AMD Ryzen 9 PRO 5945 vs Intel Core Ultra 5 235 Comparison

The AMD Ryzen 9 PRO 5945 and Intel Core Ultra 5 235 represent two very different philosophies for desktop computing, and the benchmark data reflects that split clearly. The Ryzen 9 PRO 5945 is a 12-core, 24-thread Zen 3 part designed for heavy parallel workloads, while the Core Ultra 5 235 is a 14-core, 14-thread Arrow Lake chip with a modern 3 nm process and a focus on single-thread and specialized instruction performance. The recorded data shows a near-total split: the AMD part dominates Cinebench, while the Intel part dominates most Passmark tests.

Head-to-Head Benchmarks

The most striking result in the entire comparison is in Cinebench. The Ryzen 9 PRO 5945 scores 2959 in Cinebench R15 multicore, which is 98.9% ahead of the Core Ultra 5 235's 1488. That same pattern repeats across every Cinebench version in the database. In R20 multicore, the AMD part scores 12333 versus 6202, again a 98.9% advantage. In R23 multicore, the gap is 29366 versus 14769, a 98.8% lead. The single-core Cinebench results are equally lopsided: R15 single-core shows 417 versus 210 (98.6% ahead), R20 single-core shows 1740 versus 875 (98.9% ahead), and R23 single-core shows 4145 versus 2085 (98.8% ahead). These are not small margins; the Ryzen 9 PRO 5945 essentially doubles the Intel chip's Cinebench output across the board.

The Passmark suite tells a very different story. The Core Ultra 5 235 wins the majority of those tests, often by substantial margins. In floating point math, the Intel part scores 117951 versus 70291, a 40.4% lead. In find prime numbers, it scores 371 versus 228, a 38.5% advantage. In physics, it scores 2570 versus 1772, a 31.1% lead. In single-thread performance, the Intel chip scores 4516 versus 3499, a 22.5% advantage. The Intel part also wins data encryption (29293 versus 26313, a 10.2% lead), extended instructions (32752 versus 27031, a 17.5% lead), random string sorting (48980 versus 43447, an 11.3% lead), and the Passmark multithread test (37816 versus 34549, an 8.6% lead).

The AMD chip does win a few Passmark tests, however. Its integer math score is 129768 versus 87948, a massive 47.6% lead. It also wins data compression with 416599 versus 390711, a 6.6% margin. The overall win count is close: the AMD processor takes 8 benchmark wins, while the Intel processor takes 9. The average benchmark scores reflect this near-parity: the Ryzen 9 PRO 5945 sits at 47527 with a 90th percentile ranking, while the Core Ultra 5 235 sits at 46062 with an 89th percentile ranking.

Looking at the nearest rivals in the database helps contextualize these numbers. The Ryzen 9 PRO 5945's average score of 47527 places it just 0.1% above the Intel Core Ultra X9 378H (47468) and 0.4% above the Intel Core i7-13700KF (47330). The Core Ultra 5 235's average of 46062 puts it 0.1% above the AMD Ryzen AI 9 HX 375 (46030) and 0.2% above the AMD EPYC 4364P (45970). Neither chip is an outlier; both sit in a dense pack of high-end desktop and mobile parts.

The Verdict

The data supports a clear recommendation for specific use cases, not a blanket winner. If the workload is Cinebench rendering, video encoding, or any task that scales heavily with thread count and raw core throughput, the Ryzen 9 PRO 5945 is the obvious choice. Its 98.9% lead in Cinebench R20 multicore and 98.8% lead in R23 multicore are not just wins, they are functional dominance. The AMD part's 24 threads versus the Intel part's 14 threads is the core reason for this, and the benchmark results show exactly what that thread count difference delivers in practice.

If the workload is single-threaded or relies on modern instruction sets, the Core Ultra 5 235 is the better pick. Its 22.5% lead in Passmark single-thread, 40.4% lead in floating point math, and 38.5% lead in find prime numbers show a processor that is significantly faster per thread. The Intel part also wins the Passmark multithread test despite having fewer threads, which suggests its per-core efficiency and higher clock speeds matter even in parallel workloads that are not purely Cinebench-style.

For a mixed-use desktop, the choice depends on which benchmark family you trust more. The Cinebench results are overwhelmingly AMD, and the Passmark results are mostly Intel. The database shows 8 wins for AMD and 9 wins for Intel, so a balanced workload would see both chips trading blows. Users who prioritize legacy single-thread performance or encryption and instruction-heavy tasks should lean Intel. Users who prioritize rendering and integer-heavy parallel work should lean AMD.

Architecture Differences

The two processors come from entirely different architectural eras. The Ryzen 9 PRO 5945 uses AMD's Zen 3 architecture under the codename Vermeer, built on a 7 nm process at TSMC. It has 12 cores and 24 threads, with a base clock of 3.00 GHz and a boost clock of 4.70 GHz. The Core Ultra 5 235 uses Intel's Arrow Lake architecture under the codename Arrow Lake-S, built on a 3 nm process, also at TSMC. It has 14 cores and 14 threads, with a base clock of 3.40 GHz and a boost clock of 5.00 GHz.

The transistor counts and die sizes show the process advantage of the Intel part. The Core Ultra 5 235 packs 17,800 million transistors into a 243 mm² die, while the Ryzen 9 PRO 5945 uses 8,300 million transistors across a dual-die design with each die at 74 mm². The smaller 3 nm node allows Intel to fit more transistors and run higher clocks, which explains the single-thread and floating point advantages.

