AMD Ryzen 9 PRO 8945HS vs Intel Core Ultra 5 338H Comparison

Where Each One Wins

The benchmark split between the AMD Ryzen 9 PRO 8945HS and the Intel Core Ultra 5 338H is not a matter of overall superiority, but of workload class. The AMD part takes 12 of the 17 recorded head-to-head wins, while the Intel part claims 5. That distribution tells a clear story: the Ryzen 9 PRO dominates threaded productivity and memory-intensive workloads, while the Intel chip wins in specific compute patterns that favor its architecture.

The AMD Ryzen 9 PRO 8945HS wins every Cinebench test in the database. The multicore margins range from a modest 1.3% in Cinebench R15 multicore to a massive 54.1% in Cinebench R23 multicore and 73.8% in Cinebench R23 singlecore. Single-core performance is not close: the AMD chip leads by 17% in Cinebench R15 singlecore and 3.5% in Cinebench R20 singlecore. For rendering, video encoding, and any workload that scales across threads, the Ryzen 9 PRO is the clear pick.

The AMD chip also wins the PassMark integer math test by 59.2%, data compression by 33.4%, random string sorting by 31.1%, and extended instructions by 15.9%. The multithread score favors AMD by 5.8%, and data encryption is a narrow AMD win at 2.1%. These results point to strong general-purpose compute, especially in compression, sorting, and encryption tasks.

The Intel Core Ultra 5 338H wins the remaining five tests. The most dramatic Intel victories are in PassMark find prime numbers, where it leads by 70.1%, and PassMark physics, where it leads by 47%. It also wins floating point math by 24.5% and single-thread performance by 6.2%. These wins suggest the Intel architecture handles branch-heavy integer loops, physics simulation, and floating-point throughput better than the AMD chip, despite losing most other tests.

The PassMark single-thread score is interesting: Intel leads 4180 to 3920, a 6.2% advantage, even though the AMD chip wins every Cinebench singlecore test. That discrepancy reflects different workload characteristics between the two benchmark suites, not a contradiction in the data. The AMD chip has a higher boost clock at 5.20 GHz versus 4.70 GHz, but the Intel part still edges out the PassMark single-thread test.

The Verdict

The data supports a straightforward split. Users whose workloads are dominated by Cinebench-style rendering, data compression, encryption, sorting, and integer math should choose the AMD Ryzen 9 PRO 8945HS. The 54.1% lead in Cinebench R23 multicore is the single largest margin in the entire comparison, and the 59.2% lead in integer math reinforces that this chip is built for sustained multi-threaded throughput.

Users whose workloads involve physics simulation, prime number finding, floating-point math, or single-threaded PassMark tasks should choose the Intel Core Ultra 5 338H. The 70.1% win in find prime numbers and 47% win in physics are decisive, and the 24.5% advantage in floating-point math shows a real architectural strength. The Intel chip also runs at a lower 25 W TDP versus 45 W for the AMD chip, which matters for thermally constrained mobile designs.

The overall average benchmark scores confirm the AMD chip as the stronger all-round performer. The Ryzen 9 PRO 8945HS averages 41963 across all recorded benchmarks, placing it in the 88th percentile of all CPUs. The Core Ultra 5 338H averages 33989, placing it in the 84th percentile. The AMD chip sits between the Intel Core i7-14700T (41914, 0.1% behind) and the AMD Ryzen 5 7400 (42055, 0.2% ahead). The Intel chip sits between the Intel Core Ultra 7 165H (34083, 0.3% ahead) and the Intel Core i7-12800HX (33875, 0.3% behind).

The AMD chip also leads in memory bandwidth at 89.6 GB/s versus 136.5 GB/s for Intel, which is a counterintuitive result: the Intel chip has higher raw memory bandwidth, yet the AMD chip wins the memory-sensitive compression and sorting tests. The AMD chip supports ECC memory, while the Intel chip does not. For professional or reliability-focused deployments, that is a meaningful differentiator.

Head-to-Head Benchmarks

The largest AMD win is Cinebench R23 singlecore, where the Ryzen 9 PRO 8945HS scores 3552 against 2044 for the Core Ultra 5 338H, a 73.8% margin. That is not a small edge; it is a generational gap in single-threaded rendering performance. The Cinebench R23 multicore result is nearly as stark: 25165 versus 16331, a 54.1% advantage. These two tests alone establish the AMD chip as the rendering processor in this comparison.

PassMark integer math shows a 59.2% AMD lead, with scores of 103390 against 64934. Data compression goes to AMD by 33.4%, with 368884 against 276539. Random string sorting is an AMD win by 31.1%, at 44668 against 34082. Extended instructions favor AMD by 15.9%, at 27714 versus 23906. The multithread test is closer, but still an AMD win: 30378 versus 28717, a 5.8% margin. Data encryption is the tightest AMD win at 2.1%, with 21820 versus 21367.

