AMD Ryzen 9 5900XT vs Intel Core Ultra X7 368H Comparison

Head-to-Head Benchmarks

The recorded benchmark data shows a clear overall advantage for the AMD Ryzen 9 5900XT, which wins 12 of the 17 shared tests compared to 5 for the Intel Core Ultra X7 368H. The average benchmark score for the AMD part is 50,718, placing it in the 90th percentile of all CPUs, while the Intel chip scores 40,518 and sits in the 87th percentile. The AMD processor also holds a 25.2% lead in average benchmark score over the Intel part.

The most lopsided victories for the AMD Ryzen 9 5900XT come in memory and integer workloads. In PassMark integer math, the AMD part scores 177,566 versus 88,083 for the Intel chip, a 101.6% advantage that more than doubles the output. Data compression shows a 91.1% gap, with the AMD processor at 597,862 and the Intel chip at 312,927. Random string sorting also favors AMD by 64.2%, with scores of 62,537 versus 38,093. Extended instructions run 52.5% faster on the AMD side, and data encryption shows a 49.9% edge.

Cinebench results consistently favor the AMD Ryzen 9 5900XT across all six recorded tests. In Cinebench R23 multi-core, the AMD part scores 37,373 against 28,221 for Intel, a 32.4% lead. The same 32.4% delta appears in Cinebench R23 single-core (5,276 versus 3,984), R20 multi-core (15,696 versus 11,852), R20 single-core (2,215 versus 1,673), R15 multi-core (3,767 versus 2,844), and R15 single-core (531 versus 401). PassMark multi-thread also goes to AMD by 32.9%, with 43,810 versus 32,956.

The Intel Core Ultra X7 368H takes its wins in physics, prime number finding, floating-point math, and single-thread tests. The largest Intel margin is in PassMark physics, where it scores 2,857 against 1,715 for AMD, a 40% advantage. In PassMark find prime numbers, Intel leads 321 to 205, a 36.1% gap. PassMark single-thread shows Intel ahead by 13.3%, with 4,005 versus 3,474. Floating-point math is the narrowest Intel win, at 105,681 versus 99,398, a 5.9% margin.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen 9 5900XT records an average benchmark score of 50,718, while the Intel Core Ultra X7 368H scores 40,518. The AMD part also ranks in the 90th percentile of all CPUs, three points above the Intel chip's 87th percentile.

Q: How large is the AMD lead in multi-core rendering?

A: Across all three Cinebench R15, R20, and R23 multi-core tests, the AMD Ryzen 9 5900XT leads by 32.4%. In Cinebench R23 multi-core specifically, AMD scores 37,373 versus 28,221 for Intel.

Q: In which test does the Intel chip have its biggest win?

A: The Intel Core Ultra X7 368H shows its largest advantage in PassMark physics, scoring 2,857 against AMD's 1,715, a 40% lead.

Q: Does the Intel chip win any single-thread tests?

A: Yes, the Intel Core Ultra X7 368H wins PassMark single-thread with 4,005 versus 3,474 for AMD, a 13.3% advantage. However, AMD wins all three Cinebench single-core tests (R15, R20, R23) by 32.4% each.

Q: What is the closest benchmark result between the two?

A: The narrowest margin is in PassMark floating-point math, where the Intel chip scores 105,681 and the AMD part scores 99,398, a 5.9% difference.

Q: Which processor has the higher multi-thread score in PassMark?

A: The AMD Ryzen 9 5900XT records 43,810 in PassMark multi-thread, which is 32.9% higher than the Intel Core Ultra X7 368H's 32,956.

Where Each One Wins

The AMD Ryzen 9 5900XT dominates in almost every throughput-oriented workload. Its 16 cores with 32 threads give it a structural edge in parallel tasks, and the data confirms this across rendering, compression, encryption, and integer math. For users running Cinebench-style render workloads, the AMD part delivers 32.4% higher scores in every multi-core test. The 91.1% advantage in data compression and the 101.6% lead in integer math indicate strong performance in archive handling, data processing, and general number-crunching. The 49.9% edge in data encryption and 52.5% lead in extended instructions also position it well for security and SIMD-heavy applications. Random string sorting at 64.2% ahead suggests faster database and text-processing tasks.