Cache layouts are also fundamentally different. The Ryzen 9 PRO 5945 has 64 KB of L1 cache per core, 512 KB of L2 cache per core, and 64 MB of shared L3 cache. The Core Ultra 5 235 has 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3 cache. The AMD chip's much larger L3 pool (64 MB versus 24 MB) helps its Cinebench and integer math performance, while the Intel chip's larger per-core L1 and L2 caches support its per-thread speed.

Memory support differs as well. The Ryzen 9 PRO 5945 supports DDR4 with dual-channel memory and a bandwidth of 51.2 GB/s, and it supports ECC memory. The Core Ultra 5 235 supports DDR5 with dual-channel memory and a much higher bandwidth of 102.4 GB/s, but it does not support ECC. The Intel part also includes integrated graphics, specifically Arc Xe-LPG Graphics with 24 execution units, while the AMD part has no integrated graphics listed.

Specification Differences

The core and thread counts are the most obvious difference: the Ryzen 9 PRO 5945 has 12 cores and 24 threads, while the Core Ultra 5 235 has 14 cores and 14 threads. Clock speeds differ too, with the AMD part running at 3.00 GHz base and 4.70 GHz boost, and the Intel part running at 3.40 GHz base and 5.00 GHz boost. Both have a 65 W TDP.

The sockets are incompatible: the AMD chip uses Socket AM4, while the Intel chip uses Socket 1851. The process nodes differ (7 nm versus 3 nm), as do the transistor counts (8,300 million versus 17,800 million) and die sizes (2x 74 mm² versus 243 mm²). Memory support is DDR4 versus DDR5, with bandwidth figures of 51.2 GB/s versus 102.4 GB/s. ECC support is present on the AMD part and absent on the Intel part. PCIe generations differ: the AMD chip supports Gen 4 with 20 lanes, while the Intel chip supports Gen 5 with 20 lanes. The Intel part has integrated graphics; the AMD part does not. The Core Ultra 5 235 has a launch MSRP of $257. The AMD part was released in 2022, while the Intel part was released in 2025.

FAQ

Q: Which processor is faster in Cinebench R23?

A: The AMD Ryzen 9 PRO 5945 scores 29366 in R23 multicore versus 14769 for the Intel Core Ultra 5 235, a 98.8% lead. In R23 single-core, the AMD part scores 4145 versus 2085, also a 98.8% lead.

Q: Does the Intel Core Ultra 5 235 win any benchmarks?

A: Yes, it wins 9 of the 17 head-to-head benchmarks. It leads in Passmark single-thread (4516 versus 3499), floating point math (117951 versus 70291), physics (2570 versus 1772), find prime numbers (371 versus 228), data encryption, extended instructions, random string sorting, and the Passmark multithread test.

Q: Which chip has more threads?

A: The AMD Ryzen 9 PRO 5945 has 24 threads from 12 cores. The Intel Core Ultra 5 235 has 14 threads from 14 cores.

Q: What memory types do these processors support?

A: The AMD chip supports DDR4 with dual-channel memory and 51.2 GB/s bandwidth. The Intel chip supports DDR5 with dual-channel memory and 102.4 GB/s bandwidth. The AMD part also supports ECC memory; the Intel part does not.

Q: Which processor has a higher boost clock?

A: The Intel Core Ultra 5 235 boosts to 5.00 GHz, while the AMD Ryzen 9 PRO 5945 boosts to 4.70 GHz. The Intel part also has a higher base clock at 3.40 GHz versus 3.00 GHz.

Q: Do either of these chips include integrated graphics?

A: Only the Intel Core Ultra 5 235 includes integrated graphics, specifically Arc Xe-LPG Graphics with 24 execution units. The AMD Ryzen 9 PRO 5945 has no integrated graphics listed in the database.

Where Each One Wins

The Ryzen 9 PRO 5945 is the pick for Cinebench-style rendering and heavy parallel compute. Its 24 threads and 64 MB of L3 cache drive a 98.9% lead in R15 and R20 multicore, and a 98.8% lead in R23 multicore. It also wins integer math by a 47.6% margin (129768 versus 87948) and data compression by 6.6% (416599 versus 390711). Any workload that scales with thread count and large shared cache will favor the AMD part.

The Core Ultra 5 235 is the pick for single-threaded responsiveness and modern instruction-heavy tasks. Its 22.5% lead in Passmark single-thread (4516 versus 3499) and 40.4% lead in floating point math (117951 versus 70291) make it the stronger choice for everyday desktop use, scientific floating point workloads, and applications that use AVX-style instructions. Its 38.5% lead in find prime numbers (371 versus 228) and 31.1% lead in physics (2570 versus 1772) reinforce this pattern. The Intel part also wins the Passmark multithread test (37816 versus 34549), which means even some parallel workloads favor the higher-clocked Intel design.

The database shows a genuine split: 8 wins for AMD, 9 for Intel. The Ryzen 9 PRO 5945 is a rendering monster with a 90th percentile ranking, while the Core Ultra 5 235 is a per-thread powerhouse at the 89th percentile. Choose by workload, not by brand loyalty. The AMD chip is the rendering and integer compute champion; the Intel chip is the single-thread and floating point champion.