AMD Ryzen Threadripper 2970WX vs Intel Core i9-9940X Comparison

The Verdict

The benchmark data splits this comparison cleanly by workload type. The AMD Ryzen Threadripper 2970WX wins six of eight head-to-head tests, taking every Cinebench iteration across R15, R20, and R23 in both single-core and multi-core. The Intel Core i9-9940X counters with decisive wins in Geekbench, both single-core and multi-core. For users whose primary tools resemble Cinebench’s rendering-style load, the 2970WX is the stronger pick. For those whose applications behave more like Geekbench’s mixed integer/float patterns, the 9940X has the edge. The 2970WX also carries the advantage of 24 cores versus 14, though the 9940X shows that raw core count alone does not guarantee superior performance in every benchmark. The 2970WX remains in active production while the 9940X is end-of-life, which may factor into platform longevity decisions. The 2970WX carries a launch MSRP of $1299, and the 9940X carries a launch MSRP of $1387.

Architecture Differences

The two processors represent fundamentally different design philosophies. The 2970WX uses AMD’s Zen+ architecture on a 12 nm process from GlobalFoundries, built across four dies with a combined die size of 4x 213 mm² and 19,200 million transistors. It operates on AMD Socket SP3r2 within the Ryzen Threadripper 2000 series, codenamed Colfax. The 9940X uses Intel’s Skylake architecture on a 14 nm process, fabricated on a single 484 mm² die, and uses Intel Socket 2066 as part of the Core i9 X-Series 9th Generation.

Core counts diverge sharply: the 2970WX has 24 cores and 48 threads, while the 9940X has 14 cores and 28 threads. That is a 10-core and 20-thread gap in AMD’s favor. Clock speeds tell a different story. The 9940X has a base clock of 3.30 GHz and boosts to 4.50 GHz, while the 2970WX runs at 3.00 GHz base and 4.20 GHz boost. Intel’s higher clocks on fewer cores explain part of its Geekbench advantage.

Cache configurations also differ structurally. The 2970WX allocates 96 KB of L1 per core, 512 KB of L2 per core, and 64 MB of L3. The 9940X uses 64 KB of L1 per core, a larger 1 MB L2 per core, and 19.25 MB of shared L3. Intel’s larger per-core L2 may help latency-sensitive workloads, while AMD’s much larger L3 total favors data-heavy parallel jobs.

Memory bandwidth favors AMD at 93.9 GB/s versus Intel’s 85.3 GB/s, though both use quad-channel DDR4. PCIe lane availability also differs: the 2970WX provides 60 Gen 3 lanes from the CPU, while the 9940X provides 44. Neither processor includes integrated graphics, and neither supports ECC memory. Both have unlocked multipliers, but the production statuses differ — the 2970WX is active, and the 9940X is end-of-life.

Head-to-Head Benchmarks

The Cinebench suite is a clean sweep for the 2970WX, and the margin is remarkably consistent. In Cinebench R15 multi-core, AMD scores 2651 against Intel’s 2400, a 10.5% advantage. The single-core R15 result is nearly identical in percentage: 374 versus 338, also 10.7% ahead. Cinebench R20 repeats the pattern — multi-core 11048 versus 10000 (10.5% ahead) and single-core 1559 versus 1411 (10.5% ahead). Cinebench R23 follows suit exactly: multi-core 26305 versus 23810 (10.5% ahead) and single-core 3713 versus 3361 (10.5% ahead). The consistency of that 10.5% delta across three generations of Cinebench is striking; it suggests a stable architectural difference in how each CPU handles that specific rendering workload, independent of test version.

The 9940X’s wins are not marginal — they are substantial. In Geekbench multi-core, Intel scores 10466 to AMD’s 7195, a 31.3% advantage. That is the largest single delta in the entire comparison. In Geekbench single-core, Intel scores 1468 against 1236, a 15.8% margin. These are not small victories; they indicate a fundamentally different performance profile in Geekbench’s workload mix. The 9940X manages to beat a CPU with 10 more cores in Geekbench multi-core by nearly a third, which points to significant per-core efficiency or memory latency advantages in that benchmark’s specific pattern.

