CPU Comparison

# Head-to-Head Benchmarks

The benchmark data presents a consistent picture: the Intel Core i3-7100H wins all five head-to-head comparisons against the AMD FX-9830P, though the margins are uniformly narrow. The largest advantage appears in Cinebench R15 multi-core, where the Intel part scores 295 against AMD's 282, a 4.6% delta. That pattern repeats across every other tested workload, with the Intel chip holding a 4.5% lead in Cinebench R20 multi-core (1231 vs. 1178) and an identical 4.5% edge in Cinebench R23 multi-core (2932 vs. 2806).

Single-threaded performance tells the same story. The i3-7100H posts 173 in Cinebench R20 single-core versus 166 for the FX-9830P, a 4.2% advantage. In Cinebench R23 single-core, the gap widens slightly to 4.5%, with scores of 414 and 396 respectively. These are not dramatic victories, but they are consistent across every generation of the Cinebench test suite, suggesting a fundamental per-clock efficiency advantage rather than a workload-specific quirk.

The overall benchmark averages reinforce this narrow but real separation. The i3-7100H carries an average benchmark score of 1009, while the FX-9830P sits at 1006. That three-point gap places the Intel part in the 28th percentile of all CPUs, compared to the AMD part's 27th percentile. For context, both processors land in the same performance tier as the Intel Core i5-10210Y (1006 average score) and the Intel Processor N100 (1007). The i3-7100H's nearest rival, the Intel Core i7-2675QM, scores 1012, just 0.3% higher. The FX-9830P's closest competitor, the Intel Core i7-880, matches its 1006 average exactly.

It is worth noting that the AMD processor has additional benchmark data beyond the Cinebench suite. In Geekbench, the FX-9830P scores 1606 multi-core and 605 single-core. No direct Intel comparison is available in the head-to-head dataset, but these numbers provide a standalone reference point for the AMD part's capabilities outside the Cinebench workload family.

# Architecture Differences

The two processors come from fundamentally different design philosophies and manufacturing eras. The Intel Core i3-7100H is built on a 14 nm process at Intel's foundry, using the Kaby Lake architecture (specifically Kaby Lake-H). It packs two physical cores with four threads via Hyper-Threading. The AMD FX-9830P, by contrast, uses GlobalFoundries' 28 nm process and the Excavator architecture under the Bristol Ridge codename. It offers four physical cores but only four threads, with no simultaneous multi-threading.

The memory subsystem highlights a key divergence. The Intel chip supports both DDR3 and DDR4 memory, while the AMD part is limited to DDR4. Both use dual-channel memory buses, but the FX-9830P specifies a memory bandwidth of 38.4 GB/s, a figure not listed for the Intel part. Cache layouts also differ substantially. The i3-7100H provides 64 KB of L1 cache per core, 256 KB of L2 per core, and 3 MB of shared L3 cache. The FX-9830P has 320 KB of total L1 cache and 1 MB of L2 per module, with no L3 cache at all. This absence of L3 cache on the AMD part is a significant architectural disadvantage, particularly for workloads that benefit from larger shared pools of fast memory.

The integrated graphics solutions differ as well. Intel pairs the i3-7100H with HD Graphics 630, while AMD equips the FX-9830P with Radeon R7 8CU. Neither part includes support for ECC memory. PCIe connectivity shows another gap: the Intel chip provides 16 PCIe Gen 3 lanes from the CPU, while the AMD part halves that to 8 lanes.

The process node difference, 14 nm versus 28 nm, is reflected in physical characteristics. The Intel die measures 126 mm², while the AMD die is substantially larger at 250 mm². AMD lists 3,100 million transistors, whereas Intel does not disclose a transistor count. Both parts carry a 35 W TDP, which is notable given the Intel chip's smaller process node and die size. The AMD processor also offers a boost clock of 3.70 GHz against a base clock of 3.00 GHz, while the Intel part lists only a 2.80 GHz base clock with no boost figure provided.

# FAQ

Q: Which processor wins more benchmark comparisons?

A: The Intel Core i3-7100H wins all five head-to-head benchmarks against the AMD FX-9830P. The AMD part records zero wins in the head-to-head dataset.

Q: How large is the performance gap between the two?

