AMD Ryzen 5 40 vs NVIDIA DGX Spark (MediaTek AHJ11488B) Comparison

Head-to-Head Benchmarks

The recorded data presents an unusual comparison. The AMD Ryzen 5 40 has a full suite of benchmark results across Cinebench and Passmark tests, while the NVIDIA DGX Spark (MediaTek AHJ11488B) has no recorded benchmark scores in the database. This makes a direct numerical head-to-head impossible. The only comparative inference available comes from the broader database percentile rankings. The Ryzen 5 40 sits at the 70th percentile of all CPUs, while the DGX Spark sits at the 50th percentile. That gap indicates that, based on aggregate database standings, the Ryzen 5 40 outperforms the DGX Spark in general CPU workloads.

Focusing solely on the available numbers for the Ryzen 5 40, its strongest results appear in specific Passmark tests. The data compression score reaches 141,533, and the multithread score is 9,341. Integer math finishes at 31,598, and floating point math at 15,194. Single-thread performance is recorded at 2,477. In Cinebench, the R23 multicore score is 4,841 and the single-core score is 1,150. The older R15 tests show 790 for multicore and 165.5 for single-core. These figures establish a baseline for a 4-core, 8-thread mobile processor.

Because the DGX Spark lacks any benchmark entries, no delta percentages can be computed between the two. The database lists no head-to-head benchmark wins for either product. The analysis must rely on architectural specifications and the percentile placement to draw conclusions about relative standing.

Architecture Differences

The two processors diverge substantially in their underlying design. The AMD Ryzen 5 40 uses the Zen 2 architecture with the Mendocino codename, built on a 6 nm process at TSMC. It contains 4 cores and 8 threads. The NVIDIA DGX Spark uses the GB10 codename with a 3 nm TSMC process, and it contains 20 cores with 20 threads. The core count difference is significant: the DGX Spark has 5 times the core count, though without simultaneous multithreading, its thread count equals its core count at 20.

Cache hierarchies differ as well. The Ryzen 5 40 has 64 KB of L1 per core, 512 KB of L2 per core, and 4 MB of shared L3. The DGX Spark has 128 KB of L1 per core, 2 MB of L2 per core, and 8 MB of shared L3. The per-core cache allocations are larger on the DGX Spark, and its shared L3 is double that of the Ryzen. The die size also differs: the Ryzen measures 100 mm², while the DGX Spark measures 208 mm².

Clock speeds show a mixed picture. Both processors have a base clock of 2.80 GHz. The Ryzen 5 40 boosts to 4.30 GHz, while the DGX Spark boosts to 4.00 GHz. The Ryzen has a 0.30 GHz higher boost clock, which can benefit lightly threaded workloads. Power envelopes diverge sharply. The Ryzen 5 40 has a TDP of 15 watts, while the DGX Spark has a TDP of 140 watts. That is a nearly tenfold difference in thermal design power.

Memory support also separates the two. The Ryzen 5 40 uses LPDDR5 memory on a dual-channel bus with 88.0 GB/s of bandwidth. The DGX Spark uses LPDDR5X on a quad-channel bus with 273.1 GB/s of bandwidth. The DGX Spark delivers more than three times the memory bandwidth. Neither processor supports ECC memory. PCIe connectivity is limited to Gen 3 with 4 lanes on the Ryzen, while the DGX Spark lists PCIe as N/A. Integrated graphics differ as well: the Ryzen uses Radeon 610M, while the DGX Spark uses the GB20B "Blackwell" GPU.

The release dates sit close together. The Ryzen 5 40 launched on September 30, 2025, and the DGX Spark followed on October 14, 2025. Both are listed as active production parts. The DGX Spark has a launch MSRP of $3999, while the Ryzen has no recorded launch MSRP. The market segment for both is mobile.

Where Each One Wins

The data supports distinct strengths for each processor based on specifications and recorded performance. The Ryzen 5 40 wins in single-thread clock speed. Its 4.30 GHz boost clock exceeds the DGX Spark's 4.00 GHz boost. In lightly threaded tasks that depend on peak frequency, the Ryzen has an advantage. Its recorded single-thread Passmark score of 2,477 and Cinebench R23 single-core score of 1,150 provide a concrete performance baseline, even without a direct rival score to compare.

The Ryzen 5 40 also wins on power efficiency. At 15 watts TDP, it consumes a fraction of the DGX Spark's 140 watts. For mobile workloads where battery life and thermals matter, the Ryzen is clearly the more restrained option. Its smaller die size of 100 mm² further supports a compact, low-power design.

The DGX Spark wins decisively on raw core count and memory bandwidth. With 20 cores versus 4, and 273.1 GB/s memory bandwidth versus 88.0 GB/s, the DGX Spark is built for parallel throughput. The quad-channel memory bus and LPDDR5X support give it a clear edge in memory-intensive workloads. Its 3 nm process node is more advanced than the Ryzen's 6 nm node, which allows for the higher core count within a 208 mm² die. The larger L2 cache per core (2 MB versus 512 KB) and larger L3 cache (8 MB versus 4 MB) also favor the DGX Spark in workloads with large working sets.

