Intel Core i5-110 vs NVIDIA DGX Spark (MediaTek AHJ11488B) Comparison

Where Each One Wins

The recorded data separates these two processors into distinct use cases with almost no overlap. The Intel Core i5-110 is a desktop part built around 6 cores and 12 threads, with a base clock of 2.90 GHz and a boost clock of 4.30 GHz. It uses a 65 W TDP and targets the Intel Socket 1200 platform. The NVIDIA DGX Spark (MediaTek AHJ11488B) is a mobile-class processor with 20 cores and 20 threads, running at 2.80 GHz base and 4.00 GHz boost, with a 140 W TDP. The core count difference alone shifts the workload balance: the i5-110 favors tasks that rely on higher per-core frequency and desktop platform flexibility, while the DGX Spark favors throughput-oriented workloads that scale across many cores.

Benchmark results indicate that neither part has recorded wins in the head-to-head database, as both entries show zero wins and zero average benchmark scores. However, the structural data provides a clear qualitative split. The i5-110's 12 threads and 4.30 GHz boost clock position it for lightly threaded workloads, where clock speed dominates. The DGX Spark's 20 threads and 2 MB L2 cache per core suggest a design meant for parallel execution, particularly in AI or data center contexts, given its "Spark (GB10)" codename and integrated GB20B "Blackwell" graphics. The i5-110 includes UHD Graphics 630, a modest integrated GPU for desktop output, whereas the DGX Spark's integrated graphics carry a Blackwell architecture label, indicating a different compute focus.

Memory bandwidth further separates them. The DGX Spark delivers 273.1 GB/s over a quad-channel LPDDR5X bus, which is roughly six times the i5-110's 42.7 GB/s dual-channel DDR4 bandwidth. That bandwidth advantage directly benefits memory-intensive workloads such as large model inference or data streaming. The i5-110's 12 MB shared L3 cache and 256 KB per-core L2 cache suit conventional desktop applications, while the DGX Spark's 8 MB shared L3 and 2 MB per-core L2 favor workloads that repeatedly access working sets larger than typical desktop tasks.

The production status for both is Active, and both sit at the 50th percentile versus all CPUs in the database, meaning neither is an outlier in overall performance ranking. But the percentile does not capture the workload split. The i5-110 wins on desktop compatibility, socket availability, and lower power draw. The DGX Spark wins on core count, cache per core, memory bandwidth, and process node (3 nm from TSMC versus 14 nm from Intel). Each part holds a clear advantage in its domain, and the data does not suggest any crossover where one would outperform the other outside its intended segment.

The Verdict

The data directs different buyers to different parts. The Intel Core i5-110 suits a desktop builder who needs a Socket 1200 processor with 6 cores, 12 threads, and a 4.30 GHz boost clock. The 65 W TDP allows standard cooling solutions, and the dual-channel DDR4 support at 42.7 GB/s is adequate for mainstream computing. The launch MSRP is $200, which positions it as a conventional mid-range desktop option.

The NVIDIA DGX Spark (MediaTek AHJ11488B) suits a user who prioritizes parallel throughput and memory bandwidth. Its 20 cores, 20 threads, 273.1 GB/s memory bandwidth, and quad-channel LPDDR5X support point to workloads that saturate memory and thread count. The 3 nm TSMC process node and 208 mm² die size indicate a modern, dense design. The launch MSRP is $3999, reflecting a different market tier entirely.

Neither part outperforms the other in the recorded head-to-head benchmarks because no benchmark scores exist in the database. The verdict must rest on the architectural data. A user with a desktop motherboard and standard power envelope should choose the i5-110. A user needing maximum core count and memory throughput in a mobile form factor should choose the DGX Spark. The data does not support a recommendation for one over the other outside these contexts.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark entries for this pairing. Both the winsA and winsB fields are zero, and the headToHeadBenchmarks array is empty. The avgBenchmarkScore for each part is also zero. This means no direct performance comparison exists from actual test runs.

What the data does provide is a comparison of specifications that imply performance differences. The DGX Spark has 20 cores versus 6 cores for the i5-110, a 14-core advantage. The DGX Spark also has 8 more threads (20 versus 12). The i5-110 counters with a 0.30 GHz higher boost clock (4.30 GHz versus 4.00 GHz) and a 0.10 GHz higher base clock (2.90 GHz versus 2.80 GHz).

Cache distribution differs significantly. The DGX Spark allocates 128 KB L1 per core and 2 MB L2 per core, totaling 40 MB of L2 across all cores. The i5-110 allocates 64 KB L1 and 256 KB L2 per core, totaling 1.5 MB of L2 across all cores. The DGX Spark's L2 advantage is substantial, though it also has a smaller shared L3 (8 MB versus 12 MB). Memory bandwidth favors the DGX Spark by a wide margin: 273.1 GB/s versus 42.7 GB/s, a 230.4 GB/s difference.