Looking at the broader benchmark averages, the two CPUs land close together. The 2970WX has an average benchmark score of 6760, while the 9940X sits at 6657. Both occupy the 62nd percentile against all CPUs. The nearest rivals for the 2970WX include the Intel Core i7-12700E (avg 6776, -0.2% delta), the Intel Xeon Gold 6154 (avg 6803, -0.6%), the Intel Xeon D-2775TE (avg 6711, +0.7%), and the Intel Xeon Gold 5317 (avg 6710, +0.7%). For the 9940X, the nearest rivals are the AMD Ryzen Threadripper 2950X (avg 6647, +0.1%), the Intel Core i5-13400E (avg 6638, +0.3%), the AMD EPYC 72F3 (avg 6700, -0.6%), and the Intel Core i9-12900E (avg 6611, +0.7%). These rival deltas show both CPUs sitting within a 1% band of several other high-end processors, making the head-to-head results more meaningful than the overall averages.

FAQ

Q: Which CPU wins more benchmarks overall?

A: The AMD Ryzen Threadripper 2970WX wins 6 of 8 head-to-head tests. The Intel Core i9-9940X wins the remaining 2.

Q: How large is the 2970WX’s Cinebench advantage?

A: Across all six Cinebench tests (R15, R20, R23 in both single-core and multi-core), the 2970WX leads by 10.5% or 10.7% in every case. The single-core R15 result is 10.7%, and all others are 10.5%.

Q: In which benchmark does the Intel CPU perform best relative to AMD?

A: Geekbench multi-core is the 9940X’s strongest result. It scores 10466 versus 7195, a 31.3% advantage — the largest delta in the entire comparison.

Q: How do the two CPUs compare in Geekbench single-core?

A: The 9940X scores 1468 against the 2970WX’s 1236, giving Intel a 15.8% lead. This is its second consecutive win after the multi-core Geekbench result.

Q: Are the CPUs similar in overall benchmark averages?

A: Yes. The 2970WX averages 6760 and the 9940X averages 6657. Both sit at the 62nd percentile against all CPUs, and both have nearest rivals within roughly 1% of their average score.

Q: What are the production statuses of these CPUs?

A: The 2970WX is listed as active production. The 9940X is listed as end-of-life.

Where Each One Wins

The 2970WX is the clear choice for rendering-style workloads. Its six Cinebench victories — each at a consistent 10.5% margin — indicate that applications built around the same rendering pipeline as Cinebench will consistently favor AMD. The 24-core, 48-thread configuration with 64 MB of L3 cache and 93.9 GB/s memory bandwidth provides the resources that such workloads exploit. The 2970WX also offers 60 PCIe Gen 3 lanes, which matters for systems with multiple GPUs or high-bandwidth storage devices. With its active production status, it remains a viable current purchase.

The 9940X wins where Geekbench’s workload pattern dominates. Its 31.3% multi-core and 15.8% single-core Geekbench margins are decisive, and they come despite the 2970WX having 10 more cores. The 9940X’s higher clock speeds (3.30 GHz base, 4.50 GHz boost) and larger per-core L2 cache (1 MB versus 512 KB) likely contribute to this. Users running applications with Geekbench-like mixed integer and floating-point patterns, or those sensitive to per-core throughput rather than raw core count, should favor Intel. The 9940X also has a lower TDP of 165 watts versus 250 watts, which may simplify cooling requirements despite its end-of-life status.

For users deciding between the two, the benchmark data suggests a straightforward rule: rendering and multi-threaded throughput favor AMD; mixed workload efficiency and per-core performance favor Intel. The 2970WX’s 6-2 win count in head-to-head tests gives it the overall edge, but the magnitude of the 9940X’s Geekbench victories means that applications resembling that test will see a materially better experience on Intel. Both CPUs occupy the same 62nd percentile overall, meaning neither is a universal upgrade over the other — the choice depends entirely on the target software profile.