CPU Comparison

The Intel Xeon E-2388G and AMD Ryzen Threadripper 2920X present a fascinating split, with each dominating entirely different benchmark families. The data shows a clear divergence: AMD’s 12-core processor wins every Cinebench test, while Intel’s 8-core chip delivers a decisive victory in both Geekbench tests. This is not a close contest across the board but rather a story of two distinct performance profiles that cater to different workloads. The overall average benchmark scores are close, with the Xeon at 5805 and the Threadripper at 5730, placing them within 1.3% of each other, yet the component scores could not be more different.

Head-to-Head Benchmarks

The Cinebench suite is entirely AMD’s territory. In Cinebench R15 multicore, the Threadripper scores 2154 against the Xeon’s 2009, a 6.7% advantage. The single-core R15 test shows a similar gap, with AMD at 304 versus Intel’s 283, a 6.9% lead. This pattern persists across the newer R20 and R23 versions. The R20 multicore result has AMD at 8979 and Intel at 8374, again a 6.7% delta. In R20 single-core, AMD wins 1267 to 1181, a 6.8% edge. The R23 multicore scores are 21380 for the Threadripper and 19939 for the Xeon, maintaining that consistent 6.7% gap. The R23 single-core test also shows AMD ahead, 3018 to 2814, a 6.8% difference. Across six Cinebench tests, the Threadripper’s victory margin is remarkably uniform, hovering between 6.7% and 6.9%. This consistency suggests a fundamental architectural advantage in rendering workloads, not a test-specific quirk.

The Geekbench results invert the narrative completely. In Geekbench multicore, the Xeon E-2388G scores 9615, while the Threadripper manages only 7473. That translates to a massive 28.7% lead for Intel. The single-core Geekbench result is even more lopsided: Intel scores 2227 versus AMD’s 1266, a staggering 75.9% advantage. This is not a minor edge; it is a near-doubling of the single-threaded score. The data implies that Intel’s Rocket Lake architecture, with its high boost clock of 5.10 GHz, is exceptionally strong in the specific tasks Geekbench measures, while AMD’s Zen+ design, with its lower 4.30 GHz boost, falls far behind in that particular benchmark’s single-thread workload. It is notable that the Threadripper wins all Cinebench tests by roughly the same percentage, while the Xeon wins Geekbench by a wide and inconsistent margin, indicating that the two benchmarks stress very different aspects of the processors.

Architecture Differences

The underlying designs explain these benchmark splits. The Intel Xeon E-2388G is built on the Rocket Lake architecture, specifically Rocket Lake-E, using a 14 nm process from Intel’s own fabs. It has 8 cores and 16 threads, with a base clock of 3.20 GHz and a boost clock of 5.10 GHz. Its cache layout is 80 KB of L1 per core, 512 KB of L2 per core, and 16 MB of shared L3 cache. The die size is 276 mm². This is a server/workstation part with a 95 W TDP, supporting DDR4 memory over a dual-channel bus with 51.2 GB/s of bandwidth. It also includes integrated UHD Graphics P750 and supports ECC memory, which is typical for its market segment.

The AMD Ryzen Threadripper 2920X is a different beast entirely. It uses the Zen+ architecture, codenamed Colfax, fabricated on a 12 nm process by GlobalFoundries. It packs 12 cores and 24 threads, with a base clock of 3.50 GHz and a boost clock of 4.30 GHz. The cache hierarchy is larger in some respects: 96 KB of L1 per core, 512 KB of L2 per core, but a significantly larger 32 MB of L3 cache. The die is composed of two 213 mm² chiplets, totaling 9,600 million transistors. It has a 180 W TDP, uses a quad-channel DDR4 memory bus with 93.9 GB/s of bandwidth, and lacks integrated graphics. It also does not support ECC memory. The Threadripper is an unlocked multiplier part, while the Xeon is locked.

These differences are stark. The Threadripper has 50% more cores and threads, a larger L3 cache, and nearly double the memory bandwidth. The Xeon counters with a much higher boost clock, a newer process node (14 nm vs 12 nm, though Intel’s 14 nm is an older design), and integrated graphics. The process node comparison is not straightforward, but the data shows Intel’s smaller core count with higher clocks achieves a different performance balance. The Xeon also has PCIe Gen 4 with 20 lanes, while the Threadripper uses PCIe Gen 3 with 60 lanes. The Xeon’s higher single-thread clock speed is likely a major factor in its Geekbench dominance, while the Threadripper’s core count and cache size drive its Cinebench wins.

Where Each One Wins

The benchmark data points to clear use-case scenarios. The AMD Ryzen Threadripper 2920X is the winner in all Cinebench iterations, which are classic multi-threaded rendering workloads. Its advantage of 6.7% to 6.9% across R15, R20, and R23, in both single and multicore tests, indicates that its 12-core, 24-thread configuration is better suited for video rendering, 3D animation, and other tasks that scale with core count. The larger 32 MB L3 cache and higher memory bandwidth of 93.9 GB/s also support this, as those tasks often benefit from larger data pools being accessible. If a workload is primarily Cinebench-like, the Threadripper is the clear choice from this data.

