AMD Ryzen Threadripper PRO 9955WX vs Intel Core Ultra 9 290HX Plus Comparison

Where Each One Wins

The recorded benchmark data splits this comparison into two very different computing personalities. The AMD Ryzen Threadripper PRO 9955WX dominates across most multi-threaded and memory-sensitive workloads, taking 12 of the 17 head-to-head tests. Its wins include every Cinebench multi-core test, all but one PassMark math workload, and both sorting and compression tests. The Intel Core Ultra 9 290HX Plus counters with 5 wins, but they cluster in specific areas: single-threaded PassMark performance, floating-point math, prime number finding, and data encryption.

This is not a subtle margin in most cases. The Threadripper PRO 9955WX delivers a 49.9% lead in Cinebench R23 multi-core, a 39.8% advantage in data compression, and a 43.2% edge in integer math. Those are workload classes where the AMD processor simply outruns the Intel part by a wide margin. The Intel chip's wins are narrower in percentage terms but still meaningful: an 8.5% lead in PassMark single-thread, an 11.2% edge in data encryption, and a 22.6% advantage in floating-point math.

The use-case split is clear from the data. The AMD part is built for sustained heavy compute, particularly rendering, compression, and integer-heavy tasks. The Intel part, despite being a mobile processor, shows strength in math coprocessor-style workloads and single-threaded responsiveness. For a desktop workstation user, the AMD wins align with the tasks that dominate professional workflows. For a mobile user handling encryption or floating-point simulation, the Intel part holds its own.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen Threadripper PRO 9955WX uses the Zen 5 architecture with the codename Shimada Peak, manufactured on a 4 nm process by TSMC. The Intel Core Ultra 9 290HX Plus uses Arrow Lake-HX Refresh, built on a 3 nm process, also by TSMC. Both are active production parts, but their transistor counts and die sizes differ: the AMD chip has 16,630 million transistors across 2x 70.6 mm² dies, while the Intel chip has 17,800 million transistors on a single 243 mm² die.

Core configuration tells a large part of the story. The AMD part has 16 cores and 32 threads, meaning it uses simultaneous multithreading to double its thread count. The Intel part has 24 cores and 24 threads, so every core is a physical core with no SMT. The AMD base clock is 4.50 GHz with a 5.40 GHz boost, while the Intel base clock is 2.70 GHz with a 5.50 GHz boost. The Intel chip has a higher peak single-core clock, but the AMD chip starts from a much higher base.

Cache hierarchies diverge sharply. AMD allocates 64 KB L1 and 1 MB L2 per core, plus a 64 MB shared L3. Intel allocates 192 KB L1 and 3 MB L2 per core, but only 36 MB shared L3. The much larger L3 on the AMD side likely contributes to its compression and integer math advantages, where working sets exceed the Intel cache capacity.

Memory architecture is the most extreme difference. The AMD part supports eight-channel DDR5 with a memory bandwidth of 409.6 GB/s. The Intel part supports dual-channel DDR5 with 102.4 GB/s. That is a 4x difference in theoretical bandwidth, and it shows in the data compression and multithread results. Both support ECC memory. PCIe lane counts also differ massively: AMD provides 128 Gen 5 lanes from the CPU, Intel provides 20 Gen 5 lanes. The Intel part includes integrated Arc Xe-LPG Graphics 64EU, while the AMD part has no integrated graphics.

The market segments confirm the intent: AMD targets desktop, Intel targets mobile. The AMD socket is sTR5, the Intel socket is BGA 2114. The AMD TDP is 350 watts, the Intel TDP is 55 watts. These are not competing in the same physical class, but the benchmark data still allows a direct comparison of compute capability.

Head-to-Head Benchmarks

The Cinebench results establish the AMD processor as the clear multi-core leader. In Cinebench R15 multi-core, the AMD scores 5996 against the Intel 5981, a narrow 0.3% margin. That is essentially a tie, suggesting the workload fits both processors equally at that short duration. But Cinebench R20 multi-core shows a 17.9% AMD lead with 24987 against 21198, and Cinebench R23 multi-core extends that to a 49.9% margin with 59494 against 39684. The longer the workload runs, the more the AMD part pulls ahead, which points to sustained throughput advantages from memory bandwidth and higher base clocks.

Single-core Cinebench results are unusual and require careful reading. The AMD scores 846 in R15 single-core against the Intel 340, a 148.8% lead. In R20 single-core, AMD scores 3527 against 2992, a 17.9% lead. In R23 single-core, AMD scores 8399 against 2356, a 256.5% lead. The Intel single-core scores vary wildly across Cinebench versions, while the AMD scores scale predictably. The database records these numbers as measured, and they indicate the Intel part's single-core Cinebench performance is inconsistent across test versions.

PassMark results show a more balanced picture. The AMD wins data compression with 920954 against 658724, a 39.8% margin. The AMD wins extended instructions with 76363 against 51290, a 48.9% margin. The AMD wins integer math with 236120 against 164839, a 43.2% margin. The AMD wins multithread with 67035 against 59439, a 12.8% margin. The AMD wins physics with 4156 against 3387, a 22.7% margin. The AMD wins random string sorting with 99813 against 80327, a 24.3% margin.

