AMD Ryzen AI Max+ 388 vs AMD Ryzen AI Max 390 Comparison

Where Each One Wins

The benchmark split between these two Strix Halo parts is decisively one-sided, but the two wins recorded by the AMD Ryzen AI Max+ 388 land in a category that matters for everyday responsiveness. The Ryzen AI Max+ 388 takes the single-threaded PassMark tests, scoring 4185 in both passmark_single_thread and passmark_singlethread, against 4028 for the Ryzen AI Max 390. That is a 3.9% advantage in each case. The data suggests the Max+ 388 holds a small edge in lightly threaded workloads, the kind that dominate typical desktop interaction and legacy application performance.

Every other recorded benchmark goes to the AMD Ryzen AI Max 390. The 390 wins all three Cinebench tests where both parts have matching records, plus eight of the ten PassMark workloads. The margin is rarely small. In Cinebench R23 multi-core, the 390 posts 36064 against 18759, a 48% lead. The single-core R23 result is even more lopsided: 5091 versus 1960, a 61.5% gap. Those are enormous deltas for two processors from the same family, and they point to a fundamentally different positioning between the two SKUs.

The use-case split is therefore clear. The Max+ 388 is the choice for applications that are latency-sensitive and rely on single-thread speed, where its 3.9% PassMark single-thread win is the only advantage it holds. The Max 390 dominates threaded rendering, physics simulation, compression, encryption, and math workloads. For Cinebench R15 multi-core, the 390 leads by 21%, and for passmark_physics it leads by 33.2%. The 390 also wins passmark_find_prime_numbers by 54.1%, a workload that scales aggressively with core count and cache. There is no scenario in the recorded data where the Max+ 388 wins a multi-threaded test.

The overall average benchmark scores reinforce this. The Max 390 records an average benchmark score of 56273, compared to 49796 for the Max+ 388. That is a roughly 13% gap in aggregate performance. The percentile rankings differ by only one point, with the Max 390 at the 91st percentile of all CPUs and the Max+ 388 at the 90th, but the raw score difference is substantial. The Max 390 also sits in a faster peer group in the database: its nearest rivals include the AMD Ryzen AI 9 HX PRO 470 with an average score of 56306 and the Intel Core i9-14900HX at 56004, while the Max+ 388 is bracketed by the Intel Core 9 273PE at 49845 and the Intel Core i5-14600KF at 49394.

Architecture Differences

Both processors are built on Zen 5 architecture with the Strix Halo codename, and both are manufactured on a 4 nm process at TSMC. The die size is identical at 2x 70.6 mm², and both use AMD Socket FP11. The architectural split comes down to core count, cache allocation, and the integrated graphics solution.

The Ryzen AI Max+ 388 uses 8 cores and 16 threads. The Ryzen AI Max 390 uses 12 cores and 24 threads. That 50% increase in core and thread count is the primary driver of the multi-threaded benchmark gaps. Both parts share the same per-core cache structure, with 80 KB of L1 per core and 1 MB of L2 per core, but the shared L3 differs significantly. The Max+ 388 has 32 MB of shared L3 cache, while the Max 390 has 64 MB of shared L3 cache. Doubling the shared cache alongside a 50% core increase explains why the 390 scales so much better in cache-sensitive workloads like passmark_find_prime_numbers and passmark_random_string_sorting.

Clock speeds are close but not identical. The Max+ 388 has a base clock of 3.60 GHz and a boost clock of 5.00 GHz. The Max 390 has a base clock of 3.20 GHz and the same 5.00 GHz boost clock. The lower base clock on the 390 is offset by the additional cores and cache, and the equal boost clock means single-thread ceiling is theoretically identical, though the recorded single-thread PassMark scores favor the Max+ 388 by 3.9%.

The integrated graphics differ as well. The Max+ 388 carries the Radeon 8060S, while the Max 390 carries the Radeon 8050S. The database does not record separate GPU benchmark results for these parts, so the comparison is limited to naming and positioning. Both CPUs support LPDDR5X memory over a quad-channel bus with 256.0 GB/s of bandwidth, and both support ECC memory. PCIe is Gen 4 with 16 lanes on the CPU for both.

