AMD Ryzen 9 9900X vs Intel Core Ultra 5 338H Comparison

Where Each One Wins

The benchmark record is unambiguous: AMD Ryzen 9 9900X wins all 15 head-to-head comparisons against Intel Core Ultra 5 338H. There is no workload category in the database where the Intel mobile part takes a single victory. The wins span every major benchmark family, including Cinebench R15 and R23 in both multi-core and single-core, plus the full PassMark suite covering integer math, floating-point math, encryption, compression, physics, prime number finding, extended instructions, random string sorting, and single-threaded performance.

The use-case split is therefore not about which processor wins a given task, but rather about the scale of the margin. For lightly threaded workloads, the AMD part holds a moderate but consistent edge. Single-core Cinebench R23 shows 2253 versus 2044, a 10.2% advantage. PassMark single-thread shows 4672 versus 4180, an 11.8% delta. These are meaningful gains for daily responsiveness and applications that rely on one or two threads.

For heavily threaded workloads, the gap widens dramatically. Cinebench R23 multi-core delivers 32172 for the AMD part versus 16331 for the Intel part, a 97% difference. PassMark integer math shows 181056 versus 64934, a 178.8% delta. The pattern holds across compression, encryption, extended instructions, and random string sorting, where AMD leads by margins ranging from 56.4% to 147.2%. The data indicates the Ryzen 9 9900X is decisively stronger in multi-threaded rendering, compilation, data processing, and scientific workloads.

The Intel Core Ultra 5 338H does have one notable advantage in the system-level context: its thermal envelope. The database records a 25 watt TDP for the mobile part, which positions it for thin-and-light laptops. The AMD part carries a 120 watt TDP and is a desktop processor. The Intel part's 136.5 GB/s memory bandwidth exceeds the AMD part's 89.6 GB/s, which may help in memory-bandwidth-sensitive tasks, though the benchmark results do not show this translating into a win on any recorded test.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen 9 9900X is a desktop part built on the Zen 5 architecture with the Granite Ridge codename, manufactured on a 4 nm process by TSMC. The Intel Core Ultra 5 338H is a mobile processor using the Panther Lake architecture (Panther Lake-H generation), built on a 3 nm process at Intel's own fabs.

Both parts have 12 physical cores, but the thread counts diverge sharply. The AMD processor supports 24 threads via simultaneous multithreading. The Intel processor has no SMT and runs 12 threads total. This structural difference explains much of the multi-threaded benchmark gap: the AMD part can execute two threads per core, effectively doubling its scheduling capacity in heavily threaded workloads.

Cache hierarchies also differ. The AMD part uses 80 KB L1 per core, 1 MB L2 per core, and a 64 MB shared L3 cache. The Intel part uses 192 KB L1 per core, 2.5 MB L2 per core, and only 18 MB shared L3. The AMD processor's larger L3 (64 MB versus 18 MB) provides substantially more on-die capacity for working sets, which benefits database-style workloads, large simulations, and repeated data access patterns.

Clock speeds are another differentiator. The AMD part has a 4.40 GHz base clock and 5.60 GHz boost clock. The Intel part has a 1.90 GHz base clock and 4.70 GHz boost clock. The AMD base clock is more than double the Intel base clock, reflecting the desktop power budget versus mobile efficiency targets. Boost clocks are closer, but the AMD part still holds a 0.90 GHz advantage.

Memory support differs: AMD uses DDR5 with dual-channel configuration and ECC support enabled. Intel uses LPDDR5X with dual-channel configuration and no ECC support. The AMD part provides PCIe Gen 5 with 24 CPU lanes, while the Intel part provides PCIe Gen 5 with only 4 CPU lanes. The AMD part includes Radeon Graphics as integrated graphics; the Intel part includes Arc B370. The AMD processor has an unlocked multiplier; the Intel one is locked. The AMD part targets the AM5 socket, while the Intel part uses BGA 2540, which is soldered for mobile systems.

Head-to-Head Benchmarks

The largest single margin in the database is PassMark integer math, where the AMD Ryzen 9 9900X scores 181056 against 64934 for the Intel Core Ultra 5 338H, a 178.8% advantage. This benchmark exercises heavy integer arithmetic, branch prediction, and instruction-level parallelism. The combination of higher clocks, SMT, and larger caches gives the AMD part a commanding lead in this core compute metric.

Data compression shows the second-largest gap: 683579 for AMD versus 276539 for Intel, a 147.2% delta. Compression workloads are highly parallel and benefit from large caches and many threads. The AMD part's 64 MB L3 and 24 threads align well with this pattern. Extended instructions follow closely: 55243 versus 23906, a 131.1% delta. This benchmark measures AVX and other extended instruction set throughput, where the Zen 5 architecture's wider execution resources and higher boost clocks deliver substantial gains.

