AMD Ryzen 9 9900X vs AMD Ryzen AI 9 HX 470 Comparison

AMD
AMD

AMD Ryzen 9 9900X

CORE STATE Granite Ridge
CORE SPECS 12 Cores / 24 Threads
CLOCK SPEED 4.4 Base / 5.6 GHz Turbo
CACHE 64 MB
MAX TDP 120W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2024
VS
AMD
AMD

Ryzen AI 9 HX 470

CORE STATE Gorgon Point
CORE SPECS 12 Cores / 24 Threads
CLOCK SPEED 2 Base / 5.2 GHz Turbo
CACHE 16 MB
MAX TDP 28W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2026

PERFORMANCE BENCHMARKS

3dmark_16_threads
12,552
N/A
3dmark_2_threads
2,519
N/A
3dmark_4_threads
4,886
N/A
3dmark_8_threads
9,114
N/A
3dmark_max_threads
13,929
N/A
3dmark_single_thread
1,286
N/A
cinebench_cinebench_r15_multicore
5,008
3,518
cinebench_cinebench_r15_singlecore
353
317
cinebench_cinebench_r23_multicore
32,172
23,491.5
cinebench_cinebench_r23_singlecore
2,253
2,066
geekbench_multicore
22,174
N/A
geekbench_singlecore
3,010
N/A
passmark_data_compression
683,579
484,584
passmark_data_encryption
33,421
23,746
passmark_extended_instructions
55,243
34,365
passmark_find_prime_numbers
436
125
passmark_floating_point_math
120,083
81,320
passmark_integer_math
181,056
130,171
passmark_multithread
54,643
37,010
passmark_physics
3,381
1,851
passmark_random_string_sorting
72,013
51,576
passmark_single_thread
4,672
4,124
passmark_singlethread
4,672
4,124

Analysis: AMD Ryzen 9 9900X vs AMD Ryzen AI 9 HX 470

The AMD Ryzen AI 9 HX 470 and AMD Ryzen 9 9900X are both 12-core, 24-thread Zen 5 parts, but they target entirely different platforms. The data shows a decisive sweep for the desktop-class 9900X across every single benchmark in the comparison, while the mobile-focused HX 470 brings its own set of platform advantages that the raw scores do not capture.

Head-to-Head Benchmarks

The benchmark results are unambiguous: the AMD Ryzen 9 9900X wins all 15 head-to-head comparisons listed, with no victories for the AMD Ryzen AI 9 HX 470. The largest single margin comes in the passmark_find_prime_numbers test, where the 9900X scores 436 against the HX 470’s 125, a delta of -71.3%. This is a massive gap in a workload that stresses raw integer throughput and memory latency, suggesting the desktop chip’s higher base and boost clocks are doing heavy lifting. The 9900X's clock advantage is evident here; its 5.60 GHz boost and 4.40 GHz base far outpace the HX 470's 5.20 GHz and 2.00 GHz respectively.

In multi-threaded rendering, the 9900X maintains a commanding lead. In cinebench_r23_multicore, it scores 32172 versus 23491.5, a -27% delta. A similar pattern appears in cinebench_r15_multicore, where the 9900X posts 5008 against 3518, a -29.8% difference. These results are partly explained by the 9900X’s 120W TDP compared to the HX 470’s 28W, allowing sustained high clocks across all cores. The desktop chip’s physics score in passmark_physics reinforces this, with 3381 versus 1851—a -45.3% delta, indicating a strong advantage in simulation workloads that rely on sustained multi-core performance.

Single-threaded performance is closer but still favors the desktop part. In cinebench_r23_singlecore, the 9900X scores 2253 against 2066, a -8.3% delta. The passmark_single_thread test shows a similar -11.7% gap, with scores of 4672 and 4124. While the 9900X wins, the HX 470 is not far behind, which is notable given its dramatically lower power envelope. The mobile chip’s 5.20 GHz boost clock is close to the desktop’s 5.60 GHz, so the difference likely comes down to sustained clock behavior under load rather than peak capability.

Memory-heavy and encryption workloads follow the same trend. In passmark_data_compression, the 9900X scores 683579 versus 484584, a -29.1% delta. Data encryption shows a -28.9% gap, with scores of 33421 and 23746. Extended instruction performance is where the 9900X pulls further ahead: passmark_extended_instructions shows 55243 versus 34365, a -37.8% delta. This large margin suggests the desktop part benefits from higher memory bandwidth in practice, even though both list the same 89.6 GB/s memory bandwidth rating. The 9900X’s access to 64 MB of L3 cache versus the HX 470’s 16 MB likely plays a significant role in these compute-heavy tasks.

