AMD A8-7670K vs Intel Celeron N5105 Comparison

Head-to-Head Benchmarks

The recorded data shows a clean sweep for the Intel Celeron N5105 across every shared benchmark. In Cinebench R15 multi-core, the N5105 scores 341 against 276 for the AMD A8-7670K, a 23.6% advantage. The same pattern holds in Cinebench R20 multi-core, where Intel leads 1422 to 1152, a 23.4% margin.

Single-core results are equally one-sided. The N5105 posts 200 in Cinebench R20 single-core versus 162 for the A8-7670K, a 23.5% lead. In Cinebench R23, the Intel part scores 478 single-core and 3388 multi-core, while the AMD chip manages 387 and 2745 respectively. Both deltas sit at 23.5% and 23.4% in favor of Intel.

The consistency of these margins is notable. Across five head-to-head tests, the Celeron N5105 wins all five, with every delta falling between 23.4% and 23.6%. This is not a case of one workload favoring a particular architecture; the Intel processor simply executes Cinebench workloads faster across the board.

Looking at average benchmark scores in the database, the N5105 sits at 955, slightly ahead of the A8-7670K at 944. The Intel chip's nearest rivals include the Intel Xeon W3570 at 955, the Intel Xeon W3550 at 956, and the AMD Athlon X4 870K at 956, all within 0.2% of its score. The AMD part's closest competitors are the Intel Core i3-1110G4 at 945, the AMD Athlon X4 860K at 943, and the AMD Phenom II X6 1045T at 946. Both CPUs land in the 25th percentile of all processors in the database, indicating they occupy a similar overall performance tier despite the Intel chip's clear win in these specific tests.

Architecture Differences

The two processors come from different design eras and philosophies. The Intel Celeron N5105 uses the Tremont architecture on the Jasper Lake codename, built on a 10 nm process by Intel. It has 4 cores and 4 threads, with a base clock of 2000 MHz and a boost clock of 2900 MHz. The AMD A8-7670K uses the Steamroller architecture on the Godaveri codename, fabricated on a 28 nm process by GlobalFoundries. It also has 4 cores and 4 threads, but runs at a higher base clock of 3600 MHz and a boost clock of 3900 MHz.

Cache configurations differ substantially. The N5105 packs 64 KB of L1 cache per core, 1.5 MB of shared L2 cache, and 4 MB of shared L3 cache. The A8-7670K has 256 KB of L1 cache and 4 MB of L2 cache, with no L3 cache at all. The die size tells a story of process efficiency: the Intel chip measures 63.8 mm², while the AMD die spans 245 mm² and contains 2,411 million transistors.

Memory support diverges as well. The N5105 supports DDR4 and LPDDR4 memory on a dual-channel bus, delivering 46.9 GB/s of bandwidth. The A8-7670K is limited to DDR3 on a dual-channel bus, with 34.1 GB/s of bandwidth. PCIe connectivity also differs: the Intel part offers Gen 3 with 8 lanes from the CPU, while the AMD part offers Gen 3 with 16 lanes.

Integrated graphics represent another split. The N5105 includes UHD Graphics with 24 execution units, while the A8-7670K carries a Radeon R7 GPU. The Intel chip uses the BGA 1338 socket and targets the mobile segment, whereas the AMD chip uses Socket FM2+ and is a desktop part. The AMD processor has an unlocked multiplier, a feature absent on the Intel chip. The A8-7670K was released in July 2015, while the N5105 has no recorded release date in the database. Both are end-of-life products.

FAQ

Q: Which processor is faster in multi-core Cinebench workloads?

A: The Intel Celeron N5105 wins every multi-core test. It leads by 23.6% in Cinebench R15, 23.4% in R20, and 23.4% in R23.

Q: Does the AMD A8-7670K win any benchmark?

A: No. Across the five head-to-head tests recorded, the AMD chip wins zero. The Intel chip wins all five.

Q: How do their average benchmark scores compare?

A: The Celeron N5105 averages 955, while the A8-7670K averages 944. Both sit in the 25th percentile of all CPUs in the database.

