AMD EPYC 72F3 vs Intel Core i3-1115G4 Comparison

The Intel Core i3-1115G4 and the AMD EPYC 72F3 occupy opposite ends of the processor spectrum, yet the recorded benchmark data places them in a head-to-head comparison that reveals a complete dominance by one side. The Core i3 is a mobile, low-power part built for thin-and-light laptops, while the EPYC 72F3 is a server-class behemoth designed for datacenter workloads. The database shows the EPYC winning every single benchmark in this pairing, with a consistent performance gap that underscores the fundamental differences in their design goals. The average benchmark scores are surprisingly close (6851 for the Intel versus 6700 for the AMD), but that aggregate figure masks the fact that the two chips are evaluated on entirely different workloads, and the direct comparisons tell a much more lopsided story.

Head-to-Head Benchmarks

The six head-to-head benchmarks in the database all use the Cinebench suite, which measures rendering performance across single-core and multi-core tests. In every one of these tests, the AMD EPYC 72F3 emerges victorious, and the margin is not just a minor edge but a massive chasm. The delta percentage is identical across all six tests: the EPYC scores 78.7% higher than the Core i3 in each case. This uniformity is striking; it suggests that the performance gap is not workload-specific but rather a fundamental throughput advantage that scales across both single-threaded and multi-threaded execution.

Looking at the multi-core results, the EPYC 72F3 delivers 2334 points in Cinebench R15 multi-core, while the Core i3 manages only 497. That is a 78.7% deficit for the Intel part, meaning the EPYC completes the same rendering task in roughly one-fifth of the time. The gap widens further in Cinebench R20 multi-core, where the EPYC scores 9728 versus the Core i3's 2071. In R23 multi-core, the EPYC hits 23164, while the Core i3 finishes at 4933. These numbers are not incremental improvements; they represent a different class of capability. The EPYC's eight cores and sixteen threads simply overwhelm the Core i3's two cores and four threads in any workload that can use parallel processing.

The single-core results are where the comparison becomes more nuanced, because single-core performance often reflects architectural efficiency rather than raw core count. Here, the EPYC still wins decisively. In Cinebench R15 single-core, the EPYC scores 329, while the Core i3 scores just 70. In R20 single-core, the EPYC scores 1373 versus 292. In R23 single-core, the EPYC scores 3270 versus 696. The 78.7% delta holds even in these tests, which is remarkable. It indicates that the Zen 3 architecture in the EPYC is not only more parallel but also more efficient per clock cycle, at least as measured by Cinebench. The Core i3's boost clock is 4.10 GHz, identical to the EPYC's boost clock of 4.10 GHz, yet the EPYC still produces roughly four times the single-core score. This points to a substantial instructions-per-clock advantage for the Zen 3 core, which is a critical finding for anyone evaluating these two processors.

The database also records that the Core i3 has zero wins in this head-to-head, while the EPYC has six. That is a clean sweep, and the data leaves no room for interpretation. The Core i3 is not competitive in any Cinebench metric, whether it is a short rendering task or a longer one, and whether it uses one thread or all available threads. The only context where the Core i3 might appear closer is in the average benchmark score, which includes PassMark results that are not part of the head-to-head set. Those PassMark tests, however, are not directly compared here, so the Cinebench data stands as the definitive measure for this pairing.

The Verdict

Based strictly on the recorded data, the AMD EPYC 72F3 is the superior processor for any workload that the Cinebench suite represents. The 78.7% lead in all six benchmarks is not a marginal advantage; it is a generational and architectural gap. The EPYC's eight cores, sixteen threads, and 256 MB of shared L3 cache provide a level of compute throughput that the Core i3 cannot approach. The Core i3, with its two cores, four threads, and 6 MB of L3 cache, is built for efficiency and mobility, not for rendering or heavy compute tasks.

Who should pick which? The data suggests that the EPYC 72F3 is the choice for server, workstation, or any environment where rendering, simulation, or multi-threaded processing is the primary task. The EPYC's 180 W TDP and server socket (AMD Socket SP3) indicate it is designed for always-on, high-load scenarios. The Core i3, with its 15 W TDP and mobile socket (Intel BGA 1449), is suited for portable devices where battery life and thermal management are paramount. The Core i3's integrated Iris Xe-LP Graphics G4 also makes it a self-contained solution for a laptop, whereas the EPYC has no integrated graphics and requires a discrete GPU.

