KB (per core) | 2 MB |

| L3 Cache | 4 MB (shared) | None |

| ECC Memory | Yes | No |

| PCIe | Gen 3, 16 Lanes (CPU only) | Gen 3, 8 Lanes (CPU only) |

| Integrated Graphics | UHD Graphics 630 | Radeon R7 |

| Release Date | 2019-02-19 | 2017-07-26 |

| Launch MSRP | $93 | Not listed |

| Part Number | SR3YC | AD9700AGM44AB |