The Intel Core Ultra X7 368H wins in a narrower set of workloads, but those wins are meaningful for specific use cases. PassMark physics at 40% higher output points to better performance in simulation and physics-based calculations. The 36.1% lead in find prime numbers indicates efficiency in prime-number generation, a workload that benefits from specific instruction patterns. Floating-point math at 5.9% higher shows a modest edge in FP-heavy scientific computing. The single-thread victory at 13.3% in PassMark single-thread suggests snappier response in lightly threaded applications, though Cinebench single-core results contradict that picture with AMD winning by 32.4% across all three versions. The Intel part's 16 threads versus 32 on AMD explains its multi-thread deficits, but its higher boost clock of 5.00 GHz versus 4.80 GHz contributes to its single-thread PassMark win.

Users prioritizing multi-core rendering, data compression, encryption, and integer workloads should look to the AMD Ryzen 9 5900XT. Users with workloads centered on physics simulation, prime finding, or specific floating-point tasks may see better results from the Intel Core Ultra X7 368H, despite its lower overall average score.

Specification Differences

The two processors differ in nearly every core specification. The AMD Ryzen 9 5900XT uses 16 cores with 32 threads, while the Intel Core Ultra X7 368H also has 16 cores but only 16 threads, meaning no simultaneous multithreading. Base clocks differ significantly: the AMD chip runs at 3.30 GHz, while the Intel chip starts at 2.00 GHz. Boost clocks reverse the order, with Intel reaching 5.00 GHz and AMD topping out at 4.80 GHz.

Thermal design power shows a major gap. The AMD Ryzen 9 5900XT has a TDP of 105 watts, while the Intel Core Ultra X7 368H draws only 25 watts. The AMD part uses AMD Socket AM4, while the Intel chip uses Intel BGA 2540. Memory support differs completely: AMD uses DDR4 with dual-channel access and 51.2 GB/s bandwidth, while Intel uses LPDDR5X with dual-channel access and 153.6 GB/s bandwidth. The AMD chip supports ECC memory; the Intel chip does not.

PCI Express connectivity also differs. The AMD Ryzen 9 5900XT provides Gen 4 with 20 lanes from the CPU, while the Intel Core Ultra X7 368H provides Gen 5 with only 4 lanes from the CPU. The AMD part has no integrated graphics, while the Intel chip includes Arc B390 graphics. The AMD multiplier is unlocked, allowing overclocking, while the Intel multiplier is locked. The AMD processor launched with an MSRP of $349 on July 30, 2024. The Intel chip has no recorded launch MSRP and released January 4, 2026.

Architecture Differences

The AMD Ryzen 9 5900XT is built on the Zen 3 architecture with the Vermeer codename, part of the Ryzen 9 generation. It uses a 7 nm process from TSMC and contains 8,300 million transistors across a dual-die design with each die measuring 74 mm². Cache configuration includes 64 KB of L1 per core, 512 KB of L2 per core, and 64 MB of L3 cache. It targets the desktop market segment.

The Intel Core Ultra X7 368H uses the Panther Lake architecture with the Panther Lake-H codename, part of the Core Ultra Series 3 and Ultra X7 generation. It is fabricated on Intel's 3 nm process. The database does not record transistor count or die size for this part. Cache sizes are larger per core: 192 KB of L1 per core and 2.5 MB of L2 per core, but L3 is only 18 MB shared, significantly less than the AMD chip's 64 MB. It targets the mobile segment with integrated Arc B390 graphics.

The architectural split explains the benchmark behavior. The AMD part's 32 threads double the Intel chip's thread count, enabling its multi-core wins across Cinebench and PassMark multi-thread. The Intel chip's smaller 25-watt TDP and mobile positioning align with its 3 nm process, while the AMD part's 105-watt desktop design allows higher sustained throughput. The Intel chip's larger per-core L1 and L2 caches help its single-thread PassMark win and physics performance, while the AMD part's 64 MB L3 cache supports its large data-set workloads like compression and encryption. The Intel chip's 153.6 GB/s memory bandwidth is three times the AMD part's 51.2 GB/s, yet the AMD processor still wins memory-intensive tests like data compression by 91.1%, suggesting that thread count and cache size matter more in those workloads than raw bandwidth.