AMD Ryzen AI 5 440G vs Intel Core Ultra 9 285H Comparison

Head-to-Head Benchmarks

The recorded data shows a decisive sweep for the Intel Core Ultra 9 285H across every shared PassMark test. The Intel part wins all 11 head-to-head comparisons, with the AMD Ryzen AI 5 440G failing to capture a single benchmark. The largest gap appears in passmark_find_prime_numbers, where Intel scores 330 against AMD's 90, a 72.7% deficit for the Ryzen part. That workload, which stresses integer iteration and branch handling, exposes the largest relative weakness in the AMD data.

Floating-point math follows a similar pattern. Intel delivers 109190 in passmark_floating_point_math, while AMD manages 45711, a 58.1% shortfall. The gap is substantial enough to classify the Ryzen AI 5 440G as clearly behind in numerically intensive tasks. Data encryption shows Intel at 26140 versus AMD's 13585, a 48% difference. Physics simulation also leans heavily toward Intel: 2513 against 1329, a 47.1% deficit. These four workloads collectively indicate that the Intel Core Ultra 9 285H holds a commanding lead in compute-heavy scenarios, not just a narrow edge.

Multithreaded performance follows the same trend. The passmark_multithread score for Intel is 34171, while AMD records 23487, a 31.3% gap. Integer math shows a smaller but still meaningful difference: Intel at 85922, AMD at 71251, a 17.1% deficit. Extended instructions land at 26794 for Intel and 20648 for AMD, a 22.9% gap. Random string sorting, a test that stresses memory access patterns and pointer chasing, gives Intel 40931 versus AMD's 31106, a 24% difference. Data compression shows Intel at 335859 against AMD's 292735, a 12.8% gap, which is the narrowest margin outside of single-thread results.

Single-thread performance is the closest contest in the entire comparison. Intel scores 4415 in passmark_single_thread, while AMD scores 4060, an 8% deficit. This indicates that the AMD core design holds up reasonably well on lightly threaded tasks, but the Intel part still leads in every measured category. The average benchmark score reinforces the overall picture: AMD averages 46187 across its recorded tests, while Intel averages 38312. That average can be misleading, however, because the two processors are tested across different benchmark suites. AMD's average includes only PassMark tests, while Intel's average also includes Cinebench and Geekbench results, which are not present in the AMD data. The head-to-head PassMark results, where both parts share identical workloads, provide the cleaner comparison.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI 5 440G is a 6-core, 12-thread desktop part built on TSMC's 4 nm node, with a die size of 195 mm². Its codename is Gorgon Point, and it belongs to the Ryzen AI 400 generation, which combines Zen 5 and Zen 5c cores. The Intel Core Ultra 9 285H, by contrast, is a 16-core, 16-thread mobile part built on TSMC's 3 nm node, with the codename Arrow Lake-H and the architecture name Arrow Lake. Both use TSMC as the foundry, but the process difference matters: 4 nm for AMD versus 3 nm for Intel.

Core counts tell a clear story. Intel packs 16 cores and 16 threads, which matches its hybrid design without hyper-threading. AMD offers 6 cores and 12 threads, relying on simultaneous multithreading to reach its thread count. The Intel part also has a higher boost clock at 5.40 GHz versus AMD's 4.80 GHz, and a higher base clock at 2.90 GHz versus AMD's 2.00 GHz. Despite the higher clocks, Intel's TDP is lower at 45 watts, while AMD's desktop-oriented part sits at 65 watts.

Cache hierarchies differ substantially. Intel provides 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3. AMD provides 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3. The Intel L3 cache is three times larger, which helps explain its advantage in data compression and random string sorting, workloads that benefit from larger on-die storage. Memory bandwidth also favors Intel: 102.4 GB/s versus AMD's 89.6 GB/s. Both support DDR5, but Intel additionally supports LPDDR5X, while AMD lists only DDR5. Both support ECC memory.

The socket and platform targets differ sharply. AMD uses Socket AM5, a desktop socket, with PCIe Gen 4 and 12 CPU lanes. Intel uses BGA 2049, a soldered mobile package, with PCIe Gen 5 and 8 CPU lanes. The AMD part has an unlocked multiplier, while the Intel part is locked. Integrated graphics also differ: AMD pairs with the Radeon 840M, while Intel uses the Arc Graphics 140T. The release dates show Intel arrived in January 2025, while AMD's part is dated February 2026.

FAQ

Q: Which processor wins the most head-to-head benchmarks?

A: The Intel Core Ultra 9 285H wins all 11 shared PassMark tests. The AMD Ryzen AI 5 440G records zero wins in the head-to-head data.

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

A: Intel scores 4415 in passmark_single_thread, while AMD scores 4060, an 8% deficit. This is the smallest gap in the entire comparison.

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 9 285H has 16 cores and 16 threads. The AMD Ryzen AI 5 440G has 6 cores and 12 threads.

Q: What is the difference in process node?

A: Intel uses TSMC's 3 nm process, while AMD uses TSMC's 4 nm process. Both are manufactured by TSMC.

