CPU Comparison

The Intel Core i9-10900X and AMD Ryzen Threadripper 1920 are two legacy high-end desktop parts that target different philosophies: Intel’s 10-core Cascade Lake-X versus AMD’s 12-core first-generation Zen. Benchmark data shows the i9-10900X sweeping all six head-to-head tests, but by margins so narrow, exactly 1.1% in every single test, that the results are effectively a statistical tie. The Threadripper 1920 counters with more cores, more threads, and a larger L3 cache, while the i9-10900X relies on higher clocks and newer architecture. The data suggests the Intel part edges out the AMD chip for raw single-threaded and lightly-threaded work, while the Threadripper holds its own in multi-threaded scenarios despite the loss. The verdict is clear: if you need the absolute best per-core performance from these two, pick the i9-10900X; if you want more parallel throughput headroom and a platform with more PCIe lanes, the Threadripper 1920 is the reasonable alternative, though it loses every measured benchmark.

The Verdict

The data points to a decisive, if narrow, victory for the Intel Core i9-10900X. Across six benchmark tests, spanning Cinebench R15, R20, and R23 in both multi-core and single-core modes, the i9-10900X wins every single one, with a consistent deltaPct of 1.1% over the Threadripper 1920. This consistency is remarkable: from Cinebench R15 multicore (1910 vs 1890) to Cinebench R23 singlecore (2676 vs 2647), the margin never wavers. The Intel part’s average benchmark score of 5555 places it in the 61st percentile of all CPUs, while the Threadripper’s 5425 average sits at the 60th percentile, essentially the same tier.

Who should pick which? The i9-10900X is the choice for anyone prioritizing single-core responsiveness and the latest Cinebench results, even if the lead is marginal. The Threadripper 1920, despite losing every head-to-head, offers 12 cores and 24 threads versus the Intel’s 10 cores and 20 threads, but the benchmark results show that extra hardware does not translate into a win. The Threadripper’s production status is end-of-life, whereas the i9-10900X is still active. For builders who already have an Intel Socket 2066 platform or prefer Intel’s ecosystem, the i9-10900X is the data-backed pick. For those who value the Threadripper’s larger L3 cache and 60 PCIe Gen 3 lanes, the 1920 remains a viable, though losing, option.

Architecture Differences

The two processors come from different architectural generations and foundries. The Intel Core i9-10900X is built on Cascade Lake-X, using a 14 nm process at Intel’s own foundry. The AMD Ryzen Threadripper 1920 uses the original Zen architecture, codenamed “Zen” (Whitehaven), also on a 14 nm node but fabricated by GlobalFoundries. This is a significant divergence: both use 14 nm, but the design philosophies differ greatly.

The Threadripper 1920 is a dual-die design, with a die size of 2x 213 mm² and a transistor count of 9,600 million. It packs 12 cores and 24 threads, with a larger per-core L1 cache of 96 KB (versus Intel’s 64 KB) but a smaller per-core L2 cache of 512 KB (versus Intel’s 1 MB). Its L3 cache is a full 32 MB, shared, compared to the i9-10900X’s 19.25 MB shared L3. The Intel part compensates with higher clocks: a base of 3.70 GHz and boost of 4.70 GHz, against the AMD’s 3.20 GHz base and 3.80 GHz boost.

Memory support is identical in type (DDR4) and bus (quad-channel), but the Threadripper lists a specific memory bandwidth of 85.3 GB/s, a figure the Intel pack does not provide. PCIe is Gen 3 for both, but the Threadripper specifies 60 lanes (CPU only), while the Intel part does not list a lane count. Neither has integrated graphics, and both have unlocked multipliers. The Threadripper’s part number is YD1920A9UC9AE; the Intel part has no listed part number. The i9-10900X was released on 2019-10-18, while the Threadripper’s release date is null in the data.

Head-to-Head Benchmarks

The head-to-head results are uniform to a fault: the Intel i9-10900X wins all six tests, each by a deltaPct of 1.1%. Starting with Cinebench R15 multicore, the Intel scores 1910 against the AMD’s 1890, a 20-point gap. In Cinebench R15 singlecore, the Intel leads 269 to 266. Moving to Cinebench R20, the multicore test shows 7961 for Intel versus 7877 for AMD, and the singlecore test shows 1123 versus 1111. In the newest Cinebench R23, the Intel part scores 18957 multicore and 2676 singlecore, while the Threadripper manages 18756 and 2647, respectively.

The pattern is clear: the Intel chip is ahead in every rendering workload, but the margin is so small that real-world differences would be imperceptible. The Threadripper’s two extra cores and four extra threads do not overcome Intel’s higher boost clock and architectural improvements. For example, in Cinebench R23 multicore, a 201-point difference out of nearly 19,000 points is less than 1.1%, essentially noise. The single-core results are even tighter: 2676 versus 2647 in R23, a 29-point gap. The i9-10900X’s 4.70 GHz boost clock clearly helps, but the Zen architecture’s 3.80 GHz boost is not far behind. The data shows no scenario where the Threadripper wins, so the “biggest win” for Intel is technically every test, but the biggest relative win is also 1.1%, there is no outlier.

