AMD Ryzen AI 9 365 vs Intel Core Ultra 5 225F Comparison

Head-to-Head Benchmarks

The recorded data shows a clear split between the AMD Ryzen AI 9 365 and the Intel Core Ultra 5 225F across the benchmark suite, with Intel taking 8 wins and AMD taking 7. The margin of victory varies widely by workload, revealing distinct strengths rather than a single dominant performer.

The biggest win for AMD comes in PassMark integer math, where the Ryzen AI 9 365 scores 101,831 against Intel's 66,417, a 53.3% advantage. This is the largest delta in the entire comparison. The Ryzen also leads decisively in Cinebench R23 multi-core, scoring 18,698 versus 16,467, a 13.5% edge. Data compression follows the same pattern, with AMD at 354,510 and Intel at 310,843, a 14% difference. In Cinebench R15 multi-core, AMD leads 2,842 to 2,660, a 6.8% gap, and in R15 single-core the margin is 303 versus 287, or 5.6%. Random string sorting also favors AMD, 39,447 to 37,325, a 5.7% advantage.

Intel's largest win is in PassMark find prime numbers, where it scores 352 against AMD's 117, a massive 66.8% lead. Floating point math also strongly favors Intel: 92,554 versus 62,802, a 32.1% margin. Physics tests show Intel ahead 2,430 to 1,704, a 29.9% gap. Data encryption goes to Intel by 22,648 versus 18,297, a 19.2% lead. Extended instructions favor Intel at 28,027 versus 25,113, a 10.4% difference. Single-thread performance, measured in both PassMark single-thread and singlethread tests, shows Intel at 4,397 versus AMD's 3,841, a 12.6% gap. PassMark multithread also goes to Intel, 31,004 versus 29,467, a 5% difference.

The Cinebench R23 single-core result is closer, with AMD at 1,992 and Intel at 1,893, a 5.2% gap in favor of AMD. Overall, the benchmark data indicates that AMD dominates in integer-heavy and multi-core rendering workloads, while Intel counters with advantages in floating-point, prime-number calculation, encryption, and single-thread tests.

Where Each One Wins

The use-case split follows the benchmark patterns closely. AMD's Ryzen AI 9 365 is the stronger choice for workloads that rely on integer throughput and multi-core rendering. The 53.3% lead in PassMark integer math and the 13.5% lead in Cinebench R23 multi-core position it well for general productivity, code compilation, and rendering tasks that scale across threads. Data compression at 14% ahead reinforces this, as does the 5.7% advantage in random string sorting, which is a common pattern in database and text-processing workloads.

Intel's Core Ultra 5 225F takes the other side. The 66.8% lead in find prime numbers points to strength in algorithms that depend on branch prediction and single-thread latency. Floating point math at 32.1% ahead suggests an edge in scientific computing, simulations, and 3D rendering tasks that rely on FPU throughput. The physics test result, 29.9% ahead, aligns with that interpretation. Data encryption at 19.2% ahead indicates better throughput for cryptographic operations, and the extended instructions lead of 10.4% covers workloads using newer instruction set extensions. The 12.6% single-thread advantage means Intel holds the edge in lightly threaded applications where per-core speed matters most.

The multithread result is worth examining closely. Intel wins PassMark multithread by 5%, scoring 31,004 versus 29,467, even though AMD wins the Cinebench multi-core tests by 6.8% and 13.5%. This divergence reflects the different workload compositions within each test. Cinebench R23 is a pure rendering workload that scales well with threads, while PassMark multithread mixes several sub-tests, some of which favor Intel's architecture. The data shows that AMD's 10-core, 20-thread configuration delivers a stronger result in sustained all-core rendering, while Intel's 10-core, 10-thread design wins in the aggregated PassMark multithread suite.

The Verdict

The benchmark data indicates that the AMD Ryzen AI 9 365 is the preferred processor for multi-threaded rendering and integer-intensive productivity. Its Cinebench R23 multi-core score of 18,698 versus 16,467, a 13.5% advantage, is a meaningful gap for users running CPU-based rendering or video encoding. The 53.3% integer math lead and 14% data compression lead reinforce this profile. AMD also holds the edge in both Cinebench single-core tests, though by smaller margins of 5.6% and 5.2%.

The Intel Core Ultra 5 225F is the better fit for workloads that stress floating-point arithmetic, cryptographic operations, and single-thread responsiveness. The 32.1% floating point lead, 66.8% prime-number lead, and 19.2% encryption lead are substantial. The 12.6% single-thread advantage in PassMark also matters for applications that cannot use more than a few cores. The overall benchmark average tells a similar story: AMD's average score is 40,048, placing it in the 87th percentile of all CPUs, while Intel's average is 37,313, in the 85th percentile. The nearest rivals in the database for AMD include the AMD Ryzen 7 7700 with an average score of 40,081 and a delta of -0.1%, while Intel's nearest rival is the Intel Core i9-13900HK at 37,425 with a delta of -0.3%.

