AMD Ryzen 9 PRO 8945HS vs Intel Core Ultra 5 336H Comparison

Head-to-Head Benchmarks

The recorded benchmark data shows a clear split between the AMD Ryzen 9 PRO 8945HS and the Intel Core Ultra 5 336H, with AMD taking 12 of the 17 head-to-head comparisons. The most decisive AMD victories come in throughput-oriented workloads. In PassMark integer math, the Ryzen 9 PRO 8945HS scores 103,390 against 62,070 for the Intel part, a 66.6% advantage. Data compression shows a 31.6% gap (368,884 vs. 280,340), and random string sorting favors AMD by 29.8% (44,668 vs. 34,402). Extended instructions also go AMD's way by 12.9% (27,714 vs. 24,542).

The Cinebench suite is consistently in AMD's favor, though by narrower margins. Across R15, R20, and R23, the multi-core deltas sit at 4.9% in each case, with scores of 2,536 vs. 2,418 (R15), 10,569 vs. 10,076 (R20), and 25,165 vs. 23,991 (R23). Single-core Cinebench results show the same pattern: R15 at 357 vs. 341 (4.7%), R20 at 1,491 vs. 1,422 (4.9%), and R23 at 3,552 vs. 3,387 (4.9%). PassMark multithread also lands with AMD ahead by 6.4% (30,378 vs. 28,545), while data encryption is nearly even at 2.5% (21,820 vs. 21,291).

Intel secures five wins, and they are concentrated in specific compute patterns. The largest Intel victory is in prime number finding, where the Core Ultra 5 336H scores 299 against 91 for the AMD chip, a 69.6% margin. Physics simulation goes to Intel by 46.7% (2,682 vs. 1,430), and floating-point math favors Intel by 23% (82,415 vs. 63,490). Single-thread PassMark scores put Intel ahead by 2.3% (4,013 vs. 3,920), which appears twice in the data as both "single_thread" and "singlethread" entries.

Where Each One Wins

The Ryzen 9 PRO 8945HS dominates integer-heavy and data-management tasks. Integer math, data compression, random string sorting, and extended instructions are all AMD strengths, suggesting the Zen 4 architecture handles branch-heavy and data-manipulation code efficiently. The Cinebench family, which stresses sustained multi-core rendering and single-core responsiveness, also consistently favors AMD, though the margins are modest. For users running compression tools, encryption workloads, or general productivity suites, the AMD part shows a measurable edge across the board.

The Core Ultra 5 336H wins in floating-point math, physics simulation, and prime number enumeration. These results point to a different compute profile, one where the Intel Panther Lake design excels at vectorized math and scientific-style workloads. The 46.7% physics lead and the 69.6% prime number advantage are substantial, not marginal. Single-thread PassMark also goes Intel's way, indicating that for lightly threaded, short-burst tasks, the Intel chip can edge ahead, even though Cinebench single-core scores favor AMD.

The practical split is this: AMD for integer throughput, compression, encryption, and Cinebench-style rendering; Intel for floating-point math, physics, and prime sieve algorithms. The multithread PassMark score (30,378 vs. 28,545) and the overall benchmark average (41,963 vs. 34,485) both favor AMD, but the Intel part is not a clean sweep loser in every scenario.

Architecture Differences

The two processors come from fundamentally different designs. The AMD Ryzen 9 PRO 8945HS uses Zen 4 architecture on the Hawk Point codename, built on a 4 nm process at TSMC. It has 8 cores and 16 threads, with a base clock of 4.00 GHz and a boost clock of 5.20 GHz. The TDP is 45 watts, and it uses AMD Socket FP7. The Intel Core Ultra 5 336H uses Panther Lake architecture, built on a 3 nm process at Intel, with a base clock of 1.90 GHz and a boost clock of 4.60 GHz. It has 16 cores and 16 threads, with a 25 watt TDP, on Intel BGA 2540.

The cache layouts differ substantially. AMD provides 64 KB of L1 per core, 1 MB of L2 per core, and 16 MB of shared L3. Intel provides 192 KB of L1 per core, 2.5 MB of L2 per core, and 18 MB of shared L3. The Intel L3 pool is 2 MB larger, but the per-core L2 allocation is 2.5x higher on the Intel side. The transistor count for AMD is listed at 25,000 million with a die size of 178 mm², while the Intel figures are not recorded in the database.

