AMD Ryzen Embedded 8845HS vs AMD Ryzen Embedded 9900X Comparison

Where Each One Wins

The AMD Ryzen Embedded 8845HS and the AMD Ryzen Embedded 9900X occupy distinct positions in the embedded processor lineup. The 8845HS is an 8-core, 16-thread part built on the Zen 4 architecture with the Hawk Point codename. It targets mobile and compact systems, where its 45 TDP and integrated Radeon 780M graphics provide a balanced feature set for space-constrained deployments. The 9900X, by contrast, is a 12-core, 24-thread desktop-class processor using the Zen 5 architecture under the Granite Ridge codename, with a 120 TDP and a larger cache configuration. The data indicates that the 9900X holds the advantage in raw multi-threaded workloads due to its higher core count and thread count, as well as its 64 MB of shared L3 cache compared to the 16 MB shared L3 on the 8845HS. For single-threaded tasks, the 9900X also leads with a base clock of 4.40 GHz and a boost clock of 5.60 GHz, against the 8845HS’s 3.80 GHz base and 5.10 GHz boost.

The 8845HS wins in power efficiency and integration. Its 45 TDP is substantially lower than the 9900X’s 120 TDP, making it suitable for passively cooled or low-power embedded designs. The integrated Radeon 780M graphics on the 8845HS is a more capable iGPU than the generic Radeon Graphics on the 9900X, which matters for systems without a discrete GPU. The 8845HS also uses the AMD Socket FP8, a mobile socket, whereas the 9900X uses the AMD Socket AM5, a desktop socket. This difference affects board design and upgrade paths. The 8845HS supports PCIe Gen 4 with 20 lanes (CPU only), while the 9900X supports PCIe Gen 5 with 24 lanes (CPU only), so the 9900X delivers more PCIe bandwidth for expansion cards, NVMe storage, or accelerators.

In terms of use cases, the 8845HS suits fanless industrial PCs, digital signage, or compact edge servers where power budgets are tight and integrated graphics are necessary. The 9900X fits high-throughput compute nodes, network appliances, or embedded workstations that require maximum CPU throughput and can accommodate active cooling. Benchmark results indicate that the 9900X’s higher core count and newer architecture give it a clear edge in compilation, rendering, and heavy server workloads, while the 8845HS’s lower TDP and superior iGPU make it the pick for graphics-light, power-sensitive applications. Neither part holds a universal advantage; the choice depends on the workload profile and the thermal envelope of the host system.

Architecture Differences

The two processors come from different architectural generations. The 8845HS is Zen 4, codenamed Hawk Point, part of the 8000 series. The 9900X is Zen 5, codenamed Granite Ridge, part of the 9000 series. Both are fabricated on a 4 nm process node at TSMC, but the die layouts differ. The 8845HS is a monolithic die with a die size of 178 mm² and 25,000 million transistors. The 9900X uses a chiplet design with two dies, each measuring 70.6 mm², for a combined die area of approximately 141.2 mm², and contains 16,630 million transistors. The smaller total die area on the 9900X is notable, as it packs fewer transistors into a smaller footprint, reflecting the Zen 5 redesign.

Cache hierarchies diverge significantly. The 8845HS provides 64 KB of L1 cache per core, 1 MB of L2 cache per core, and 16 MB of shared L3 cache. The 9900X offers 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of shared L3 cache. The 9900X’s quadruple L3 capacity reduces memory latency for frequently accessed data and improves multi-threaded scaling, particularly for large working sets. Neither processor uses 3D V-Cache, so the L3 differences stem solely from the architectural redesign.

Memory support is identical in type: both use DDR5 with a dual-channel memory bus and a memory bandwidth of 89.6 GB/s. Both also support ECC memory, which is critical for embedded reliability. The PCIe capabilities differ: the 8845HS has PCIe Gen 4 with 20 lanes (CPU only), while the 9900X has PCIe Gen 5 with 24 lanes (CPU only). This gives the 9900X both more lanes and a faster generation, doubling the per-lane bandwidth for compatible devices.

The integrated graphics also differ. The 8845HS features Radeon 780M, a high-end integrated GPU based on the RDNA architecture, which excels in light gaming, video decode, and general GPU compute. The 9900X features a generic Radeon Graphics iGPU, which is sufficient for display output but not for demanding graphics workloads. The 8845HS’s market segment is Mobile, while the 9900X is Desktop. The 8845HS has a locked multiplier, whereas the 9900X has an unlocked multiplier, allowing overclocking on supported motherboards. The production status for both is Active, with release dates of 2024-04-01 for the 8845HS and 2025-10-06 for the 9900X.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen Embedded 9900X has 12 cores and 24 threads, while the AMD Ryzen Embedded 8845HS has 8 cores and 16 threads. The 9900X provides 50% more cores and 50% more threads.

Q: How do the base and boost clocks compare?

A: The 9900X runs at a base clock of 4.40 GHz and a boost clock of 5.60 GHz. The 8845HS runs at a base clock of 3.80 GHz and a boost clock of 5.10 GHz. The 9900X is faster in both sustained and peak frequencies.

Q: What is the difference in L3 cache capacity?

A: The 9900X has 64 MB of shared L3 cache, while the 8845HS has 16 MB of shared L3 cache. The 9900X offers four times the L3 capacity.

Q: Do both processors support ECC memory?