A: The margin is consistent but modest. The Intel chip leads by 4.6% in Cinebench R15 multi-core and 4.5% in Cinebench R20 and R23 multi-core tests. Single-core deltas range from 4.2% to 4.5% in favor of Intel.

Q: Does the AMD FX-9830P have more cores than the Intel i3-7100H?

A: Yes. The AMD part has four physical cores, while the Intel chip has two. However, the Intel processor supports four threads via Hyper-Threading, matching the AMD part's total thread count at four.

Q: What are the average benchmark scores for each processor?

A: The Intel Core i3-7100H has an average benchmark score of 1009, placing it in the 28th percentile of all CPUs. The AMD FX-9830P averages 1006, sitting in the 27th percentile.

Q: Do both processors support the same memory types?

A: No. The Intel i3-7100H supports both DDR3 and DDR4 memory, while the AMD FX-9830P supports only DDR4. Both use dual-channel memory buses.

Q: Which processor has a smaller manufacturing process?

A: The Intel i3-7100H is built on a 14 nm process, compared to the AMD FX-9830P's 28 nm process. The Intel die is also smaller at 126 mm² versus 250 mm² for AMD.

# The Verdict

The data points to a clear, if narrow, recommendation for the Intel Core i3-7100H for most use cases. It wins every benchmark comparison in the dataset, with margins between 4.2% and 4.6%. The consistency of these results across Cinebench R15, R20, and R23, both single-core and multi-core, indicates that the Intel architecture delivers better per-clock performance despite having half the physical core count of the AMD part.

For users prioritizing multi-threaded workloads, the AMD FX-9830P's four physical cores might seem appealing on paper, but the benchmark results do not support that expectation. The Intel chip's two cores with four threads actually outperform the AMD part's four cores in multi-core tests. The i3-7100H scores 2932 in Cinebench R23 multi-core versus 2806 for the FX-9830P. The absence of L3 cache on the AMD part likely contributes to this outcome, as does the older 28 nm process node.

The single-core results reinforce the Intel advantage. A 4.2% to 4.5% lead in single-threaded tests means the i3-7100H will handle lightly threaded applications like everyday productivity and legacy software with greater responsiveness. The AMD part's higher boost clock of 3.70 GHz cannot overcome the Intel chip's superior instructions-per-clock efficiency.

The FX-9830P does offer one unique advantage: it has additional benchmark data in Geekbench, scoring 1606 multi-core and 605 single-core. This does not change the head-to-head verdict, but it provides a more complete performance profile for the AMD part. Additionally, the AMD processor integrates Radeon R7 8CU graphics, which may differ in capability from Intel's HD Graphics 630, though the benchmark data does not include graphics comparisons.

For buyers choosing between these two end-of-life mobile processors, the Intel Core i3-7100H is the safer pick based purely on compute benchmarks. Its consistent wins, higher average score, and better percentile ranking make it the stronger performer across the tested workloads. The AMD FX-9830P remains a viable option only if the specific system requirements favor its socket, integrated graphics, or memory bandwidth specification, factors outside the scope of these CPU benchmarks.

# Specification Differences

| Specification | Intel Core i3-7100H | AMD FX-9830P |

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

| Cores | 2 | 4 |

| Threads | 4 | 4 |

| Base Clock | 2.80 GHz | 3.00 GHz |

| Boost Clock | Not specified | 3.70 GHz |

| Socket | Intel BGA 1440 | AMD Socket FP4 |

| Architecture | Kaby Lake | Excavator |

| Codename | Kaby Lake-H | Bristol Ridge |

| Process Node | 14 nm | 28 nm |

| Foundry | Intel | GlobalFoundries |

| Die Size | 126 mm² | 250 mm² |

| Transistors | Not specified | 3,100 million |

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

| L2 Cache | 256 KB (per core) | 1 MB (per module) |

| L3 Cache | 3 MB (shared) | None |

| Memory Support | DDR3, DDR4 | DDR4 |

| Memory Bandwidth | Not specified | 38.4 GB/s |

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

| Integrated Graphics | Intel HD Graphics 630 | Radeon R7 8CU |

| Release Date | 2017-01-02 | 2016-05-30 |

| Launch MSRP | $225 | Not specified |

| Part Number | SR32T | FM983PAEY44AB |