Memory support also differs. AMD supports DDR5 with dual-channel memory and a bandwidth of 89.6 GB/s, plus ECC memory support. Intel supports both DDR5 and LPDDR5X, also dual-channel, with a higher bandwidth of 115.2 GB/s, but no ECC. PCIe generations differ: AMD uses Gen 4 with 20 CPU lanes, while Intel uses Gen 5 with 12 CPU lanes. Integrated graphics are the Radeon 780M on the AMD side and Intel Xe3 Graphics on the Intel side. The AMD part launched in April 2024; the Intel part launched in January 2026, both in the mobile segment and both currently active.

FAQ

Q: Which processor wins more head-to-head benchmarks?

A: The AMD Ryzen 9 PRO 8945HS wins 12 of the 17 recorded comparisons, while the Intel Core Ultra 5 336H wins 5.

Q: What is the largest performance gap in either direction?

A: The biggest AMD win is 66.6% in PassMark integer math (103,390 vs. 62,070). The biggest Intel win is 69.6% in PassMark find prime numbers (299 vs. 91).

Q: How do the Cinebench scores compare?

A: AMD leads all six Cinebench tests. Multi-core margins are 4.9% in R15, R20, and R23. Single-core margins are 4.7% in R15 and 4.9% in R20 and R23.

Q: Does the Intel part have any single-thread advantage?

A: Yes, in PassMark single-thread tests the Intel chip scores 4,013 against 3,920 for AMD, a 2.3% lead. However, Cinebench single-core scores favor AMD by 4.7% to 4.9%.

Q: What are the core and thread counts?

A: The AMD part has 8 cores and 16 threads. The Intel part has 16 cores and 16 threads.

Q: Which processor has higher memory bandwidth?

A: The Intel Core Ultra 5 336H records 115.2 GB/s, compared to 89.6 GB/s for the AMD Ryzen 9 PRO 8945HS. Intel also supports LPDDR5X in addition to DDR5, while AMD supports only DDR5.

The Verdict

The data points to the AMD Ryzen 9 PRO 8945HS as the stronger overall processor. Its average benchmark score of 41,963 places it in the 88th percentile of all CPUs, while the Intel Core Ultra 5 336H sits at 34,485 in the 84th percentile. The nearest rivals for AMD include the Intel Core i7-14700T at 41,914 (0.1% behind) and the AMD Ryzen 5 7400 at 42,055 (0.2% ahead), showing the AMD chip is tightly clustered with those parts. The Intel chip's nearest rivals are the Intel Core i7-13700HX at 34,554 (0.2% ahead) and the Intel Core i5-13450HX at 34,333 (0.4% behind), indicating a lower performance tier overall.

For workloads that resemble Cinebench rendering, integer math, compression, encryption, or random string sorting, the Ryzen 9 PRO 8945HS is the pick from the recorded results. Its 12 wins include all six Cinebench tests and the multithread PassMark score. The 45 watt TDP is higher than the Intel part's 25 watts, which may matter in system design, but the performance data does not show a penalty in the benchmark scores.

For floating-point math, physics simulation, prime number finding, and PassMark single-thread tests, the Intel Core Ultra 5 336H is the better choice from the numbers. The 46.7% physics lead and the 69.6% prime number lead are too large to ignore for those specific tasks. The Intel part also offers higher memory bandwidth (115.2 GB/s vs. 89.6 GB/s) and PCIe Gen 5 support, though with fewer CPU lanes (12 vs. 20).

The percentile ranking separates the two clearly: 88th for AMD versus 84th for Intel. The AMD chip also has a higher boost clock (5.20 GHz vs. 4.60 GHz) and a higher base clock (4.00 GHz vs. 1.90 GHz). The Intel chip counters with more cores (16 vs. 8), a larger L3 cache (18 MB vs. 16 MB), and a smaller process node (3 nm vs. 4 nm). For general-purpose mobile computing, the benchmark average favors AMD. For specialized math and physics workloads, the Intel part delivers specific wins that the AMD chip cannot match in the recorded data.