Q: What memory types does each support?

A: The N5105 supports DDR4 and LPDDR4 with 46.9 GB/s bandwidth. The A8-7670K supports DDR3 with 34.1 GB/s bandwidth.

Q: Which processor has an unlocked multiplier?

A: The AMD A8-7670K has an unlocked multiplier. The Intel Celeron N5105 does not.

Q: What are their thermal design power ratings?

A: The N5105 is rated at 10 watts TDP, while the A8-7670K is rated at 95 watts TDP.

The Verdict

The benchmark data points to a clear conclusion: the Intel Celeron N5105 outperforms the AMD A8-7670K in every measured Cinebench workload. The margin is remarkably uniform, hovering around 23.5% regardless of single-core or multi-core testing. This consistency suggests a fundamental architectural advantage rather than a workload-specific quirk.

The N5105 achieves this with a much lower 10 watt TDP compared to the A8-7670K's 95 watt TDP. It also delivers higher memory bandwidth at 46.9 GB/s versus 34.1 GB/s, despite using a smaller die and a more advanced 10 nm process. The AMD chip fights back with a higher base clock (3600 MHz versus 2000 MHz) and more PCIe lanes (16 versus 8), but these advantages do not translate into Cinebench performance wins.

For anyone choosing between these two end-of-life processors, the Intel Celeron N5105 is the stronger performer in CPU-bound rendering tasks. The AMD A8-7670K's only clear advantages are its unlocked multiplier for overclocking and its larger PCIe lane count, but the recorded benchmark data shows no scenario where those features produce a win.

Specification Differences

| Specification | Intel Celeron N5105 | AMD A8-7670K |

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

| Architecture | Tremont | Steamroller |

| Codename | Jasper Lake | Godaveri |

| Process Node | 10 nm | 28 nm |

| Foundry | Intel | GlobalFoundries |

| Base Clock | 2000 MHz | 3600 MHz |

| Boost Clock | 2900 MHz | 3900 MHz |

| TDP | 10 W | 95 W |

| Socket | Intel BGA 1338 | AMD Socket FM2+ |

| L1 Cache | 64 KB (per core) | 256 KB |

| L2 Cache | 1.5 MB (shared) | 4 MB |

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

| Die Size | 63.8 mm² | 245 mm² |

| Transistors | Not recorded | 2,411 million |

| Memory Support | DDR4, LPDDR4 | DDR3 |

| Memory Bandwidth | 46.9 GB/s | 34.1 GB/s |

| PCIe | Gen 3, 8 Lanes | Gen 3, 16 Lanes |

| Integrated Graphics | UHD Graphics 24EU | Radeon R7 |

| Market Segment | Mobile | Desktop |

| Multiplier Unlocked | No | Yes |

| Release Date | Not recorded | 2015-07-19 |

Where Each One Wins

The Intel Celeron N5105 wins in all CPU rendering benchmarks based on the recorded data. Its 23.4% to 23.6% advantages across Cinebench R15, R20, and R23 make it the default choice for any workload that resembles these multi-threaded and single-threaded rendering tasks. The N5105 also offers higher memory bandwidth, which can benefit memory-sensitive applications, and a dramatically lower TDP that suits compact or thermally constrained systems.

The AMD A8-7670K has no benchmark wins in this comparison, but its specification sheet reveals areas where it holds theoretical advantages. The unlocked multiplier allows manual overclocking, which could narrow the gap in the hands of a user willing to push clock speeds. The 16 PCIe Gen 3 lanes provide more expansion headroom for add-in cards compared to the N5105's 8 lanes. The A8-7670K also uses the desktop-oriented Socket FM2+ platform, which may offer different upgrade paths or board features compared to the mobile BGA 1338 socket of the Intel part.

In practical terms, the data supports picking the Intel Celeron N5105 for raw compute performance in Cinebench-type workloads. The AMD A8-7670K remains relevant only for users who value its overclocking capability, broader PCIe connectivity, or desktop platform characteristics. For straight benchmark performance, the N5105 is the winner in every recorded test.