The Intel Xeon E-2388G wins decisively in Geekbench, particularly in the single-core test where it leads by 75.9%. This suggests the Xeon is superior for tasks that rely heavily on single-threaded performance, such as many legacy applications, certain database operations, or software that is not well-optimized for multi-threading. The Geekbench multicore win of 28.7% for Intel is also substantial, showing that even with fewer cores, the Xeon’s architecture can outperform in this specific benchmark. The Xeon’s 5.10 GHz boost clock is the standout feature driving this. It also has the advantage of integrated graphics, which could be useful for basic display output or light GPU-accelerated tasks without a dedicated card. The ECC memory support makes it a candidate for reliability-sensitive server environments, despite its lower core count.

The data does not show a single overall winner. It shows a partition. The Threadripper is for compute-heavy, multi-threaded rendering. The Xeon is for high-frequency, single-thread-sensitive tasks. The average benchmark scores being nearly identical (5805 vs 5730) is misleading, as it masks the extreme variance in individual test categories. A user should not look at the average but at the specific benchmark that matches their intended application.

The Verdict

Based strictly on the data, the AMD Ryzen Threadripper 2920X is the better choice for anyone running Cinebench-like workloads. It wins every single Cinebench test, from R15 to R23, by a consistent margin of roughly 6.7%. If the primary use is 3D rendering, video encoding, or any heavily multithreaded task that behaves like Cinebench, this processor is superior. The fact that it also has an unlocked multiplier offers flexibility for those who may want to push performance further, though the data does not include overclocking results.

The Intel Xeon E-2388G is the better choice for workloads represented by Geekbench. Its 75.9% single-core lead and 28.7% multicore lead are overwhelming. This makes it ideal for applications that are latency-sensitive, depend on high clock speeds, or are poorly multithreaded. The integrated graphics and ECC support are additional features that could be decisive in a server or workstation environment where a discrete GPU is not needed or where data integrity is paramount. The Xeon’s lower TDP of 95 W versus the Threadripper’s 180 W also suggests it will be more efficient in power-constrained scenarios, though the data does not provide direct power consumption measurements under load.

The choice hinges entirely on the benchmark suite that best represents the target software. A rendering farm should pick the Threadripper. A high-frequency trading platform or a database server might prefer the Xeon. The data does not support a universal recommendation. It supports a workload-specific one.

FAQ

Q: Which processor has a higher single-core score in Geekbench?

A: The Intel Xeon E-2388G, with a score of 2227, which is 75.9% higher than the AMD Ryzen Threadripper 2920X’s score of 1266.

Q: Does the AMD Ryzen Threadripper 2920X win all Cinebench tests?

A: Yes, it wins all six Cinebench tests listed (R15, R20, and R23, both single and multicore), with victory margins between 6.7% and 6.9%.

Q: What is the difference in average benchmark scores between the two CPUs?

A: The Intel Xeon E-2388G has an average benchmark score of 5805, while the AMD Ryzen Threadripper 2920X has an average of 5730, a difference of 1.3% in favor of Intel.

Q: How many cores and threads does each processor have?

A: The Intel Xeon E-2388G has 8 cores and 16 threads. The AMD Ryzen Threadripper 2920X has 12 cores and 24 threads.

Q: Which processor supports ECC memory?

A: The Intel Xeon E-2388G supports ECC memory, while the AMD Ryzen Threadripper 2920X does not.

Q: What is the boost clock speed for each CPU?

A: The Intel Xeon E-2388G has a boost clock of 5.10 GHz, and the AMD Ryzen Threadripper 2920X has a boost clock of 4.30 GHz.

Specification Differences

The following table details the key specifications where the two processors differ, based on the provided data.

| Specification | Intel Xeon E-2388G | AMD Ryzen Threadripper 2920X |

| :--- | :--- | :--- |

| Cores | 8 | 12 |

| Threads | 16 | 24 |

| Base Clock | 3.20 GHz | 3.50 GHz |

| Boost Clock | 5.10 GHz | 4.30 GHz |

| TDP | 95 W | 180 W |

| Socket | Intel Socket 1200 | AMD Socket SP3r2 |

| Architecture | Rocket Lake | Zen+ |

| Codename | Rocket Lake-E | Colfax |

| Process Node | 14 nm | 12 nm |

| Foundry | Intel | GlobalFoundries |

| Die Size | 276 mm² | 2x 213 mm² |

| Transistors | N/A | 9,600 million |

| L1 Cache | 80 KB (per core) | 96 KB (per core) |

| L3 Cache | 16 MB (shared) | 32 MB |

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

| Memory Bandwidth | 51.2 GB/s | 93.9 GB/s |

| ECC Memory | Yes | No |

| PCIe | Gen 4, 20 Lanes (CPU only) | Gen 3, 60 Lanes (CPU only) |

| Integrated Graphics | UHD Graphics P750 | None |

| Market Segment | Server/Workstation | Desktop |

| Release Date | 2021-09-07 | 2018-10-02 |

| Launch MSRP | $539 | $649 |

| Multiplier Unlocked | No | Yes |

| Part Number | SRKMZ | YD292XA8UC9AF |