AMD Ryzen AI Max PRO 390 vs Intel Core Ultra 9 285 Comparison

The AMD Ryzen AI Max PRO 390 and the Intel Core Ultra 9 285 are designed for different computing environments, and the benchmark data reflects that split decisively. The Intel part, a desktop processor, dominates the AMD mobile chip in every single recorded head-to-head test. The data shows a 15-0 win record for the Intel Core Ultra 9 285, but the margins vary significantly by workload, offering insight into where each architecture excels relative to its category.

Head-to-Head Benchmarks

The most dramatic differences appear in single-core performance. In Cinebench R23 single-core, the Intel Core Ultra 9 285 scores 6909 against the AMD's 1976, a 71.4% advantage. The Cinebench R15 single-core test shows a similar story, with Intel winning 696 to 303, a 56.5% gap. These results indicate a substantial difference in per-thread capability, which is a defining characteristic of the desktop part.

Multi-core performance also heavily favors Intel, but the margin is slightly less extreme. In Cinebench R23 multi-core, the Intel chip scores 48945 versus the AMD's 24828, a 49.3% lead. The Cinebench R15 multi-core test shows Intel ahead with 4933 against 3918, a 20.6% difference. This suggests that while the AMD chip's 12 cores are efficient, they cannot match the raw throughput of Intel's 24 cores in sustained all-core workloads.

In floating-point math, the Intel chip shows a commanding lead, scoring 194988 compared to the AMD's 94666, a 51.5% advantage. This is a critical metric for scientific and analytical workloads. The data encryption test also shows a significant Intel advantage, with a score of 46949 versus 26027, a 44.6% margin. These results confirm that the Intel processor is better suited for compute-heavy tasks that rely on parallel execution units.

The gap narrows considerably in some integer and data-processing tasks. In PassMark integer math, Intel wins 164869 to 148508, a 9.9% margin. The extended instructions test shows the same 9.9% delta, with Intel scoring 45357 against the AMD's 40888. Data compression is also a closer contest, with Intel winning 602121 to 506170, a 15.9% lead. These results indicate that the AMD's Zen 5 architecture is comparatively strong in general-purpose integer work, even when outmatched in raw core count.

The remaining tests follow the same pattern. PassMark multi-thread shows Intel ahead 56602 to 42912, a 24.2% margin. Random string sorting favors Intel, 73651 to 55248, a 25% difference. Physics simulation also goes to Intel, 3598 to 2773, a 22.9% lead. Prime number finding shows Intel ahead at 459 versus 323, a 29.6% margin. The single-thread PassMark test rounds out the results with Intel winning 4881 to 3967, an 18.7% advantage.

FAQ

Q: Which processor wins in Cinebench R23 multi-core?

A: The Intel Core Ultra 9 285 wins with a score of 48945, which is 49.3% higher than the AMD Ryzen AI Max PRO 390's score of 24828.

Q: How large is the single-core performance gap?

A: The largest gap is in Cinebench R23 single-core, where the Intel chip leads by 71.4%, scoring 6909 against the AMD's 1976.

Q: Where is the performance difference the smallest?

A: The smallest margins are in PassMark integer math and extended instructions, where the Intel chip leads by 9.9% in both tests, with scores of 164869 and 45357 respectively.

Q: Does the AMD processor win any benchmark in the head-to-head comparison?

A: No, the recorded data shows the Intel Core Ultra 9 285 winning all 15 head-to-head benchmark comparisons.

Q: What is the difference in average benchmark scores?

A: The Intel processor has an average benchmark score of 75488, while the AMD processor has an average of 63765. This places the Intel chip in the 95th percentile of all CPUs, compared to the 93rd percentile for the AMD.

Q: How does the Intel processor compare to its nearest rivals?

A: The Intel Core Ultra 9 285 is essentially tied with the AMD EPYC 8224P, showing a delta of -0.1%. It performs similarly to the AMD Ryzen 7 PRO 9755X3D, with a delta of -0.3%.

Architecture Differences

The two processors are built on different architectural foundations. The AMD Ryzen AI Max PRO 390 uses the Zen 5 architecture, codenamed Strix Halo, and is manufactured on a 4 nm process by TSMC. The Intel Core Ultra 9 285 uses the Arrow Lake architecture, part of the Core Ultra Series 2, and is manufactured on a smaller 3 nm process, also by TSMC.