The percentile rankings offer a broader context. The Core i3 sits at the 63rd percentile among all CPUs in the database, while the EPYC sits at the 62nd percentile. These are nearly identical, which is surprising given the lopsided head-to-head results. This suggests that the database's overall benchmark pool includes many other processors that outperform both of these, and that the average scores (6851 for Intel, 6700 for AMD) are pulled in different directions by different workload types. The Core i3's PassMark scores, which include data compression at 62444 and integer math at 17472, may boost its average relative to the EPYC, which only has Cinebench scores in this record. The data does not include any PassMark results for the EPYC, so a direct comparison on those tests is not possible.

Where Each One Wins

The EPYC 72F3 wins in all six head-to-head Cinebench tests, so the "where each one wins" section is, in this pairing, entirely one-sided. The EPYC wins in multi-core rendering, where its eight cores and sixteen threads allow it to process multiple render tasks simultaneously. It also wins in single-core rendering, which is more telling because it isolates the architectural efficiency of the Zen 3 core. The EPYC's 256 MB of shared L3 cache likely contributes to this single-core advantage by reducing memory latency for frequently accessed data.

The Core i3 does not win a single benchmark in this direct comparison. However, the broader PassMark results in its record indicate areas where it is not necessarily a poor performer in absolute terms, even if it cannot match the EPYC. For example, the Core i3 scores 2585 in PassMark single-thread and 5915 in PassMark multithread. These are respectable numbers for a mobile chip, but they are not part of the head-to-head comparison, so they cannot be used to claim any win over the EPYC. The Core i3 also has a 63rd percentile ranking, which places it above the EPYC's 62nd percentile, but that ranking is based on the full benchmark set, not the direct comparison.

In practical terms, the Core i3's wins would come in scenarios not measured here, such as power efficiency or thermal output, but the database does not provide any power or thermal benchmarks. The TDP figures (15 W for Intel, 180 W for AMD) are the only power-related data, and they are not benchmark scores. The Core i3 also supports LPDDR4X memory and has integrated graphics, which are features the EPYC lacks, but those are not performance metrics.

FAQ

Q: Why does the AMD EPYC 72F3 win every head-to-head benchmark by the same 78.7% margin?

A: The database records identical delta percentages of -78.7% for all six Cinebench tests, which means the EPYC's score is consistently 78.7% higher than the Core i3's score in each test. This uniformity suggests that the performance gap is proportional across single-core and multi-core workloads, likely due to a combination of higher core count and more efficient per-core execution.

Q: Is the Intel Core i3-1115G4 competitive in any benchmark?

A: In the direct head-to-head comparison, the Core i3 has zero wins out of six benchmarks. The database does include PassMark scores for the Core i3 (such as 2585 in single-thread and 5915 in multithread), but these are not compared directly to the EPYC, which has no PassMark scores recorded. Therefore, the data shows no benchmark where the Core i3 beats the EPYC.

Q: How do the core counts and cache sizes affect the results?

A: The EPYC has 8 cores and 16 threads, while the Core i3 has 2 cores and 4 threads. The EPYC also has 256 MB of shared L3 cache, compared to the Core i3's 6 MB. These differences directly explain the multi-core performance gap, as the EPYC can handle more threads simultaneously. The single-core gap is less obvious but may be influenced by the larger cache reducing memory latency.

Q: What is the significance of the percentile rankings for these two CPUs?

A: The Core i3 is at the 63rd percentile and the EPYC is at the 62nd percentile among all CPUs in the database. This means both are near the middle of the overall distribution, but the percentile is based on average benchmark scores, which include different test sets for each CPU. The Core i3 has PassMark scores that likely raise its average, while the EPYC only has Cinebench scores.

Q: Can the Core i3's integrated graphics compensate for its lower compute performance?

A: The Core i3 includes Iris Xe-LP Graphics G4, while the EPYC has no integrated graphics. This is a feature difference, not a benchmark score. The database does not provide any graphics benchmarks, so the data cannot quantify any advantage from the integrated GPU. For compute-heavy tasks, the EPYC's performance lead is clear.

Q: Are there any workload types where the Core i3 might be preferable based on the data?