The percentile data adds context. The Ryzen 5 40 ranks at the 70th percentile of all CPUs, while the DGX Spark ranks at the 50th percentile. Despite having far fewer cores, the Ryzen's higher percentile suggests that its measured benchmark performance places it above the DGX Spark in the database's overall CPU ranking. This is notable because the DGX Spark's architectural specifications are substantially more robust on paper, but its lack of recorded benchmark scores leaves its actual measured performance unverified.

FAQ

Q: Does the NVIDIA DGX Spark have any benchmark scores in the database?

A: No. The DGX Spark has an empty benchmark list, while the AMD Ryzen 5 40 has 15 recorded benchmark scores across Cinebench and Passmark tests.

Q: Which processor has more cores?

A: The NVIDIA DGX Spark has 20 cores and 20 threads. The AMD Ryzen 5 40 has 4 cores and 8 threads.

Q: What is the TDP difference between the two?

A: The AMD Ryzen 5 40 has a TDP of 15 watts. The NVIDIA DGX Spark has a TDP of 140 watts.

Q: Which processor has higher memory bandwidth?

A: The NVIDIA DGX Spark has 273.1 GB/s of memory bandwidth on a quad-channel LPDDR5X bus. The AMD Ryzen 5 40 has 88.0 GB/s on a dual-channel LPDDR5 bus.

Q: What are the percentile rankings?

A: The AMD Ryzen 5 40 is at the 70th percentile of all CPUs. The NVIDIA DGX Spark is at the 50th percentile.

Q: Which processor has a higher boost clock?

A: The AMD Ryzen 5 40 boosts to 4.30 GHz. The NVIDIA DGX Spark boosts to 4.00 GHz. Both have a base clock of 2.80 GHz.

Specification Differences

| Specification | AMD Ryzen 5 40 | NVIDIA DGX Spark (MediaTek AHJ11488B) |

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

| Cores | 4 | 20 |

| Threads | 8 | 20 |

| Boost Clock | 4.30 GHz | 4.00 GHz |

| TDP | 15 W | 140 W |

| Socket | AMD Socket FT6 | None listed |

| Architecture | Zen 2 | None listed |

| Codename | Mendocino | GB10 |

| Process Node | 6 nm | 3 nm |

| Die Size | 100 mm² | 208 mm² |

| L1 Cache | 64 KB (per core) | 128 KB (per core) |

| L2 Cache | 512 KB (per core) | 2 MB (per core) |

| L3 Cache | 4 MB (shared) | 8 MB (shared) |

| Memory Support | LPDDR5 | LPDDR5X |

| Memory Bus | Dual-channel | Quad-channel |

| Memory Bandwidth | 88.0 GB/s | 273.1 GB/s |

| PCIe | Gen 3, 4 Lanes (CPU only) | N/A |

| Integrated Graphics | Radeon 610M | GB20B "Blackwell" |

| Release Date | 2025-09-30 | 2025-10-14 |

| Launch MSRP | None listed | $3999 |

| Part Number | Unknown | DSE1-275-A1 |

| Percentile vs All CPUs | 70 | 50 |

| Avg Benchmark Score | 15882 | 0 |

The base clock is identical at 2.80 GHz for both. Both processors are produced by TSMC, both are mobile market segments, both are active production parts, both lack ECC memory support, and neither has an unlocked multiplier.

The Verdict

The data points to the AMD Ryzen 5 40 as the stronger choice for measured CPU performance. Its 70th percentile ranking, supported by a complete set of benchmark scores and an average benchmark score of 15,882, gives it a verified performance profile. The NVIDIA DGX Spark's 50th percentile ranking and lack of any benchmark scores leave its actual performance unmeasured in the database. Despite its architectural advantages in core count, cache, and memory bandwidth, the DGX Spark has no recorded results to substantiate its capability.

For workloads that depend on single-thread speed and power efficiency, the Ryzen 5 40 is the clear pick. Its higher boost clock of 4.30 GHz and 15 watt TDP make it suitable for mobile computing scenarios where thermal and power constraints are tight. The recorded single-core and multicore scores provide concrete evidence of its capabilities.

For workloads that demand massive parallelism and memory throughput, the DGX Spark's specifications suggest it could excel, but the database contains no measurements to confirm this. The 20-core design, 273.1 GB/s memory bandwidth, and 8 MB L3 cache are compelling on paper. The 3 nm process node and quad-channel memory bus are advanced features. However, without benchmark data, any performance claim remains unverified.

The database records no head-to-head benchmark wins for either product. The only comparative metric available is the percentile ranking, and that favors the Ryzen 5 40 by 20 percentile points. The average benchmark score of 15,882 for the Ryzen stands against a score of 0 for the DGX Spark, reflecting the absence of recorded tests rather than a literal performance of zero.

The DGX Spark's launch MSRP of $3999 places it in a different market tier than the Ryzen 5 40, which has no recorded launch price. The two processors serve fundamentally different design goals. The Ryzen 5 40 is a low-power mobile processor with verified results. The DGX Spark is a high-core-count mobile processor with advanced specifications but no recorded performance data. Based strictly on what the database contains, the Ryzen 5 40 is the only one of the two with demonstrated benchmark performance, and its higher percentile ranking supports that conclusion.