Q: Which processor has the larger L3 cache?

A: Intel has 24 MB of shared L3 cache. AMD has 8 MB of L3 cache, so Intel's L3 is three times larger.

Q: Do both processors support ECC memory?

A: Yes, both the AMD Ryzen AI 5 440G and the Intel Core Ultra 9 285H list ECC memory support in the database.

The Verdict

The data points to a clear winner for raw performance: the Intel Core Ultra 9 285H. It leads in every shared benchmark, often by wide margins. The 72.7% advantage in prime number finding, the 58.1% lead in floating-point math, and the 48% gap in encryption all indicate that Intel's architecture handles demanding computational workloads with substantially more headroom. The multithread score of 34171 versus 23487 reinforces that conclusion, showing a 31.3% advantage in heavily parallel tasks.

The AMD Ryzen AI 5 440G does not win any head-to-head test, but its single-thread score of 4060 sits only 8% behind Intel's 4415. That suggests the Zen 5-based core design is competitive on lightly threaded workloads, even though the overall package falls short. The AMD part also carries a desktop socket and an unlocked multiplier, which are characteristics typically associated with overclocking potential. The Intel part, locked and soldered to a mobile package, offers no such flexibility.

Percentile rankings complicate the simple "Intel wins" narrative. AMD sits at the 89th percentile among all CPUs, while Intel sits at the 86th. The AMD part's nearest rivals include the Intel Core i9-13900HX at a 0.2% average score delta, the Intel Core Ultra 5 235 at 0.3%, and the AMD Ryzen AI 9 HX 375 at 0.3%. Intel's nearest rivals include the Intel Core 9 270H at -0.1%, the Intel Core i5-13600HX at 0.1%, and the AMD Ryzen 7 250 at 0.2%. The average scores of the two parts differ by roughly 21%, but their percentile ranks are close because Intel's average is dragged down by the inclusion of Cinebench and Geekbench results in its benchmark set, while AMD's average reflects only PassMark tests.

For users selecting purely on measured compute performance, the Intel Core Ultra 9 285H is the stronger choice in every shared workload. For users who prioritize a desktop platform, an unlocked multiplier, and a lower price point in absolute terms, the AMD part offers a different trade-off. The launch MSRP for Intel is $651, which can be stated once as a reference point, but the database does not record a launch MSRP for AMD.

Specification Differences

The two processors differ across nearly every major specification field. The AMD Ryzen AI 5 440G uses 6 cores and 12 threads, while the Intel Core Ultra 9 285H uses 16 cores and 16 threads. Base clocks are 2.00 GHz for AMD and 2.90 GHz for Intel; boost clocks are 4.80 GHz and 5.40 GHz respectively. TDP is 65 watts for AMD and 45 watts for Intel. AMD uses Socket AM5, while Intel uses BGA 2049. The process node is 4 nm for AMD and 3 nm for Intel.

Cache structures diverge completely. AMD has 80 KB L1 per core, 1 MB L2 per core, and 8 MB L3. Intel has 192 KB L1 per core, 3 MB L2 per core, and 24 MB shared L3. Memory support lists DDR5 for AMD and DDR5 plus LPDDR5X for Intel. Memory bandwidth is 89.6 GB/s for AMD and 102.4 GB/s for Intel. PCIe support is Gen 4 with 12 lanes for AMD and Gen 5 with 8 lanes for Intel. Integrated graphics are Radeon 840M for AMD and Arc Graphics 140T for Intel. The AMD multiplier is unlocked; the Intel multiplier is locked. The AMD part number is 100-000001918, while Intel's is SRQAL. The market segments are Desktop for AMD and Mobile for Intel.

Where Each One Wins

The Intel Core Ultra 9 285H wins in every benchmark category where both parts have recorded data. Its strongest relative advantages appear in prime number finding, floating-point math, encryption, physics simulation, and multithreaded workloads. These are compute-bound tasks that benefit from the Intel part's 16 cores, higher clocks, larger L3 cache, and higher memory bandwidth. For applications that stress integer throughput, extended instruction sets, or memory-heavy data movement, the Intel part's 17.1% to 24% leads across integer math, extended instructions, and random string sorting make it the preferred option in the recorded data.

The AMD Ryzen AI 5 440G shows its best relative performance in single-thread tests, where the 8% gap is the closest margin in the entire comparison. The desktop socket, unlocked multiplier, and 65 watt TDP suggest the AMD part is positioned for a different use case than the mobile, soldered Intel chip. The AMD part also carries a higher percentile rank at 89 versus Intel's 86, reflecting its stronger standing within the full database of CPUs when all its recorded benchmarks are averaged together.

The practical split is simple: the Intel Core Ultra 9 285H is the performance leader across every shared workload, with the largest margins in floating-point, encryption, physics, and multithreaded tasks. The AMD Ryzen AI 5 440G offers a desktop platform, an unlocked multiplier, and a closer single-thread result, but it does not match Intel in any measured head-to-head test. The database records zero wins for AMD and 11 wins for Intel, which makes the directional call unambiguous.