The core configurations differ fundamentally. The AMD chip has 12 cores and 24 threads, while the Intel chip has 24 cores and also 24 threads. This indicates that Intel is not using simultaneous multithreading (SMT) in this design, instead relying on physical cores for its thread count, whereas AMD uses its SMT implementation to double its thread count from 12 to 24.

Cache hierarchies are also distinct. The AMD chip has 80 KB of L1 cache per core, 1 MB of L2 per core, and a large 64 MB shared L3 cache. The Intel chip has 192 KB of L1 per core, 3 MB of L2 per core, but a smaller 36 MB shared L3 cache. The larger L3 on the AMD part suggests a focus on data locality and latency reduction, while Intel's larger per-core L2 may better serve its higher core count.

The Intel chip's transistor count is recorded at 17,800 million on a 243 mm² die, while the AMD chip's transistor count and die size are not listed. The memory controllers also differ, with AMD supporting LPDDR5X memory over a quad-channel bus with a bandwidth of 256.0 GB/s, while Intel supports DDR5 over a dual-channel bus with a bandwidth of 102.4 GB/s. Both support ECC memory. The PCIe interfaces differ as well, with AMD using Gen 4 with 16 lanes and Intel using Gen 5 with 20 lanes.

Specification Differences

The specification sheets show clear differences in physical and electrical characteristics. The AMD Ryzen AI Max PRO 390 has a base clock of 3.20 GHz and a boost clock of 5.00 GHz, while the Intel Core Ultra 9 285 has a lower base clock of 2.50 GHz but a higher boost clock of 5.60 GHz. The AMD chip has a TDP of 55 watts, versus the Intel chip's 65 watts.

The sockets are incompatible: AMD uses Socket FP11, while Intel uses Socket 1851. The market segment differs significantly, with the AMD chip targeting mobile devices and the Intel chip targeting desktops. The AMD chip's integrated graphics are the Radeon 8050S, while the Intel chip uses Arc Xe-LPG Graphics with 64 execution units. The release dates are close, with AMD launching on January 5, 2025, and Intel on December 31, 2024. The Intel chip has a launch MSRP of $579. Both processors have locked multipliers and are currently in active production.

The Verdict

The benchmark data presents a clear picture: the Intel Core Ultra 9 285 is the superior performer in every recorded test. Its average benchmark score of 75488 places it in the 95th percentile of all CPUs, while the AMD Ryzen AI Max PRO 390's score of 63765 places it in the 93rd percentile. The Intel chip is also competitive within its nearest rival group, showing deltas of -0.1% against the AMD EPYC 8224P and -0.3% against the AMD Ryzen 7 PRO 9755X3D.

The AMD chip, however, is not without merit. Its 55-watt TDP and mobile segment designation suggest it is engineered for power efficiency within a laptop chassis, where the Intel chip's 65-watt TDP and desktop socket would not apply. The AMD chip's quad-channel LPDDR5X memory support provides a memory bandwidth of 256.0 GB/s, which is more than double the Intel chip's 102.4 GB/s. This specific advantage could be relevant for integrated graphics performance or memory-intensive tasks, though it does not translate into a win in the recorded benchmarks.

Where Each One Wins

The Intel Core Ultra 9 285 wins in all scenarios that involve raw processing power. It is the clear choice for multi-core rendering, as shown by its 49.3% lead in Cinebench R23 multi-core and 20.6% lead in R15 multi-core. It also wins decisively in floating-point math, with a 51.5% advantage, making it the better option for simulation and scientific computing. Its 44.6% lead in data encryption and 29.6% lead in prime number finding further solidify its position for cryptography and number theory workloads.

The AMD Ryzen AI Max PRO 390 does not have a single benchmark win, but its relative performance in integer math and data compression, where the margin was only 9.9% and 15.9% respectively, indicates that its Zen 5 cores are efficient for general-purpose code. Its lower TDP of 55 watts and mobile form factor make it the more suitable option for a thin-and-light laptop where sustained performance and battery life are priorities. The Intel chip, with its higher boost clock of 5.60 GHz and 24 physical cores, is the better choice for a desktop workstation where power draw is less of a concern and maximum throughput is the goal.