CPU Comparison

The AMD Ryzen Threadripper 1900X and Intel Xeon E5-2698B v3 represent two very different approaches to high-core-count computing, with the former built on a modern 14nm Zen design and the latter a 22nm Haswell-EP server part. The benchmark data shows a remarkably consistent pattern: the AMD processor wins every single head-to-head test, but by margins that are uniformly narrow, ranging from 3.2% to 3.6%. This makes the comparison less about outright dominance and more about architectural efficiency versus raw core count, with the Threadripper’s higher clock speeds and newer design compensating for the Xeon’s 2x core and 2x thread advantage.

Head-to-Head Benchmarks

The most striking finding is the consistency of the AMD Ryzen Threadripper 1900X’s victory across all six Cinebench tests. In Cinebench R15 multi-core, the Threadripper scores 1440 against the Xeon’s 1395, a 3.2% advantage. The single-core R15 test shows a slightly larger gap, with the Threadripper at 203 versus the Xeon’s 196, a 3.6% lead. This pattern repeats exactly in Cinebench R20: multi-core sees 6000 versus 5814 (3.2% delta), while single-core shows 846 versus 820 (also 3.2%). The Cinebench R23 results mirror this precisely, with the Threadripper hitting 14286 multi-core versus the Xeon’s 13843 (3.2%), and 2016 single-core versus 1954 (3.2%).

What makes these numbers notable is that the Xeon E5-2698B v3 has 16 cores and 32 threads, exactly double the Threadripper’s 8 cores and 16 threads. Yet the AMD part still manages to edge ahead in multi-threaded workloads. This suggests that the Threadripper’s 3.80 GHz base clock and 4.00 GHz boost clock, versus the Xeon’s 2.00 GHz base and 3.40 GHz boost, are more than enough to overcome the core deficit. The single-core results reinforce this: the 3.6% lead in R15 single-core and 3.2% in R20/R23 single-core indicate that the Zen architecture’s per-thread performance is decisively better than Haswell-EP’s.

The Geekbench results, available only for the Threadripper, show a multi-core score of 6588 and single-core of 1205, but no direct comparison exists for the Xeon in this test. The average benchmark score tells a similar story: the Threadripper averages 4073 across all its tested workloads, while the Xeon averages 4004. Both CPUs sit at the 57th percentile among all processors, placing them in the same performance tier despite their architectural differences. The nearest rivals for the Threadripper include the Intel Core i5-12500TE (avg 4083, -0.3% delta) and AMD Ryzen 7 PRO 2700X (avg 4114, -1% delta), while the Xeon’s closest competitors are the AMD Ryzen 5 PRO 4655G (avg 4010, -0.2%) and Intel Core i5-12400T (avg 4019, -0.4%). These figures confirm that both chips occupy a similar performance band, with the Threadripper holding a slight edge.

Where Each One Wins

The data shows a clean sweep for the AMD Ryzen Threadripper 1900X across all six head-to-head benchmark tests, so the Xeon E5-2698B v3 does not win any direct comparisons. However, the Xeon’s strengths manifest in different ways. With 16 cores and 32 threads versus the Threadripper’s 8 and 16, the Xeon offers double the parallel processing capacity. In heavily threaded workloads where clock speed is less critical, such as server virtualization or database processing, the Xeon’s extra cores could theoretically provide better scaling, even though the Cinebench results do not reflect this advantage.

The Threadripper, by contrast, wins on efficiency and single-thread performance. Its 14nm process node versus the Xeon’s 22nm yields better performance per clock, and its higher base and boost clocks make it superior for lightly threaded applications. The Threadripper also benefits from a 32 MB L3 cache versus the Xeon’s 40 MB, but the AMD part’s per-core L1 (96 KB) and L2 (512 KB) caches are larger than the Xeon’s 64 KB and 256 KB respectively. For users running mixed workloads, some single-threaded, some multi-threaded, the Threadripper’s consistency across all tests makes it the more versatile choice.

The Xeon does have one clear advantage: ECC memory support. The Threadripper does not support ECC, while the Xeon does. This makes the Xeon the technically superior choice for mission-critical server environments where data integrity is paramount. Additionally, the Xeon’s lower TDP of 135 watts versus the Threadripper’s 180 watts means it draws less power under load, a meaningful consideration for densely packed server racks. The Xeon is also end-of-life, while the Threadripper remains in active production, but that status does not affect current benchmark performance.

