AMD Ryzen AI 9 HX PRO 370 vs AMD Ryzen Embedded 9900X Comparison
AMD Ryzen AI 9 HX PRO 370
Ryzen Embedded 9900X
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen AI 9 HX PRO 370 vs AMD Ryzen Embedded 9900X
Head-to-Head Benchmarks
The recorded data shows a fundamental asymmetry between these two processors: the AMD Ryzen AI 9 HX PRO 370 has a full suite of benchmark scores, while the AMD Ryzen Embedded 9900X has no recorded benchmark entries in the database. This means head-to-head comparisons must be drawn from the HX PRO 370's absolute scores and its position against its nearest rivals, rather than direct side-by-side measurements.
The HX PRO 370 delivers a Cinebench R23 multi-core score of 18,520 and a single-core score of 1,992. Its Cinebench R15 results are 3,178 multi-core and 309 single-core. In PassMark testing, the processor shows strong throughput in integer math at 120,016 and floating point math at 73,493. Data compression reaches 403,113, while data encryption scores 20,788. Extended instructions produce 28,165, and random string sorting records 43,767. The PassMark multi-thread score is 33,349, with a single-thread score of 4,003 and a physics score of 1,892. Prime number finding is the lowest result at 131.
The average benchmark score for the HX PRO 370 is 50,448, placing it in the 90th percentile of all CPUs in the database. Its nearest rival, the Intel Core i9-13980HX, records an average score of 50,398, a delta of 0.1 percent. The AMD Ryzen 9 5900XT scores 50,718, which is 0.5 percent higher. The Intel Core i7-13850HX posts 50,761, a 0.6 percent advantage, and the Intel Core i9-14900T reaches 51,015, 1.1 percent higher. These deltas indicate that the HX PRO 370 sits in a tight performance cluster, effectively matching the top mobile HX parts from Intel while trailing them by only a single percentage point or less. The 90th percentile ranking confirms that this is a high-end mobile processor by database-wide standards.
Because the Ryzen Embedded 9900X has no benchmark scores, there are no recorded wins for either processor in direct competition. The winsA and winsB fields both stand at zero. The analysis must therefore rely on architectural and specification differences to characterize what the Embedded 9900X would bring to a comparison, while acknowledging that its performance data is absent from the database.
Architecture Differences
The two processors share the same Zen 5 architecture family, but they diverge in nearly every implementation detail. The HX PRO 370 uses the Strix Point codename and belongs to the Ryzen AI PRO 300 generation, which combines Zen 5 and Zen 5c cores in a hybrid layout. The Embedded 9900X carries the Granite Ridge codename and is listed under the Ryzen Embedded generation, also based on Zen 5. Both are manufactured on a 4 nm process at TSMC, so the transistor fabrication node is identical.
The physical implementation differs substantially. The HX PRO 370 uses a monolithic die of 233 mm² with no transistor count listed. The Embedded 9900X uses a chiplet design with two dies, each measuring 70.6 mm², for a combined die area of approximately 141.2 mm², and the database records 16,630 million transistors. This chiplet approach is typical of desktop-oriented Granite Ridge parts, while the monolithic Strix Point die is optimized for mobile integration.
Cache hierarchies are a major differentiator. Both processors provide 80 KB of L1 cache per core and 1 MB of L2 cache per core. The L3 cache, however, is four times larger on the Embedded 9900X: 64 MB versus 16 MB on the HX PRO 370. This larger L3 pool is significant for workloads that repeatedly access a large working set, as it reduces trips to system memory.
Memory support also differs. The HX PRO 370 supports DDR5 and LPDDR5X, making it flexible for mobile designs that may use low-power memory. The Embedded 9900X supports DDR5 only. Both use a dual-channel memory bus and show the same 89.6 GB/s memory bandwidth in the database. Both processors support ECC memory, which matters for embedded and reliability-focused applications.
PCIe connectivity is another split. The HX PRO 370 provides PCIe Gen 4 with 16 lanes (CPU only). The Embedded 9900X provides PCIe Gen 5 with 24 lanes (CPU only). The newer PCIe generation and higher lane count give the Embedded part more headroom for expansion cards, NVMe storage, and I/O devices. The integrated graphics differ as well: the HX PRO 370 uses the Radeon 890M, while the Embedded 9900X uses a generic Radeon Graphics solution with no specific model number.
Clock speeds show the Embedded 9900X with a clear advantage. Its base clock is 4.40 GHz and boost clock is 5.60 GHz, compared to 2.00 GHz base and 5.10 GHz boost on the HX PRO 370. The base clock gap is particularly large, reflecting the different power envelopes. The HX PRO 370 has a TDP of 28 watts, suited to thin-and-light mobile systems. The Embedded 9900X has a TDP of 120 watts, reflecting a desktop-oriented power budget. Socket compatibility is also different: the HX PRO 370 uses AMD Socket FP8, while the Embedded 9900X uses AMD Socket AM5.
FAQ
Q: Which processor has a higher boost clock?
A: The AMD Ryzen Embedded 9900X boosts to 5.60 GHz, while the AMD Ryzen AI 9 HX PRO 370 boosts to 5.10 GHz.
Q: Do both processors support ECC memory?
A: Yes, both the HX PRO 370 and the Embedded 9900X have ECC memory support listed in the database.
Q: How much L3 cache does each processor have?