Choice depends on workload. The rendering and integer-heavy user should select the AMD part. The floating-point and single-thread user should select the Intel part. Neither processor dominates the other across the full suite; the 7-to-8 win split in the head-to-head benchmarks confirms a genuine trade-off.

FAQ

Q: Which processor has the higher Cinebench R23 multi-core score?

A: The AMD Ryzen AI 9 365 scores 18,698, which is 13.5% higher than the Intel Core Ultra 5 225F's 16,467.

Q: How large is Intel's lead in floating-point math?

A: Intel scores 92,554 in PassMark floating point math against AMD's 62,802, a 32.1% advantage.

Q: Does AMD win any single-core benchmarks?

A: Yes. AMD wins Cinebench R15 single-core with 303 versus 287, a 5.6% margin, and Cinebench R23 single-core with 1,992 versus 1,893, a 5.2% margin. Intel wins PassMark single-thread with 4,397 versus 3,841, a 12.6% margin.

Q: What is the largest single benchmark delta between the two?

A: The largest delta is in PassMark find prime numbers, where Intel scores 352 against AMD's 117, a 66.8% lead.

Q: How do the overall benchmark averages compare?

A: AMD's average benchmark score is 40,048, placing it in the 87th percentile of all CPUs. Intel's average is 37,313, placing it in the 85th percentile.

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

A: Intel wins 8 of the 15 head-to-head tests, while AMD wins 7.

Architecture Differences

The two processors come from different architectural generations and design philosophies. AMD uses the Zen 5 architecture with the codename Strix Point, part of the Ryzen AI 300 generation that combines Zen 5 and Zen 5c cores. Intel uses the Arrow Lake architecture with the codename Arrow Lake-S, part of the Core Ultra Series 2.

Both processors have 10 cores, but the thread counts differ. AMD supports 20 threads through simultaneous multithreading, while Intel's Core Ultra 5 225F supports 10 threads, indicating no hyper-threading on this part. This explains why AMD wins the multi-threaded Cinebench tests despite identical core counts.

The process nodes differ. AMD is built on a 4 nm process at TSMC, while Intel is built on a 3 nm process, also at TSMC. Die sizes are close: AMD measures 233 mm², Intel measures 243 mm². Intel reports a transistor count of 17,800 million; AMD does not report a transistor count in the database.

Cache hierarchies differ substantially. AMD allocates 80 KB of L1 cache per core and 1 MB of L2 per core, with 16 MB of L3 cache. Intel allocates 192 KB of L1 per core and 3 MB of L2 per core, with 20 MB of shared L3. The larger per-core L1 and L2 on Intel may contribute to its single-thread and floating-point advantages, while the total cache distribution favors Intel in L3 capacity.

Integrated graphics also separate the two. AMD includes the Radeon 880M iGPU, while Intel has no integrated graphics, listed as N/A. This is consistent with their market segments: AMD is a mobile processor on the AMD Socket FP8, while Intel is a desktop processor on the Intel Socket 1851.

Specification Differences

The processors differ across nearly every specification field. AMD's base clock is 2.00 GHz with a boost clock of 5.00 GHz. Intel's base clock is 3.30 GHz with a boost clock of 4.90 GHz. Intel starts from a higher base frequency but boosts slightly lower.

Thermal design power differs significantly. AMD is rated at 28 W, while Intel is rated at 65 W. This reflects the mobile versus desktop split. AMD's TDP is less than half of Intel's, which is notable given that AMD still wins several multi-threaded benchmarks.

Memory support differs. AMD supports DDR5 and LPDDR5X with dual-channel memory and a measured bandwidth of 89.6 GB/s. Intel supports DDR5 only, also dual-channel, with a higher measured bandwidth of 102.4 GB/s. Neither supports ECC memory.

PCIe capabilities differ by generation and lane count. AMD uses Gen 4 with 16 CPU-only lanes. Intel uses Gen 5 with 20 CPU-only lanes. This gives Intel a substantial advantage in available PCIe bandwidth and lane count for desktop expansion.

The release dates differ. AMD was released on 2024-06-30, while Intel was released on 2025-01-06. Intel's launch MSRP is $231; AMD does not have a recorded launch MSRP. Neither processor has an unlocked multiplier. The part numbers are 100-000001530 for AMD and SRQD2SRVF9 for Intel. Both are listed as Active in production status.