AMD Ryzen AI Max PRO 390 vs Intel Core i9-13900 Comparison

Head-to-Head Benchmarks

The benchmark comparison between the AMD Ryzen AI Max PRO 390 and the Intel Core i9-13900 shows a clear split in workload types. Out of 15 head-to-head tests, the Intel processor wins 11, while the AMD chip takes 4. The overall average benchmark scores reflect this: the AMD Ryzen AI Max PRO 390 sits at 63765, while the Intel Core i9-13900 is lower at 60676, though that gap is not the whole story.

The Intel Core i9-13900 dominates in multi-threaded rendering tasks. In Cinebench R23 multi-core, it scores 37931 against the AMD chip's 24828, a 34.5% advantage. The Cinebench R15 multi-core test is closer, with Intel scoring 3823 versus AMD's 3918, giving AMD a 2.5% win there. But the R23 result is the more telling one for sustained all-core workloads. The single-core results are even more lopsided. Intel's Cinebench R23 single-core score of 5355 crushes AMD's 1976, a 63.1% deficit. In Cinebench R15 single-core, Intel wins 539 to 303, a 43.8% margin. These are massive differences in per-thread performance, and they show up across most real-world tasks.

In PassMark tests, Intel takes data compression (577285 versus 506170, a 12.3% lead), data encryption (35242 versus 26027, 26.1% ahead), floating point math (120492 versus 94666, 21.4% ahead), integer math (176107 versus 148508, 15.7% ahead), multithread (45680 versus 42912, 6.1% ahead), random string sorting (64396 versus 55248, 14.2% ahead), and single-thread tests (4309 versus 3967, 7.9% ahead). The Intel part is broadly faster in these common compute workloads.

The AMD Ryzen AI Max PRO 390 has its own strengths. Its biggest win is in PassMark find prime numbers, where it scores 323 against Intel's 186, a 73.7% advantage. That is not a typo: the AMD chip is nearly three-quarters faster at this specific math operation. It also wins PassMark extended instructions (40888 versus 32760, a 24.8% lead), which points to strength in specialized instruction sets. In PassMark physics, AMD scores 2773 versus Intel's 2484, an 11.6% win. And in Cinebench R15 multi-core, AMD's 3918 edges out Intel's 3823 by 2.5%, as noted.

The delta percentages matter here. Intel's wins are often large (34.5%, 63.1%, 26.1%), while AMD's wins are either narrow (2.5% in Cinebench R15 multi-core) or highly specialized (73.7% in prime numbers, 24.8% in extended instructions). The database's percentile rankings place the AMD part at 93rd percentile versus 92nd for Intel, so both are near the top of the overall CPU pool. But the workload distribution is the deciding factor for most users.

FAQ

Q: Which processor is faster in Cinebench R23 multi-core?

A: The Intel Core i9-13900 is significantly faster, scoring 37931 versus 24828 for the AMD Ryzen AI Max PRO 390, a 34.5% lead.

Q: Does the AMD chip beat Intel in any single-core test?

A: No. In every single-core benchmark recorded, Intel wins. Cinebench R23 single-core shows Intel at 5355 versus AMD's 1976 (63.1% ahead), and PassMark single-thread shows Intel at 4309 versus AMD's 3967 (7.9% ahead).

Q: What is the AMD processor's largest benchmark win?

A: The biggest margin is in PassMark find prime numbers, where AMD scores 323 versus Intel's 186, a 73.7% advantage. It also wins PassMark extended instructions by 24.8% and PassMark physics by 11.6%.

Q: How do the average benchmark scores compare?

A: The AMD Ryzen AI Max PRO 390 has an average benchmark score of 63765, which is higher than the Intel Core i9-13900's 60676. However, Intel's nearest rivals list shows the Core i9-13900 is close to other high-end parts like the AMD Ryzen 9 7945HX (1% delta) and Intel Core i9-14900F (1.1% delta).

Q: Which processor has more cores and threads?

A: The Intel Core i9-13900 has 24 cores and 32 threads. The AMD Ryzen AI Max PRO 390 has 12 cores and 24 threads.

Q: Is the AMD chip competitive in modern multi-threaded workloads?

A: It depends on the workload. It wins Cinebench R15 multi-core by 2.5%, but loses Cinebench R23 multi-core by 34.5%. It also loses PassMark multithread by 6.1%. The data shows it is not consistently competitive in heavy multi-threaded rendering, though it does hold its own in specific tasks like physics and extended instructions.

Architecture Differences

The two processors come from different design philosophies. The AMD Ryzen AI Max PRO 390 is built on the Zen 5 architecture with the codename Strix Halo, fabricated on a 4nm process by TSMC. It uses a chiplet-style design with 12 cores and 24 threads, and it is aimed at the mobile segment with a 55W TDP. The Intel Core i9-13900 is based on Raptor Lake, codename Raptor Lake-S, built on Intel's 10nm process (a node name that does not directly correspond to transistor size in the same way as TSMC's naming). It uses a hybrid architecture of performance and efficiency cores, totaling 24 cores and 32 threads, and it is a desktop part with a 65W TDP.

Cache configurations differ notably. Both have 80 KB of L1 cache per core. For L2, AMD provides 1 MB per core, while Intel provides 2 MB per core. The L3 cache is a major divergence: AMD has 64 MB shared, while Intel has 36 MB shared. That gives AMD nearly double the L3 capacity, which can help in cache-sensitive workloads, though the benchmark data shows Intel still wins in most tests.