A: The HX PRO 370 has 16 MB of L3 cache, while the Embedded 9900X has 64 MB.
Q: What is the TDP difference between the two?
A: The HX PRO 370 has a TDP of 28 watts, and the Embedded 9900X has a TDP of 120 watts.
Q: Which processor uses a chiplet design?
A: The Embedded 9900X uses two dies, each 70.6 mm², with 16,630 million transistors. The HX PRO 370 uses a single 233 mm² die.
Q: Are there any recorded benchmark scores for the Embedded 9900X?
A: No, the database contains no benchmark entries for the Embedded 9900X, so its average benchmark score is zero and its percentile is 50.
Specification Differences
The two processors differ in the following recorded fields:
- Base clock: 2.00 GHz (HX PRO 370) vs 4.40 GHz (Embedded 9900X)
- Boost clock: 5.10 GHz (HX PRO 370) vs 5.60 GHz (Embedded 9900X)
- TDP: 28 watts (HX PRO 370) vs 120 watts (Embedded 9900X)
- Socket: AMD Socket FP8 (HX PRO 370) vs AMD Socket AM5 (Embedded 9900X)
- Codename: Strix Point (HX PRO 370) vs Granite Ridge (Embedded 9900X)
- Die size: 233 mm² (HX PRO 370) vs 2x 70.6 mm² (Embedded 9900X)
- Transistors: not listed (HX PRO 370) vs 16,630 million (Embedded 9900X)
- L3 cache: 16 MB (HX PRO 370) vs 64 MB (Embedded 9900X)
- Memory support: DDR5, LPDDR5X (HX PRO 370) vs DDR5 only (Embedded 9900X)
- PCIe: Gen 4, 16 lanes (HX PRO 370) vs Gen 5, 24 lanes (Embedded 9900X)
- Integrated graphics: Radeon 890M (HX PRO 370) vs Radeon Graphics (Embedded 9900X)
- Market segment: Mobile (HX PRO 370) vs Desktop (Embedded 9900X)
- Multiplier unlocked: false (HX PRO 370) vs true (Embedded 9900X)
- Release date: 2025-01-05 (HX PRO 370) vs 2025-10-06 (Embedded 9900X)
Fields that are identical include cores (12), threads (24), L1 cache (80 KB per core), L2 cache (1 MB per core), process node (4 nm), foundry (TSMC), memory bus (dual-channel), memory bandwidth (89.6 GB/s), ECC support (true), and production status (active). Neither processor has a launch MSRP listed in the database.
Where Each One Wins
The HX PRO 370 wins in the mobile segment by design. Its 28-watt TDP, FP8 socket, and support for LPDDR5X memory make it suitable for portable systems where power draw and thermal output are constrained. Its benchmark results confirm strong absolute performance for this class: the Cinebench R23 multi-core score of 18,520 and the 90th percentile ranking against all CPUs show that it does not sacrifice throughput for efficiency. The integrated Radeon 890M provides a specific, named graphics solution, which is useful for systems without a discrete GPU. For workloads like data compression (403,113) and integer math (120,016), the recorded scores indicate heavy multi-threaded capability.
The Embedded 9900X wins in raw architectural headroom. Its 64 MB L3 cache, PCIe Gen 5 with 24 lanes, and unlocked multiplier position it for desktop or embedded systems that need maximum I/O bandwidth and cache capacity. The base clock of 4.40 GHz and boost clock of 5.60 GHz suggest higher sustained single-thread throughput, though no benchmark scores exist to confirm this. The 120-watt TDP and AM5 socket indicate a platform designed for larger cooling solutions and higher power delivery. The chiplet die design with 16,630 million transistors points to a more complex, desktop-class implementation. The two-die layout also implies the processor is built for scalability in embedded applications where multiple dies are acceptable.
The cache difference is the clearest functional split. The Embedded 9900X's 64 MB L3 versus the HX PRO 370's 16 MB gives it a theoretical advantage in workloads with large, reusable data sets, such as database processing, virtualization, or scientific computing. The HX PRO 370's smaller cache is offset by its lower power draw and mobile integration. The PCIe difference matters for storage and expansion: Gen 5 with 24 lanes supports more and faster devices than Gen 4 with 16 lanes.
The Verdict
The database presents two processors with identical core and thread counts, the same process node, and the same Zen 5 architecture, but the similarities end there. The HX PRO 370 is a measured performer with a complete benchmark profile, an average score of 50,448, and a 90th percentile standing. It trades base clock, cache, and PCIe capability for a 28-watt TDP and mobile compatibility. The Embedded 9900X is an unmeasured part with a 120-watt TDP, a 5.60 GHz boost clock, 64 MB of L3 cache, PCIe Gen 5, and an unlocked multiplier. Its 50th percentile and zero average score reflect the absence of data, not necessarily weak performance.
For a system that must run on battery power or fit into a thin chassis, the HX PRO 370 is the only choice with recorded evidence of performance. Its benchmark scores show it competes within one percentage point of high-end Intel mobile parts like the Core i9-13980HX and Core i7-13850HX. For a desktop or embedded deployment where power is available and I/O expansion is critical, the Embedded 9900X offers superior specifications on paper: higher clocks, quadruple the L3 cache, more PCIe lanes, and a newer PCIe generation. The data does not verify its performance, so any claim about its speed must remain speculative. Each processor wins in its own domain, and the deciding factor is the target platform rather than the silicon itself.