AMD Ryzen AI Max+ 392 vs Intel Core Ultra 7 265T Comparison

Head-to-Head Benchmarks

The recorded data shows a split decision between the AMD Ryzen AI Max+ 392 and the Intel Core Ultra 7 265T, with AMD taking 6 of 11 head-to-head comparisons and Intel taking 5. The margin of victory, however, is heavily skewed toward AMD. The largest single win belongs to the AMD part in the extended instructions test, where it scores 45666 against Intel's 28609, a 59.6% advantage. Data compression is nearly as lopsided, with AMD at 554760 versus 370158, a 49.9% lead. Integer math also favors AMD decisively, 152414 against 104943, a 45.2% edge. These three workloads demonstrate the Zen 5 architecture's strength in compute-heavy, instruction-diverse tasks.

Random string sorting is another clear AMD win, 59487 versus 44400, a 34% gap. The multithread score shows AMD ahead at 45231 versus 37084, a 22% lead, and physics simulation follows the same pattern, 2887 versus 2391, a 20.7% edge. These results indicate that the Ryzen AI Max+ 392 sustains its advantage across both integer-heavy and memory-latency-sensitive workloads.

Intel's wins are narrower but still significant. The floating point math test is its best result, with 129817 against AMD's 99548, a 23.3% improvement. Data encryption goes to Intel by a smaller margin, 29687 versus 27784, a 6.4% lead. Single-thread performance also favors Intel, 4338 versus 3927, a 9.5% advantage, and the find prime numbers test is essentially a tie, with Intel at 322 and AMD at 320, a 0.6% difference. The single-thread and singlethread entries are identical, confirming the same benchmark result under different naming conventions.

The average benchmark scores reinforce this split. AMD's average is 90541, placing it in the 96th percentile of all CPUs, while Intel's average is 47697, placing it in the 90th percentile. The nearest rivals for AMD are the Intel Xeon 654 at 90717 (0.2% below AMD), the AMD Ryzen 9 9955HX at 91199 (0.7% below), the Intel Xeon w7-2575X at 88172 (2.7% above), and the Intel Xeon 6736P at 87864 (3% above). Intel's nearest rivals include the AMD Ryzen 9 PRO 5945 at 47527 (0.4% above), the AMD Ryzen 9 7900X3D at 47908 (0.4% below), the AMD Ryzen 9 3900 at 47918 (0.5% below), and the Intel Core Ultra X9 378H at 47468 (0.5% above). The delta percentages show that AMD's average sits near the top of its comparison group, while Intel's average is tightly clustered with its rivals.

Where Each One Wins

The AMD Ryzen AI Max+ 392 dominates in workloads that stress the processor with complex instruction streams and parallel integer operations. The extended instructions score of 45666, which is 59.6% higher than Intel's, indicates a substantial advantage in applications that use AVX-512 or similar vector extensions. Data compression, with a 49.9% lead, benefits from the large 64 MB shared L3 cache, which reduces memory traffic during compression and decompression tasks. Integer math, at 45.2% ahead, shows the efficiency of the 12 Zen 5 cores with simultaneous multithreading, delivering 24 threads for parallel integer workloads. Random string sorting, a 34% win, also benefits from the large cache and high memory bandwidth of 256.0 GB/s from the quad-channel LPDDR5X interface.

Multithread performance, 22% ahead, and physics simulation, 20.7% ahead, further establish AMD's strength in heavily threaded scenarios. The 24 threads versus Intel's 20 threads, combined with the 64 MB L3 cache, give the Ryzen AI Max+ 392 a clear edge in rendering, scientific computing, and simulation tasks where memory access patterns are irregular.

The Intel Core Ultra 7 265T wins in floating point math, with a 23.3% advantage. This result is notable because the 20 Arrow Lake cores, despite lacking hyperthreading, deliver higher per-core floating point throughput. Data encryption, a 6.4% win, may reflect Intel's dedicated AES-NI implementation and its efficiency in cryptographic workloads. Single-thread performance, 9.5% ahead, is a clear Intel strength, driven by the 5.30 GHz boost clock versus AMD's 5.00 GHz. The find prime numbers test, though nearly tied at 0.6% apart, still goes to Intel, suggesting a slight advantage in simple integer loops that are latency-bound rather than throughput-bound.

Architecture Differences

The two processors take fundamentally different approaches to core design. The AMD Ryzen AI Max+ 392 uses 12 cores and 24 threads based on the Zen 5 architecture, codenamed Strix Halo, fabricated on a 4 nm process at TSMC. The die is composed of two chiplets, each measuring 70.6 mm². Each core has 80 KB of L1 cache and 1 MB of L2 cache, with a shared 64 MB L3 cache. The processor supports LPDDR5X memory over a quad-channel bus, yielding 256.0 GB/s of memory bandwidth, and it supports ECC memory. The integrated graphics are Radeon 8060S, and the socket is AMD Socket FP11. The base clock is 3.20 GHz, and the boost clock is 5.00 GHz, with a TDP of 55 watts. The part is unlocked, meaning the multiplier cannot be adjusted.