A: The data shows the Core i3 has a lower TDP (15 W versus 180 W) and a mobile socket, which implies it is designed for portable devices. The database does not include power efficiency benchmarks, so the only advantage is the TDP specification itself. For rendering tasks, the EPYC is the clear winner in every recorded test.

Architecture Differences

The two processors come from different manufacturers and use fundamentally different architectures. The Intel Core i3-1115G4 is built on the Tiger Lake architecture, specifically the Tiger Lake-U codename, and uses a 10 nm process node manufactured by Intel. The AMD EPYC 72F3 is based on the Zen 3 architecture, with the codename Milan, and uses a 7 nm process node from TSMC. This process advantage gives the EPYC a transistor density edge, with the EPYC containing 33,200 million transistors across eight chiplets, each with an 81 mm² die size. The Core i3 has a single die of 144 mm², with no transistor count recorded.

The cache hierarchies are also markedly different. The Core i3 has 80 KB of L1 cache per core, 1.25 MB of L2 per core, and 6 MB of shared L3. The EPYC has 64 KB of L1 per core, 512 KB of L2 per core, and a massive 256 MB of shared L3. That L3 cache is a defining feature of the EPYC, designed to hold entire datasets in fast memory and reduce the need to access slower system memory. The Core i3's L3 is far smaller, reflecting its mobile focus where power and die area are at a premium.

The memory controllers also differ. The Core i3 supports DDR4 and LPDDR4X memory in a dual-channel configuration, while the EPYC supports DDR4 in an eight-channel configuration with a recorded memory bandwidth of 204.8 GB/s. The EPYC also supports ECC memory, which is critical for server reliability, whereas the Core i3 does not. The PCIe capabilities are similarly lopsided: the Core i3 provides Gen 4 with 4 CPU lanes, while the EPYC provides Gen 4 with 128 CPU lanes, enabling far more expansion devices and high-speed interconnects.

The manufacturing process and foundry choices reflect the different market segments. Intel fabricates the Core i3 on its own 10 nm process, while AMD uses TSMC's 7 nm process for the EPYC. The EPYC's chiplet design, with eight dies, allows for higher yields and modular scaling, whereas the Core i3 is a monolithic die. These architectural differences explain why the EPYC can sustain a 180 W TDP and deliver server-class performance, while the Core i3 operates within a 15 W envelope for battery-powered devices.

Specification Differences

The specification table shows clear divergences in almost every field. The Core i3 has 2 cores and 4 threads, while the EPYC has 8 cores and 16 threads, a fourfold increase in both. The base clocks differ: the Core i3 runs at 2.20 GHz, while the EPYC runs at 3.70 GHz. The boost clocks are identical at 4.10 GHz for both, which is notable because it means the single-core performance gap is not due to clock speed but to architectural efficiency. The TDP is a major differentiator: 15 W for the Intel part versus 180 W for the AMD part, reflecting the mobile versus server design goals.

The sockets are incompatible: the Core i3 uses Intel BGA 1449, a soldered mobile socket, while the EPYC uses AMD Socket SP3, a server socket. The process nodes differ (10 nm for Intel, 7 nm for AMD), as do the foundries (Intel versus TSMC). The EPYC has a recorded transistor count of 33,200 million, while the Core i3 has no transistor count listed. The die sizes are 144 mm² for the Core i3 and 8x 81 mm² for the EPYC, meaning the EPYC uses eight smaller dies rather than one larger one.

Memory support is another clear split: the Core i3 supports DDR4 and LPDDR4X with a dual-channel bus, while the EPYC supports only DDR4 but with an eight-channel bus and a recorded bandwidth of 204.8 GB/s. ECC memory is available only on the EPYC. PCIe lanes are vastly different: the Core i3 has 4 CPU lanes of Gen 4, while the EPYC has 128 CPU lanes of Gen 4. The Core i3 includes integrated Iris Xe-LP Graphics G4, while the EPYC has no integrated graphics. The market segments are Mobile for Intel and Server/Workstation for AMD, and the release dates are September 2020 for the Core i3 and March 2021 for the EPYC. The launch MSRP for the Core i3 is $281, and for the EPYC it is $2468, though price is not a performance metric. The production status for both is Active, and neither has an unlocked multiplier.