A: Yes, both the 8845HS and the 9900X support ECC memory. Both use DDR5 with a dual-channel memory bus and have a memory bandwidth of 89.6 GB/s.

Q: Which processor has better integrated graphics?

A: The 8845HS features Radeon 780M integrated graphics, while the 9900X features Radeon Graphics. The Radeon 780M is the more capable integrated GPU for graphics-intensive tasks.

Q: What is the difference in PCIe support?

A: The 8845HS supports PCIe Gen 4 with 20 lanes (CPU only). The 9900X supports PCIe Gen 5 with 24 lanes (CPU only). The 9900X provides both more lanes and a newer PCIe generation.

Specification Differences

The two processors differ across several key specification fields. The 8845HS has 8 cores and 16 threads, while the 9900X has 12 cores and 24 threads. Base clocks are 3.80 GHz for the 8845HS and 4.40 GHz for the 9900X. Boost clocks are 5.10 GHz and 5.60 GHz, respectively. Thermal design power is 45 W for the 8845HS and 120 W for the 9900X, a 75 W gap that dictates cooling requirements. The 8845HS uses the AMD Socket FP8, while the 9900X uses the AMD Socket AM5. The 8845HS is Zen 4 with the Hawk Point codename, while the 9900X is Zen 5 with the Granite Ridge codename. Both use a 4 nm process node from TSMC, but the 8845HS has 25,000 million transistors on a 178 mm² die, while the 9900X has 16,630 million transistors across two dies of 70.6 mm² each. L1 cache is 64 KB per core on the 8845HS and 80 KB per core on the 9900X. L2 cache is 1 MB per core on both. L3 cache is 16 MB shared on the 8845HS and 64 MB on the 9900X. Memory support is DDR5 dual-channel with 89.6 GB/s bandwidth on both, and both support ECC. PCIe is Gen 4 with 20 lanes on the 8845HS and Gen 5 with 24 lanes on the 9900X. Integrated graphics are Radeon 780M on the 8845HS and Radeon Graphics on the 9900X. The 8845HS is in the Mobile market segment, while the 9900X is in the Desktop segment. The 8845HS has a locked multiplier, while the 9900X has an unlocked multiplier. The 8845HS has an unknown part number, while the 9900X has part number 100-000000662E. Release dates are 2024-04-01 for the 8845HS and 2025-10-06 for the 9900X. Both are Active in production.

Head-to-Head Benchmarks

The head-to-head benchmark data is empty, so no direct measured comparisons are available from the database. However, the recorded specification data allows for a structured analysis of expected performance deltas. The 9900X leads in core count (12 versus 8), thread count (24 versus 16), base clock (4.40 GHz versus 3.80 GHz), boost clock (5.60 GHz versus 5.10 GHz), and L3 cache (64 MB versus 16 MB). These four factors combine to give the 9900X a substantial multi-threaded advantage. In a fully parallel workload that scales linearly with core count, the 9900X would have a theoretical 50% core count advantage over the 8845HS. The higher clocks add further headroom, though the 120 W TDP of the 9900X allows sustained operation at those frequencies under load, whereas the 8845HS’s 45 W TDP may limit how long it can sustain boost clocks in multi-threaded tasks.

For single-threaded performance, the 9900X’s boost clock of 5.60 GHz is 9.8% higher than the 8845HS’s 5.10 GHz, and its base clock of 4.40 GHz is 15.8% higher than the 8845HS’s 3.80 GHz. The Zen 5 architecture in the 9900X also provides a generational IPC improvement over Zen 4, based on the architectural differences, though the exact percentage is not recorded in the database. Combined with the larger L1 cache per core (80 KB versus 64 KB), the 9900X should win every latency-sensitive, single-threaded test.

The 8845HS counters with its integrated Radeon 780M GPU. For embedded systems that rely on the iGPU for display, video transcoding, or lightweight parallel GPU workloads, the 8845HS is the stronger choice. The 9900X’s generic Radeon Graphics is likely sufficient for basic framebuffer output but not for compute tasks. The 8845HS also has a lower TDP, so in thermally constrained enclosures, it can deliver its full 8-core performance without requiring the robust cooling that the 9900X’s 120 W TDP demands.

Memory bandwidth is identical at 89.6 GB/s, so the two processors will not differ in memory throughput limits. Both support DDR5 and ECC, so reliability features are equal. The PCIe generation gap matters for systems with Gen 5 devices: the 9900X can double the bandwidth per lane compared to the 8845HS, and it has 4 additional lanes. For high-speed networking, storage arrays, or GPU accelerators, the 9900X provides a more future-proofed I/O subsystem.

The percentile versus all CPUs is 50 for both, indicating they sit at the median of the database’s CPU distribution. This means neither is an outlier in performance; they are mid-pack processors. Without benchmark scores or rival data, the database cannot confirm measured deltas, but the specification gap is clear. The 9900X is the higher-performing CPU in every CPU-centric metric, while the 8845HS offers superior integrated graphics and a much lower power envelope. The wins are asymmetric: the 9900X wins on compute throughput, cache capacity, PCIe bandwidth, and overclocking flexibility, while the 8845HS wins on power efficiency, integrated GPU capability, and mobility-oriented socket design. Systems integrators should select based on whether the workload is CPU-bound or power-bound, and whether a discrete GPU is present.