Memory support is another differentiator. AMD supports LPDDR5X memory with a quad-channel bus and a recorded memory bandwidth of 256.0 GB/s. Intel supports both DDR4 and DDR5, but only on a dual-channel bus, and the database does not list a memory bandwidth figure for it. Both support ECC memory. PCIe connectivity differs: AMD provides Gen 4 with 16 lanes (CPU only), while Intel provides Gen 5 with 16 lanes (CPU only). Intel's newer PCIe generation allows faster interconnect speeds for GPUs and storage, though the AMD part's memory bandwidth advantage is a counterpoint.

Integrated graphics are present on both. AMD ships a Radeon 8050S, while Intel ships UHD Graphics 770. Neither is a discrete-class solution, but the database shows both parts are active in production. The AMD part is a mobile processor on Socket FP11, while Intel is a desktop processor on Socket 1700. The release dates are about two years apart: AMD launched on January 5, 2025, while Intel launched on January 3, 2023. The Intel part has a launch MSRP of $549; the AMD part has no listed launch price.

Specification Differences

The core and thread counts are a clear differentiator: AMD has 12 cores and 24 threads, Intel has 24 cores and 32 threads. Base clocks differ wildly: AMD's base clock is 3.20 GHz, while Intel's is listed as 2000.00 MHz (2.00 GHz). Boost clocks are closer: AMD boosts to 5.00 GHz, Intel boosts to 5.60 GHz, giving Intel a 0.6 GHz advantage. TDP also differs: AMD is 55W, Intel is 65W. The socket is different: AMD uses Socket FP11, Intel uses Socket 1700. The process nodes are different: AMD is 4nm TSMC, Intel is 10nm Intel. The die size is only recorded for Intel at 257 mm², with no figure for AMD. L2 cache per core is 1 MB for AMD versus 2 MB for Intel. L3 cache is 64 MB shared for AMD versus 36 MB shared for Intel. Memory support: AMD uses LPDDR5X only, Intel uses DDR4 and DDR5. Memory bus: AMD is quad-channel, Intel is dual-channel. Memory bandwidth: AMD is 256.0 GB/s, Intel has no recorded figure. PCIe: AMD is Gen 4 with 16 lanes, Intel is Gen 5 with 16 lanes. Integrated graphics: AMD has Radeon 8050S, Intel has UHD Graphics 770. Market segment: AMD is mobile, Intel is desktop. Release date: AMD is January 5, 2025, Intel is January 3, 2023. Launch MSRP: Intel is $549, AMD has none listed. Part numbers differ: AMD is 100-000001421, Intel is SRMB6.

The Verdict

The benchmark data points to a clear choice for most users: the Intel Core i9-13900 is the stronger processor for general and multi-threaded performance. It wins 11 of 15 head-to-head tests, including every single-core test, every major rendering test except one, and most PassMark compute workloads. Its 34.5% lead in Cinebench R23 multi-core and 63.1% lead in Cinebench R23 single-core are decisive for productivity and creative work. The Intel part also has a higher boost clock (5.60 GHz versus 5.00 GHz), more cores (24 versus 12), more threads (32 versus 24), and support for PCIe Gen 5. If you are building a desktop and need maximum throughput in common applications, the data says Intel wins.

The AMD Ryzen AI Max PRO 390 is not without merits. It has a higher average benchmark score (63765 versus 60676) and a slightly better percentile ranking (93rd versus 92nd). Its 64 MB of L3 cache is nearly double Intel's 36 MB. It wins PassMark find prime numbers by 73.7%, extended instructions by 24.8%, physics by 11.6%, and Cinebench R15 multi-core by 2.5%. It also has a lower TDP (55W versus 65W) and higher memory bandwidth (256.0 GB/s on quad-channel LPDDR5X). For a mobile platform, that combination is meaningful. But the data does not show it beating Intel in the workloads most users care about, such as rendering, compression, encryption, and floating-point math.

The verdict is workload-dependent but leans heavily Intel. If you prioritize raw compute in standard desktop scenarios, the Core i9-13900 is the choice. If you need a mobile part with strong memory bandwidth, a large L3 cache, and specific strengths in prime-number math and extended instructions, the Ryzen AI Max PRO 390 is the only option between the two, since Intel's part is desktop-only.

Where Each One Wins

Pick the Intel Core i9-13900 for multi-threaded rendering and productivity. It wins Cinebench R23 multi-core by 34.5%, Cinebench R23 single-core by 63.1%, and Cinebench R15 single-core by 43.8%. It also wins PassMark data compression (12.3% ahead), data encryption (26.1% ahead), floating point math (21.4% ahead), integer math (15.7% ahead), multithread (6.1% ahead), random string sorting (14.2% ahead), and single-thread (7.9% ahead). This covers most content creation, scientific computing, and office workloads. The Intel part also offers PCIe Gen 5 connectivity, which matters for modern GPUs and NVMe storage.

Pick the AMD Ryzen AI Max PRO 390 for specific compute patterns. It wins PassMark find prime numbers by a massive 73.7%, which indicates a strong integer-prime calculation pipeline. It wins PassMark extended instructions by 24.8%, meaning it handles specialized instruction sets (like AVX-512 or similar) more efficiently. It wins PassMark physics by 11.6%, which can help in simulation or physics-based workloads. It also edges out Intel in Cinebench R15 multi-core by 2.5%, though that is a narrow margin. The AMD chip's 256.0 GB/s memory bandwidth and 64 MB L3 cache make it a better fit for memory-bound tasks, even if the benchmark data does not show a consistent overall win.

For mobile users, the AMD part is the only realistic choice because it is a mobile processor on Socket FP11, while Intel is a desktop processor on Socket 1700. For desktop users, the Intel part has more cores, more threads, a higher boost clock, and wins the majority of benchmarks. The data does not support any other conclusion.