The overall average benchmark scores reflect this dominance. The HX 470 has an average benchmark score of 58826, while the 9900X sits at 57498. This is a curious inversion—the mobile chip has a higher average score despite losing every head-to-head test. This is explained by the different benchmark suites in each part’s full profile; the HX 470 includes passmark_data_encryption and other tests where it scores relatively well, but the shared head-to-head tests all favor the 9900X. Both parts sit at the 92nd percentile against all CPUs, indicating they are both near the top of the performance stack, but the desktop part is clearly stronger in the tests where they directly compete.

FAQ

Q: Which processor wins in multi-core benchmarks?

A: The AMD Ryzen 9 9900X wins decisively. In cinebench_r23_multicore, it scores 32172 versus the HX 470’s 23491.5, a -27% delta. The passmark_multithread test shows a -32.3% gap, with scores of 54643 and 37010.

Q: Is the single-core performance much different?

A: The 9900X is ahead, but not by a huge margin. In cinebench_r23_singlecore, the 9900X scores 2253 against 2066, a -8.3% delta. The passmark_single_thread test shows a -11.7% gap, with scores of 4672 and 4124.

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

A: The largest gap is in passmark_find_prime_numbers, where the 9900X scores 436 versus 125, a -71.3% delta. This suggests a significant difference in integer-heavy, latency-sensitive workloads.

Q: Do both chips have the same core and thread counts?

A: Yes, both the AMD Ryzen AI 9 HX 470 and the AMD Ryzen 9 9900X have 12 cores and 24 threads. Both use the Zen 5 architecture and are built on TSMC’s 4 nm process.

Q: What are the TDP differences?

A: The 9900X has a TDP of 120W, while the HX 470 is rated at 28W. This is a substantial difference that allows the desktop part to sustain higher clocks under load.

Q: How do their cache sizes compare?

A: The 9900X has 64 MB of L3 cache, while the HX 470 has 16 MB. Both have 80 KB of L1 and 1 MB of L2 per core.

Architecture Differences

Both processors share the fundamental Zen 5 architecture and are fabricated on TSMC’s 4 nm process, but their physical designs and platform targets diverge sharply. The HX 470 uses the Gorgon Point codename and fits the AMD Socket FP8, indicating a mobile-first design. Its die size is listed as 233 mm², which is a single, unified die. The 9900X, codenamed Granite Ridge, uses the AMD Socket AM5 for desktop systems and has a die size of 2x 70.6 mm², indicating a chiplet-based design with two CCDs. The 9900X also lists 16,630 million transistors, while the HX 470 has no transistor count listed.

The cache hierarchy is a major differentiator. The HX 470 has 16 MB of L3 cache, while the 9900X has 64 MB, a fourfold increase. Both have identical L1 and L2 allocations per core at 80 KB and 1 MB respectively. The 9900X also supports ECC memory, while the HX 470 does not. Memory support differs as well: the HX 470 supports both DDR5 and LPDDR5X, while the 9900X only lists DDR5. Both have dual-channel memory buses and the same rated memory bandwidth of 89.6 GB/s.

PCIe connectivity is another clear split. The HX 470 offers Gen 4 with 16 lanes (CPU only), while the 9900X provides Gen 5 with 24 lanes (CPU only). This makes the 9900X significantly more capable for high-bandwidth add-in cards and NVMe storage. The integrated graphics also differ: the HX 470 features a Radeon 890M, while the 9900X has a more generic Radeon Graphics solution. The HX 470’s mobile segment designation and 28W TDP contrast with the 9900X’s desktop segment and 120W TDP, which is the core reason for the performance differential.

The Verdict

The data points to a clear choice for users who prioritize raw compute performance: the AMD Ryzen 9 9900X is the superior processor in every head-to-head benchmark recorded. With a 27% lead in cinebench_r23_multicore and a 71.3% lead in passmark_find_prime_numbers, the 9900X is the pick for rendering, physics simulation, and heavy integer workloads. Its 64 MB of L3 cache and 120W TDP enable sustained performance that the 28W HX 470 cannot match. The 9900X also offers ECC memory support, PCIe Gen 5 with 24 lanes, and an unlocked multiplier, making it a more flexible and future-proof desktop part.