Specification Differences

Comparing the two specification sheets reveals where they diverge beyond the benchmarks:

• Cores: Intel 10, AMD 12
• Threads: Intel 20, AMD 24
• Base Clock: Intel 3.70 GHz, AMD 3.20 GHz
• Boost Clock: Intel 4.70 GHz, AMD 3.80 GHz
• TDP: Intel 165 W, AMD 140 W
• Socket: Intel Socket 2066, AMD Socket SP3r2
• Architecture: Cascade Lake vs Zen (Whitehaven)
• Process Node: Both 14 nm, but Intel foundry vs GlobalFoundries
• Transistors: Intel null, AMD 9,600 million
• Die Size: Intel null, AMD 2x 213 mm²
• L1 Cache: Intel 64 KB per core, AMD 96 KB per core
• L2 Cache: Intel 1 MB per core, AMD 512 KB per core
• L3 Cache: Intel 19.25 MB shared, AMD 32 MB
• Memory Bandwidth: Intel null, AMD 85.3 GB/s
• PCIe: Intel Gen 3 (no lane count), AMD Gen 3 with 60 lanes (CPU only)
• Production Status: Intel Active, AMD End-of-life
• Release Date: Intel 2019-10-18, AMD null
• Part Number: Intel null, AMD YD1920A9UC9AE

Both support DDR4 memory, use quad-channel buses, lack ECC support and integrated graphics, and have unlocked multipliers. The i9-10900X has a higher TDP (165 W vs 140 W), which aligns with its higher clocks. The Threadripper’s larger L3 cache and more cores are its spec advantages, but they do not translate into benchmark wins.

FAQ

Q: Which processor wins in Cinebench R23 multicore?

A: The Intel Core i9-10900X wins with a score of 18957, compared to the AMD Ryzen Threadripper 1920’s 18756, a deltaPct of 1.1% in Intel’s favor.

Q: Does the Threadripper 1920 have more cores than the i9-10900X?

A: Yes, the Threadripper 1920 has 12 cores and 24 threads, while the i9-10900X has 10 cores and 20 threads. Despite this, the Intel part wins every benchmark test.

Q: What are the boost clock differences between the two?

A: The Intel i9-10900X boosts to 4.70 GHz, while the AMD Threadripper 1920 boosts to 3.80 GHz. The Intel part also has a higher base clock at 3.70 GHz versus 3.20 GHz.

Q: Which CPU has a larger L3 cache?

A: The AMD Ryzen Threadripper 1920 has a 32 MB L3 cache, while the Intel Core i9-10900X has 19.25 MB shared L3 cache.

Q: Is the Threadripper 1920 still in production?

A: No, the data lists the Threadripper 1920 as end-of-life, whereas the Intel i9-10900X is listed as active production.

Q: How do their average benchmark scores compare?

A: The i9-10900X has an average benchmark score of 5555, placing it in the 61st percentile of all CPUs. The Threadripper 1920 averages 5425, at the 60th percentile.

Where Each One Wins

The data shows the Intel Core i9-10900X winning all six head-to-head benchmarks, so the “where each one wins” section is lopsided. The Intel part wins in every Cinebench test, R15, R20, and R23, in both single-core and multi-core modes. Its higher boost clock (4.70 GHz) and newer Cascade Lake architecture give it a 1.1% edge across the board. This makes the i9-10900X the better pick for any workload where Cinebench scores are representative, including 3D rendering, video encoding, and general productivity. The Intel part also benefits from being an active product, meaning it may be easier to source, and its higher TDP (165 W) suggests more headroom for sustained all-core loads, though this is not a benchmark win.

The AMD Ryzen Threadripper 1920, despite zero benchmark wins, has its own theoretical advantages. It offers 12 cores and 24 threads, which is 20% more cores and threads than the Intel part, and a 32 MB L3 cache versus 19.25 MB. In workloads that scale with cache capacity or raw thread count, the Threadripper could be expected to win, but the provided benchmark data does not reflect that, it loses all Cinebench tests. It also has a lower TDP (140 W) and specifies 60 PCIe Gen 3 lanes, which is a concrete advantage for builders needing many expansion cards or NVMe drives. The Threadripper’s memory bandwidth is listed at 85.3 GB/s, a figure absent for Intel, so it may have an edge in memory-heavy tasks. However, for the benchmarks provided, the Intel i9-10900X is the clear winner, and the Threadripper 1920 is the fallback for those who prioritize its core count, cache size, or PCIe lane count over measured performance.