The Intel Core Ultra 7 265T uses 20 cores and 20 threads, with no hyperthreading, based on the Arrow Lake architecture, codenamed Arrow Lake-S, fabricated on a 3 nm process at TSMC. The die is a monolithic 243 mm² with 17,800 million transistors. Each core has 192 KB of L1 cache and 3 MB of L2 cache, with a shared 30 MB L3 cache. The processor supports DDR5 memory over a dual-channel bus, yielding 102.4 GB/s of memory bandwidth, and it does not support ECC memory. The integrated graphics are Arc Xe-LPG Graphics with 64 execution units. The socket is Intel Socket 1851. The base clock is 1.50 GHz, and the boost clock is 5.30 GHz, with a TDP of 35 watts. The launch MSRP is $384. The multiplier is also locked.

The process node difference, 4 nm versus 3 nm, gives Intel a density and power efficiency advantage at the transistor level, but the AMD design compensates with a larger L3 cache and significantly higher memory bandwidth. The quad-channel LPDDR5X interface on AMD delivers 2.5 times the bandwidth of Intel's dual-channel DDR5. The core count difference, 20 versus 12, is offset by AMD's SMT, which provides 24 threads versus Intel's 20. The L1 and L2 cache per core are larger on Intel, 192 KB and 3 MB versus 80 KB and 1 MB, but the shared L3 is more than double on AMD, 64 MB versus 30 MB.

The PCIe configuration also differs: AMD offers Gen 4 with 16 lanes, while Intel offers Gen 5 with 20 lanes. Intel's newer PCIe standard provides higher bandwidth for discrete GPUs and NVMe storage, but AMD's memory subsystem is more capable for CPU-bound tasks. The market segments differ as well, with AMD targeting mobile and Intel targeting desktop, which aligns with the TDP difference of 55 watts versus 35 watts.

FAQ

Q: Which processor has a higher average benchmark score?

A: The AMD Ryzen AI Max+ 392 has an average score of 90541, while the Intel Core Ultra 7 265T has an average score of 47697.

Q: How do the core and thread counts compare?

A: The AMD part has 12 cores and 24 threads, while the Intel part has 20 cores and 20 threads, with no hyperthreading.

Q: What is the difference in memory bandwidth?

A: The AMD Ryzen AI Max+ 392 delivers 256.0 GB/s over a quad-channel LPDDR5X bus, while the Intel Core Ultra 7 265T delivers 102.4 GB/s over a dual-channel DDR5 bus.

Q: Which processor wins in single-thread performance?

A: The Intel Core Ultra 7 265T wins the single-thread test with a score of 4338 versus AMD's 3927, a 9.5% advantage.

Q: Are both processors factory unlocked for overclocking?

A: No, both have locked multipliers, so the multiplier cannot be adjusted on either part.

Q: What are the cache configurations?

A: The AMD part has 80 KB L1 and 1 MB L2 per core, with 64 MB shared L3. The Intel part has 192 KB L1 and 3 MB L2 per core, with 30 MB shared L3.

Q: Which processor has a higher boost clock?

A: The Intel Core Ultra 7 265T boosts to 5.30 GHz, while the AMD Ryzen AI Max+ 392 boosts to 5.00 GHz.

The Verdict

The data points to distinct use cases for each processor. The AMD Ryzen AI Max+ 392 is the stronger choice for multi-threaded, memory-intensive, and vector-heavy workloads. Its wins in extended instructions, data compression, integer math, multithread, physics, and random string sorting are all substantial, ranging from 20.7% to 59.6% over Intel. The 64 MB L3 cache and 256.0 GB/s memory bandwidth are the primary drivers of these results. The 96th percentile ranking among all CPUs, with an average score of 90541, places it near the top of the database, within 0.7% of the AMD Ryzen 9 9955HX and within 0.2% of the Intel Xeon 654.

The Intel Core Ultra 7 265T is the better option for single-thread performance and floating point math. The 5.30 GHz boost clock delivers a 9.5% single-thread advantage, and the floating point score is 23.3% higher, indicating a strong per-core FPU design. Data encryption also favors Intel, though by a smaller margin. The 90th percentile ranking, with an average score of 47697, is competitive with its nearest rivals, all within 0.5% of each other. The lower TDP of 35 watts versus 55 watts also makes it a more power-efficient choice for desktop builds with limited cooling.

The overall average scores are not directly comparable because the AMD part includes a different mix of benchmark tests, but the head-to-head results are clear. For users prioritizing parallel compute, large datasets, and cache-sensitive workloads, the AMD Ryzen AI Max+ 392 is the data-supported winner. For users prioritizing single-thread responsiveness, floating point throughput, and lower power consumption, the Intel Core Ultra 7 265T is the better fit. The choice comes down to workload type rather than overall capability.