The AMD Ryzen AI 9 HX 470 is not a poor performer—it sits at the 92nd percentile against all CPUs and comes within 8.3% of the 9900X in cinebench_r23_singlecore. Its advantage lies in efficiency and platform integration. With a 28W TDP, it is designed for mobile systems where power consumption and thermals are critical. It supports LPDDR5X memory alongside DDR5, which is a mobile-specific feature, and fits the FP8 socket for thin-and-light laptops. For users building a desktop system with no power constraints, the 9900X is the obvious choice. For a laptop that needs strong multi-core performance without the desktop power draw, the HX 470 is the only viable option of the two. The 9900X’s launch MSRP is $499, though that figure does not factor into performance comparisons.

Specification Differences

The following table highlights only the fields where the two processors differ:

| Specification | AMD Ryzen AI 9 HX 470 | AMD Ryzen 9 9900X |

|---|---|---|

| Base Clock | 2.00 GHz | 4.40 GHz |

| Boost Clock | 5.20 GHz | 5.60 GHz |

| TDP | 28W | 120W |

| Socket | AMD Socket FP8 | AMD Socket AM5 |

| Codename | Gorgon Point | Granite Ridge |

| Generation | Ryzen AI 400 (Zen 5 / Zen 5c) | Ryzen 9 (Zen 5 (Granite Ridge)) |

| Die Size | 233 mm² | 2x 70.6 mm² |

| Transistors | Not listed | 16,630 million |

| L3 Cache | 16 MB | 64 MB |

| Memory Support | DDR5, LPDDR5X | DDR5 |

| ECC Memory | No | Yes |

| PCIe | Gen 4, 16 Lanes (CPU only) | Gen 5, 24 Lanes (CPU only) |

| Integrated Graphics | Radeon 890M | Radeon Graphics |

| Market Segment | Mobile | Desktop |

| Release Date | 2025-12-31 | 2024-08-14 |

| Multiplier Unlocked | No | Yes |

| Launch MSRP | Not listed | $499 |

DETAILED SPECIFICATIONS

SPECIFICATION
9 9900X
AI 9 HX 470
Core Specs
Cores
12
12 0.0%
Threads
24
24 0.0%
Base Clock (GHz)
4.4
2 -54.5%
Boost Clock (GHz)
5.6
5.2 -7.1%
Frequency (GHz)
4.4
2 -54.5%
Turbo Clock (GHz)
5.6
5.2 -7.1%
Multiplier
44
20 -54.5%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
80 KB (per core)
80 KB (per core)
L2 Cache
1 MB (per core)
1 MB (per core)
L3 Cache
64 MB
16 MB
Power
TDP (W)
120
28 -76.7%
PPT
162 W
—
Configurable TDP
—
15-54 W
Architecture
Architecture
Zen 5
Zen 5
Codename
Granite Ridge
Gorgon Point
Generation
Ryzen 9 (Zen 5 (Granite Ridge))
Ryzen AI 400 (Zen 5 / Zen 5c)
Process Size
4 nm
4 nm
Transistors
16,630 million
—
Die Size
2x 70.6 mm²
233 mm²
Foundry
TSMC
TSMC
Memory
Memory Support
DDR5
DDR5, LPDDR5X
Memory Bus
Dual-channel
Dual-channel
Memory Bandwidth
89.6 GB/s
89.6 GB/s
ECC Memory
Yes
No
Platform
Socket
AMD Socket AM5
AMD Socket FP8
Chipsets
X870E, X870, B850, B840, X670E, X670, B650E, B650, A620
—
PCIe
Gen 5, 24 Lanes(CPU only)
Gen 4, 16 Lanes(CPU only)
Intel Hybrid
Hybrid Cores
—
4 + 8
E-Core Frequency
—
2000 MHz up to 3.3 GHz
AMD Multi-Die
IO Process Size
6 nm
—
AI/NPU
NPU
—
Yes / 55 TOPS
Graphics
Integrated Graphics
Radeon Graphics
Radeon 890M
Other
Market
Desktop
Mobile
Production Status
Active
Active
Launch Price
$499
—
Part Number
100-000000662
100-000001936
Package
FC-LGA1718
FP8
Tj Max
95°C
100°C
Bundled Cooler
None
—
View Ryzen 9 9900X Details View Ryzen AI 9 HX 470 Details