Random string sorting shows 72013 versus 34082, a 111.3% delta. This workload involves pointer chasing and memory access patterns that favor the AMD part's cache hierarchy. Cinebench R23 multi-core shows 32172 versus 16331, a 97% delta, confirming the rendering workload advantage. Cinebench R15 multi-core shows 5008 versus 2504, exactly 100% delta, meaning the AMD part scores double.

PassMark multi-thread shows 54643 versus 28717, a 90.3% delta. Data encryption shows 33421 versus 21367, a 56.4% delta. Find prime numbers shows 436 versus 304, a 43.4% delta. Floating-point math shows 120083 versus 84067, a 42.8% delta. Physics shows 3381 versus 2697, a 25.4% delta.

Single-threaded margins are smaller but still consistently favor AMD. Cinebench R23 single-core: 2253 versus 2044, 10.2% delta. Cinebench R15 single-core: 353 versus 305, 15.7% delta. PassMark single-thread: 4672 versus 4180, 11.8% delta. These results indicate the Zen 5 core design holds a measurable IPC and clock advantage even without the benefits of SMT or large caches.

The average benchmark score comparison reinforces the pattern. The AMD part records an average score of 57498 across its benchmark suite, while the Intel part records 33989. The AMD part sits at the 92nd percentile among all CPUs in the database, while the Intel part sits at the 84th percentile. The nearest rivals for the AMD part include Intel Core i9-14900 with a 1.1% lower average score and AMD EPYC 7313 with a 0.2% higher score. The nearest rivals for the Intel part include Intel Core Ultra 7 165H with a 0.3% lower score and AMD EPYC 4244P with a 0.7% higher score.

The Verdict

The data supports a clear functional split. The AMD Ryzen 9 9900X is the choice for desktop systems where power budget is not the primary constraint and where maximum multi-threaded throughput, large cache capacity, and high clock speeds are required. Its 24 threads, 64 MB L3, 5.60 GHz boost, and 120 W TDP make it suitable for rendering, compiling, data processing, and heavy scientific computing. The 92nd percentile ranking and average score of 57498 place it in the upper tier of all recorded processors.

The Intel Core Ultra 5 338H serves a different role entirely. Its 25 W TDP, 12 threads, and 1.90 GHz base clock target mobile systems where battery life and thermal limits dominate. The LPDDR5X memory support with 136.5 GB/s bandwidth and the Arc B370 integrated graphics make it a viable option for thin-and-light laptops. Its 84th percentile ranking and average score of 33989 show it is still a capable mobile processor, but it does not compete with the desktop-class AMD part in raw performance.

The benchmark record shows no scenario where the Intel part wins. The closest margins are in single-threaded tests (10.2% to 15.7% deltas), while multi-threaded tests show deltas of 90% to 178.8%. Users who need maximum compute throughput should select the AMD Ryzen 9 9900X. Users who need a low-power mobile processor with integrated graphics and modern memory support should consider the Intel Core Ultra 5 338H, understanding that its performance envelope is categorically different.

FAQ

Q: Which processor has more threads?

A: The AMD Ryzen 9 9900X has 24 threads from 12 cores via SMT. The Intel Core Ultra 5 338H has 12 threads from 12 cores with no SMT.

Q: What is the largest performance gap between the two?

A: PassMark integer math shows the largest delta: 181056 for AMD versus 64934 for Intel, a 178.8% advantage. Cinebench R15 multi-core shows exactly 100% delta, with AMD doubling the Intel score.

Q: How do single-threaded scores compare?

A: The AMD part leads in all single-threaded tests. Cinebench R23 single-core shows 2253 versus 2044 (10.2% delta). Cinebench R15 single-core shows 353 versus 305 (15.7% delta). PassMark single-thread shows 4672 versus 4180 (11.8% delta).

Q: What memory types do these processors support?

A: The AMD Ryzen 9 9900X supports DDR5 with dual-channel configuration and ECC enabled. The Intel Core Ultra 5 338H supports LPDDR5X with dual-channel configuration and no ECC.

Q: What are the thermal design power ratings?

A: The AMD Ryzen 9 9900X has a 120 W TDP. The Intel Core Ultra 5 338H has a 25 W TDP.

Q: What are the cache sizes?

A: The AMD part has 80 KB L1 per core, 1 MB L2 per core, and 64 MB shared L3. The Intel part has 192 KB L1 per core, 2.5 MB L2 per core, and 18 MB shared L3.