Architecture Differences

The architectural gap between these two processors is substantial. The AMD Ryzen Threadripper 1900X uses the Zen architecture, built on a 14nm process at GlobalFoundries, with a die size of 2x 213 mm² and 9,600 million transistors. The Intel Xeon E5-2698B v3 uses the Haswell-EP architecture, built on Intel’s 22nm process, with a single 356 mm² die and 2,600 million transistors. This difference in node size explains much of the performance gap: the smaller 14nm process allows for higher clock speeds and better power efficiency, while the older 22nm process limits the Xeon to a 2.00 GHz base clock.

The cache hierarchies also differ significantly. The Threadripper has 96 KB of L1 cache per core, 512 KB of L2 per core, and 32 MB of shared L3 cache. The Xeon has 64 KB of L1 per core, 256 KB of L2 per core, and 40 MB of shared L3 cache. While the Xeon has more total L3, the Threadripper’s larger per-core caches likely contribute to its single-thread performance advantage. Memory support is similar, both use DDR4 with quad-channel memory buses, but the Threadripper offers 85.3 GB/s of memory bandwidth versus the Xeon’s 68.3 GB/s, a 25% difference that favors the AMD part.

PCIe connectivity is another differentiator. The Threadripper provides 60 PCIe Gen 3 lanes (CPU only), while the Xeon provides 40 lanes. This makes the Threadripper better suited for multi-GPU configurations or high-bandwidth storage arrays. The Threadripper also has an unlocked multiplier, allowing for overclocking, whereas the Xeon is locked. The Threadripper’s socket is AMD Socket SP3r2, while the Xeon uses Intel Socket 2011-3. The Threadripper is a desktop part with a launch MSRP of $549, while the Xeon is a server/workstation part with no listed launch MSRP. The Xeon does support ECC memory, which the Threadripper does not.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon E5-2698B v3 has 16 cores and 32 threads, exactly double the AMD Ryzen Threadripper 1900X’s 8 cores and 16 threads.

Q: Why does the Threadripper win multi-core benchmarks despite having half the cores?

A: The Threadripper’s 3.80 GHz base and 4.00 GHz boost clocks, combined with the newer 14nm Zen architecture, compensate for the core deficit. In Cinebench R23 multi-core, it scores 14286 versus the Xeon’s 13843, a 3.2% margin.

Q: Does the Xeon support ECC memory?

A: Yes, the Intel Xeon E5-2698B v3 supports ECC memory. The AMD Ryzen Threadripper 1900X does not support ECC memory.

Q: What are the power consumption figures?

A: The Threadripper has a TDP of 180 watts, while the Xeon has a TDP of 135 watts.

Q: Which processor has higher memory bandwidth?

A: The AMD Ryzen Threadripper 1900X offers 85.3 GB/s of memory bandwidth, compared to the Intel Xeon E5-2698B v3’s 68.3 GB/s.

Q: Are both processors in the same performance percentile?

A: Yes, both the Threadripper and the Xeon sit at the 57th percentile among all CPUs. The Threadripper’s average benchmark score is 4073, slightly above the Xeon’s 4004.

The Verdict

The benchmark data is unambiguous: the AMD Ryzen Threadripper 1900X outperforms the Intel Xeon E5-2698B v3 in every tested workload, winning all six head-to-head comparisons with margins between 3.2% and 3.6%. For users prioritizing raw performance in Cinebench-style rendering workloads, the Threadripper is the clear choice. Its higher clock speeds and modern 14nm architecture deliver better single-thread and multi-thread results, despite having half the cores and threads of the Xeon.

However, the Xeon’s strengths lie outside synthetic benchmarks. Its 16 cores and 32 threads provide double the parallel capacity, which could benefit workloads that scale perfectly with core count, even if those gains do not appear in the provided Cinebench results. The Xeon’s ECC memory support and lower 135-watt TDP make it more suitable for server environments where data integrity and power efficiency are critical. The Xeon is also end-of-life, while the Threadripper remains active in production.

The Threadripper offers more PCIe lanes (60 versus 40), higher memory bandwidth (85.3 GB/s versus 68.3 GB/s), and an unlocked multiplier for overclocking. It also has a launch MSRP of $549. For desktop users, workstation builders, or anyone needing a balance of single-thread responsiveness and multi-thread capability, the Threadripper is the superior choice based on the data. For server administrators prioritizing ECC support and lower power draw, the Xeon remains a viable option, but the benchmark results favor AMD. Choose the Threadripper for performance, the Xeon for ECC and power efficiency.