The release dates are one year apart. The Max 390 was released on 2025-01-05, and the Max+ 388 on 2026-01-05. Both are listed as Active in production status, and both have locked multipliers. The part numbers are 100-000001980 for the Max+ 388 and 100-000001423 for the Max 390.

Head-to-Head Benchmarks

The largest single benchmark gap in the entire comparison is Cinebench R23 single-core, where the Max 390 leads by 61.5%. The Max+ 388 records 1960 and the Max 390 records 5091. That is a massive delta for a single-core test, especially given that both CPUs share a 5.00 GHz boost clock and the same Zen 5 architecture. The result suggests that the database sample for the Max+ 388 may have been constrained, or that the Max 390 sustains its boost under this particular workload far better. Regardless, the recorded data is unambiguous: the 390 wins this test by a wide margin.

Cinebench R23 multi-core shows a 48% lead for the Max 390, with 36064 versus 18759. Cinebench R15 multi-core follows the same pattern, with the 390 at 3635 and the Max+ 388 at 2872, a 21% gap. Cinebench R15 single-core is even more severe at 41.9%, with 513 versus 298. The Max 390 also records wins in Cinebench R20 multi-core and single-core, though the Max+ 388 has no matching R20 entries in the database, so those results cannot be compared directly.

In PassMark workloads, the Max 390 wins passmark_data_compression with 487145 versus 400887, a 17.7% lead. Data encryption goes to the 390 by 19.9%, with 25097 versus 20092. Extended instructions favor the 390 by 15.5%, and floating point math by 19.7%. The integer math gap is 25.2%, with the 390 at 146519 and the Max+ 388 at 109588. PassMark multi-thread shows a 19.8% lead for the 390, and physics shows a 33.2% lead. Random string sorting goes to the 390 by 18.7%, and find prime numbers by 54.1%.

The only wins for the Max+ 388 are the two identical single-thread PassMark tests, both with a 3.9% margin. The Max+ 388 also has no recorded Cinebench R20 results, so it cannot be compared on that workload.

Specification Differences

The two processors share their socket, architecture, process node, foundry, die size, memory type, memory bus width, memory bandwidth, ECC support, PCIe configuration, TDP, and boost clock. The differences are concentrated in core count, cache, base clock, integrated graphics, release date, and part number.

The Ryzen AI Max+ 388 has 8 cores and 16 threads, while the Ryzen AI Max 390 has 12 cores and 24 threads. The Max+ 388 has a base clock of 3.60 GHz, the Max 390 a base clock of 3.20 GHz, and both boost to 5.00 GHz. L3 cache is 32 MB shared on the Max+ 388 and 64 MB shared on the Max 390. Per-core L1 and L2 are identical at 80 KB and 1 MB respectively.

The integrated graphics differ: Radeon 8060S on the Max+ 388, Radeon 8050S on the Max 390. The Max 390 was released on 2025-01-05, the Max+ 388 on 2026-01-05. Both are Active and both have locked multipliers. The part numbers are 100-000001980 for the Max+ 388 and 100-000001423 for the Max 390. Neither part has a recorded launch MSRP in the database.

FAQ

Q: Which processor has more cores?

A: The AMD Ryzen AI Max 390 has 12 cores and 24 threads, while the AMD Ryzen AI Max+ 388 has 8 cores and 16 threads.

Q: Which processor wins in single-threaded PassMark tests?

A: The AMD Ryzen AI Max+ 388 wins both passmark_single_thread and passmark_singlethread with a score of 4185, against 4028 for the AMD Ryzen AI Max 390, a 3.9% margin.

Q: How large is the Cinebench R23 multi-core gap?

A: The AMD Ryzen AI Max 390 leads by 48%, scoring 36064 versus 18759 for the AMD Ryzen AI Max+ 388.

Q: Do both processors use the same amount of L3 cache?

A: No. The AMD Ryzen AI Max+ 388 has 32 MB of shared L3 cache, while the AMD Ryzen AI Max 390 has 64 MB of shared L3 cache.

Q: What is the TDP of both processors?

A: Both the AMD Ryzen AI Max+ 388 and the AMD Ryzen AI Max 390 have a TDP of 55 watts.

Q: Which processor has a higher average benchmark score?

A: The AMD Ryzen AI Max 390 has an average benchmark score of 56273, compared to 49796 for the AMD Ryzen AI Max+ 388.