Process node and die size also differ. The DGX Spark uses a 3 nm process from TSMC with a 208 mm² die. The i5-110 uses Intel's 14 nm process with no listed die size. The DGX Spark's integrated graphics are labeled GB20B "Blackwell", while the i5-110 uses UHD Graphics 630. Neither part has ECC memory support.

The absence of benchmark scores means the largest wins are purely specification-based. The DGX Spark wins on core count, thread count, L2 cache, memory bandwidth, process node, and integrated graphics tier. The i5-110 wins on clock speeds, L3 cache, and TDP efficiency (65 W versus 140 W). Without measured scores, the data cannot confirm which part delivers better real-world performance in any specific application.

FAQ

Q: Which processor has more cores?

A: The NVIDIA DGX Spark (MediaTek AHJ11488B) has 20 cores, while the Intel Core i5-110 has 6 cores. The DGX Spark also has 20 threads versus 12 threads for the i5-110.

Q: What are the boost clock speeds of each part?

A: The Intel Core i5-110 boosts to 4.30 GHz, which is 0.30 GHz higher than the NVIDIA DGX Spark's 4.00 GHz boost clock.

Q: How much memory bandwidth does each processor support?

A: The NVIDIA DGX Spark supports 273.1 GB/s over a quad-channel LPDDR5X bus. The Intel Core i5-110 supports 42.7 GB/s over a dual-channel DDR4 bus.

Q: What process nodes do the two processors use?

A: The NVIDIA DGX Spark uses a 3 nm process from TSMC with a 208 mm² die size. The Intel Core i5-110 uses Intel's 14 nm process with no die size listed.

Q: Do either of these processors support ECC memory?

A: No. Both the Intel Core i5-110 and the NVIDIA DGX Spark have eccMemory set to false in the database.

Q: What is the TDP for each processor?

A: The Intel Core i5-110 has a 65 W TDP. The NVIDIA DGX Spark has a 140 W TDP, which is 75 W higher.

Architecture Differences

The two processors diverge at every architectural level. The Intel Core i5-110 belongs to the Comet Lake architecture, built on Intel's 14 nm process. It uses the Intel Socket 1200 platform and targets the desktop market segment. The NVIDIA DGX Spark (MediaTek AHJ11488B) carries the GB10 codename, uses TSMC's 3 nm process, and targets the mobile market segment. The process node difference is substantial: 14 nm versus 3 nm, a gap of 11 nm in feature size.

Core and thread counts set the fundamental compute shape. The i5-110 has 6 cores and 12 threads, meaning each core supports two threads. The DGX Spark has 20 cores and 20 threads, meaning each core supports one thread. This indicates different execution models: simultaneous multithreading on Intel versus a wider single-threaded core array on NVIDIA.

Cache hierarchies are organized differently. The i5-110 uses 64 KB L1 per core, 256 KB L2 per core, and 12 MB shared L3. The DGX Spark uses 128 KB L1 per core, 2 MB L2 per core, and 8 MB shared L3. The per-core L2 allocation is 8 times larger on the DGX Spark, which suggests a design that keeps more working data closer to each core. The shared L3 is 4 MB larger on the i5-110, offering more cross-core cache capacity but with lower per-core bandwidth.

Memory subsystems differ in width and technology. The i5-110 uses dual-channel DDR4 with 42.7 GB/s bandwidth. The DGX Spark uses quad-channel LPDDR5X with 273.1 GB/s bandwidth. The DGX Spark's memory bus is double the width and the technology is newer, resulting in a 230.4 GB/s bandwidth advantage. PCIe support also differs: the i5-110 provides Gen 3 with 16 lanes from the CPU, while the DGX Spark lists PCIe as N/A.

Integrated graphics represent another split. The i5-110 includes UHD Graphics 630, a conventional desktop iGPU. The DGX Spark includes GB20B "Blackwell" graphics, which ties into NVIDIA's GPU architecture lineage. The market segment classification reinforces this: Desktop for Intel, Mobile for NVIDIA.

Production status is Active for both, and release dates are close, with the i5-110 at 2025-09-10 and the DGX Spark at 2025-10-14. Neither has an unlocked multiplier. The DGX Spark has a listed die size of 208 mm², while the i5-110 has no die size recorded. Transistor counts are not recorded for either part.

The architecture differences align with each part's intended role. The i5-110 is a conventional desktop CPU with a familiar socket, moderate power draw, and standard memory support. The DGX Spark is a dense, high-bandwidth mobile processor with a modern process node, large per-core cache, and Blackwell graphics. The data shows no architectural overlap between the two.