The Intel wins are concentrated in four PassMark sub-tests. Data encryption favors Intel with 50008 against 44389, an 11.2% margin. Prime number finding favors Intel with 519 against 337, a 35.1% margin. Floating-point math favors Intel with 201773 against 156215, a 22.6% margin. PassMark single-thread favors Intel with 4951 against 4530, an 8.5% margin. These Intel wins suggest strong execution units for encryption, primality testing, and floating-point throughput, plus a higher boost clock for single-threaded tasks.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 9 290HX Plus has 24 cores and 24 threads. The AMD Ryzen Threadripper PRO 9955WX has 16 cores and 32 threads. The Intel part has more physical cores, but the AMD part has more total threads due to simultaneous multithreading.

Q: What is the memory bandwidth difference?

A: The AMD processor supports eight-channel DDR5 with 409.6 GB/s bandwidth. The Intel processor supports dual-channel DDR5 with 102.4 GB/s bandwidth. The AMD part has four times the theoretical memory bandwidth.

Q: Which processor wins more benchmarks overall?

A: The AMD Ryzen Threadripper PRO 9955WX wins 12 of the 17 head-to-head benchmark comparisons. The Intel Core Ultra 9 290HX Plus wins 5.

Q: Does the Intel processor have integrated graphics?

A: Yes, the Intel Core Ultra 9 290HX Plus includes Arc Xe-LPG Graphics 64EU. The AMD Ryzen Threadripper PRO 9955WX has no integrated graphics, listed as N/A.

Q: What is the largest single benchmark margin?

A: The largest margin is in Cinebench R23 single-core, where the AMD scores 8399 against the Intel 2356, a 256.5% lead. The largest Intel lead is 35.1% in PassMark find prime numbers.

Q: How do the average benchmark scores compare?

A: The AMD processor has an average benchmark score of 101041 and sits at the 97th percentile of all CPUs. The Intel processor has an average benchmark score of 79574 and sits at the 95th percentile.

The Verdict

The data points to a clear division of roles. The AMD Ryzen Threadripper PRO 9955WX is the stronger processor for multi-threaded compute, sustained rendering, compression, integer math, and memory-heavy workloads. Its 49.9% lead in Cinebench R23 multi-core and 39.8% lead in data compression are decisive. The 12-5 win count in head-to-head benchmarks confirms this direction. Its nearest rivals in the database, including the AMD EPYC 7513 and AMD Ryzen Threadripper PRO 5965WX, sit within a few percent of its average score, placing it in serious server-class territory.

The Intel Core Ultra 9 290HX Plus is a different kind of performer. It wins floating-point math by 22.6%, prime number finding by 35.1%, and encryption by 11.2%. Its PassMark single-thread score of 4951 is 8.5% higher than the AMD's 4530, which indicates better raw single-threaded throughput in that specific measurement. With a 55 watt TDP and integrated graphics, it is designed for mobile systems, and its results should be read in that context. Its nearest rivals include the Intel Core i9-14900KF and Core i9-14900K, which it matches within 0.6%, showing it performs like a high-end desktop chip in a mobile envelope.

The recorded data does not support a single "best" conclusion. It supports two conclusions based on workload. For multithreaded compute and memory bandwidth, the AMD Ryzen Threadripper PRO 9955WX is the clear choice. For floating-point math, encryption, and single-threaded PassMark performance, the Intel Core Ultra 9 290HX Plus demonstrates advantages that matter in those specific tasks.

Specification Differences

| Specification | AMD Ryzen Threadripper PRO 9955WX | Intel Core Ultra 9 290HX Plus |

|---|---|---|

| Cores | 16 | 24 |

| Threads | 32 | 24 |

| Base clock | 4.50 GHz | 2.70 GHz |

| Boost clock | 5.40 GHz | 5.50 GHz |

| TDP | 350 W | 55 W |

| Socket | AMD Socket sTR5 | Intel BGA 2114 |

| Architecture | Zen 5 | Not specified |

| Codename | Shimada Peak | Arrow Lake-HX Refresh |

| Process node | 4 nm | 3 nm |

| Foundry | TSMC | TSMC |

| Transistors | 16,630 million | 17,800 million |

| Die size | 2x 70.6 mm² | 243 mm² |

| L1 cache | 64 KB (per core) | 192 KB (per core) |

| L2 cache | 1 MB (per core) | 3 MB (per core) |

| L3 cache | 64 MB | 36 MB (shared) |

| Memory bus | Eight-channel | Dual-channel |

| Memory bandwidth | 409.6 GB/s | 102.4 GB/s |

| ECC memory | Yes | Yes |

| PCIe | Gen 5, 128 lanes | Gen 5, 20 lanes |

| Integrated graphics | N/A | Arc Xe-LPG Graphics 64EU |

| Market segment | Desktop | Mobile |

| Release date | 2025-07-22 | 2026-03-16 |

| Launch MSRP | $1649 | Not listed |

| Unlocked multiplier | Yes | Yes |