The Intel wins are concentrated in compute patterns that reward its core topology. The find prime numbers test shows Intel at 304 against AMD at 91, a 70.1% Intel lead. Physics favors Intel by 47%, with 2697 against 1430. Floating point math goes to Intel by 24.5%, at 84067 versus 63490. The PassMark single-thread test is an Intel win by 6.2%, with 4180 versus 3920. The Cinebench results tell a different story for single-threaded performance, but the PassMark single-thread score is the recorded data point that favors Intel.

The Cinebench R15 and R20 multicore results are close, with AMD winning by 1.3% and 3.5% respectively. Cinebench R15 singlecore is a 17% AMD win, and Cinebench R20 singlecore is a 3.5% AMD win. These smaller margins show that the gap narrows in older rendering workloads, but the direction never flips: AMD wins every Cinebench test in the database.

FAQ

Q: Which processor is faster in Cinebench R23 multicore?

A: The AMD Ryzen 9 PRO 8945HS scores 25165, which is 54.1% ahead of the Intel Core Ultra 5 338H at 16331.

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

A: Yes. The Intel Core Ultra 5 338H wins the PassMark single-thread test with 4180 against 3920, a 6.2% margin. However, the AMD chip wins every Cinebench singlecore test, including a 73.8% margin in Cinebench R23 singlecore.

Q: Which chip has more cores and threads?

A: The Intel Core Ultra 5 338H has 12 cores and 12 threads. The AMD Ryzen 9 PRO 8945HS has 8 cores and 16 threads, meaning it uses simultaneous multithreading while the Intel chip does not.

Q: How do the average benchmark scores compare?

A: The AMD Ryzen 9 PRO 8945HS averages 41963 across all recorded benchmarks, placing it in the 88th percentile of all CPUs. The Intel Core Ultra 5 338H averages 33989, placing it in the 84th percentile.

Q: Which chip supports ECC memory?

A: The AMD Ryzen 9 PRO 8945HS supports ECC memory. The Intel Core Ultra 5 338H does not.

Q: What is the memory bandwidth difference?

A: The Intel Core Ultra 5 338H has a memory bandwidth of 136.5 GB/s, while the AMD Ryzen 9 PRO 8945HS has 89.6 GB/s. Despite the higher Intel bandwidth, the AMD chip wins the data compression test by 33.4%.

Architecture Differences

The two processors come from different manufacturing nodes and design philosophies. The AMD Ryzen 9 PRO 8945HS uses TSMC's 4 nm process and packs 25,000 million transistors into a 178 mm² die. The Intel Core Ultra 5 338H uses Intel's 3 nm process. The AMD chip is part of the 8000 series, built on Zen 4 architecture with the Hawk Point codename. The Intel chip belongs to the Core Ultra Series 3, built on Panther Lake architecture with the Panther Lake-H codename.

Core counts differ significantly. The AMD chip has 8 cores and 16 threads, while the Intel chip has 12 cores and 12 threads. The AMD chip relies on simultaneous multithreading to reach 16 threads from 8 cores; the Intel chip has no multithreading, so 12 cores means 12 threads. This explains why the AMD chip wins most multithreaded tests despite having fewer physical cores.

Cache hierarchies also differ. The AMD chip has 64 KB of L1 per core, 1 MB of L2 per core, and 16 MB of shared L3. The Intel chip has 192 KB of L1 per core, 2.5 MB of L2 per core, and 18 MB of shared L3. The Intel chip has more L3 cache overall and a larger per-core L1 and L2, which likely contributes to its wins in prime finding and physics tests.

Clock speeds favor AMD. The Ryzen 9 PRO 8945HS has a base clock of 4.00 GHz and a boost clock of 5.20 GHz. The Core Ultra 5 338H has a base clock of 1.90 GHz and a boost clock of 4.70 GHz. The AMD chip also carries a higher TDP at 45 W versus 25 W for Intel, which explains part of its performance advantage in sustained workloads.

Memory support differs. The AMD chip uses DDR5 on a dual-channel bus with 89.6 GB/s bandwidth and supports ECC. The Intel chip uses LPDDR5X on a dual-channel bus with 136.5 GB/s bandwidth and does not support ECC. The Intel chip has higher raw bandwidth, but the AMD chip wins the memory-sensitive tests in the database.

PCIe support also diverges. The AMD chip offers Gen 4 with 20 lanes from the CPU. The Intel chip offers Gen 5 with 4 lanes from the CPU. The integrated graphics differ as well: the AMD chip uses the Radeon 780M, while the Intel chip uses the Arc B370. The AMD chip uses AMD Socket FP7, while the Intel chip uses Intel BGA 2540. The AMD chip has a release date of 2024-04-15, while the Intel chip has a release date of 2026-01-04.