SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Embedded P164 + AMD Radeon 890M

Get a comprehensive performance analysis of your gaming rig with detailed benchmarks, bottleneck detection, and upgrade recommendations

90 / 100
ULTIMATE READY

Apex Performer

Top 10% of systems. Capable of 4K Ultra gaming and advanced rendering.

4K 60+ FPSVR ReadyRay Tracing

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
95%
VS
GPU
86%
PROCESSOR

AMD Ryzen AI Embedded P164

52,901 Benchmark Score
Top 5% Market Ranking
View Full Specs →
GRAPHICS CARD

AMD Radeon 890M

9,210 Benchmark Score
Top 14% Market Ranking
View Full Specs →

Market Position

How your build compares to others
Budget
0-30
Mid-Range
30-60
High-End
60-85
Enthusiast
85-100
Your Build

Game Performance Benchmarks

Real-world 4K FPS in popular titles
View All Games →

Performance Insights

Tips to maximize your system

Optimal Performance

Your system is in the top tier. You can run any modern game at maximum settings.

4K Gaming Ready

Consider a 4K 144Hz monitor to fully utilize your hardware capabilities.

Compatible Games See what you can play Compare CPUs Find upgrades Compare GPUs Find upgrades

Performance Tiers Explained

90-100

Ultimate

4K Ultra gaming, VR ready, ray tracing enabled, professional workloads

4K 60+ FPS VR Ready
70-89

High-End

1440p Ultra or 4K High settings, excellent for modern AAA titles

1440p Ultra 4K High
50-69

Mid-Range

1080p Ultra or 1440p Medium, great value for most gamers

1080p Ultra 1440p Med
30-49

Entry Level

1080p Medium settings, suitable for eSports and older titles

1080p Med eSports
0-29

Legacy

Basic gaming, older titles, consider upgrading for modern games

720p-1080p Low Older Games

The AMD Ryzen AI Embedded P164 pairs an 8-core, 16-thread Zen 5 / Zen 5c processor with the Radeon 890M integrated graphics, creating a laptop platform that targets a specific balance of CPU throughput and GPU capability without a discrete graphics card. The CPU’s average benchmark score of 52901 places it in the 91st percentile of all CPUs, while the GPU’s average score of 9210 sits in the 45th percentile of all GPUs. This combination yields an overall build percentile of 68, indicating a system that leans heavily on processor performance rather than graphical muscle. The data shows a platform designed for productivity, development, and light-to-moderate creative work, with gaming possible but limited by the integrated GPU’s mid-pack standing. The following analysis breaks down what these scores mean for real workloads, the upgrade path on this socket, and which users should consider this configuration.

CPU Analysis

The AMD Ryzen AI Embedded P164 is built on the Gorgon Point codename, using a 4 nm TSMC process with a die size of 233 mm². It packs 8 cores and 16 threads, with a base clock of 2.00 GHz and a boost clock of 5.00 GHz. The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3 shared across the chip. This is a mobile-market processor with a 28 W TDP, indicating efficiency-focused design rather than raw multi-core dominance. The architecture is Zen 5 / Zen 5c, a hybrid arrangement that likely mixes full-performance cores with compact cores, though the FACT PACK does not specify the core distribution.

Benchmark results reveal a strong single-threaded performer. The PassMark single-thread score is 4029, which is a healthy figure for a mobile chip and suggests responsive everyday use, snappy application launches, and solid performance in lightly threaded tasks like office suites or web browsing. The multithread score of 25889 reflects the 8-core/16-thread design, and the average benchmark score of 52901 places it in the 91st percentile of all CPUs. Compared to the nearest rivals, the P164 is essentially neck-and-neck with the AMD Ryzen 5 9500F (avg score 52873, delta 0.1%), the Intel Xeon 634 (avg score 52974, delta -0.1%), the AMD EPYC 7313P (avg score 53206, delta -0.6%), and the AMD Ryzen 9 7900X (avg score 53288, delta -0.7%). This means the P164 sits within 1% of these desktop-class processors, which is remarkable for a 28 W mobile chip, but it also means that in multi-threaded workloads, it is not breaking away from the pack—it is merely matching established desktop parts.

The integer math score of 87940 and floating point math score of 55799 indicate balanced arithmetic performance, useful for data processing and scientific calculations. The extended instructions score of 24193 suggests good SIMD capability, while the data compression score of 327891 is a standout, indicating strong throughput in compression workloads like archiving or database compression. The data encryption score of 16055 is moderate, and the random string sorting score of 34801 points to competent sorting algorithms. The physics score of 1210 is relatively low, which may impact simulation-heavy tasks, but the find prime numbers score of 71 is a niche metric. For real workloads, the CPU excels in single-threaded responsiveness and multi-threaded office-to-creative tasks, but it is not a rendering monster—the multithread score, while decent, does not rival high-core-count desktop parts beyond the ones listed as rivals.

Upgrade Path and Platform

The P164 uses AMD Socket FP8, which is a mobile socket, meaning upgrades are limited to what fits that socket—typically other embedded or mobile Ryzen parts, though the FACT PACK does not list specific compatible CPUs. The platform supports DDR5 and LPDDR5X memory with a dual-channel bus and 89.6 GB/s of memory bandwidth. It also supports ECC memory, which is a notable feature for reliability in workstation or server-like scenarios. PCIe connectivity is Gen 4 with 16 lanes from the CPU, which provides ample bandwidth for NVMe storage or a discrete GPU if the laptop chassis allows it, though the integrated Radeon 880M (note: the GPU section lists the Radeon 890M, not 880M) is the primary graphics solution in this config.

The CPU’s TDP is 28 W, and the GPU’s TDP is 15 W, so the combined power draw is modest. The FACT PACK does not provide a suggested PSU for this laptop build, but given the low TDPs, a standard laptop power brick will suffice. The upgrade path here is not about swapping the CPU—it is about memory and storage. The dual-channel 89.6 GB/s bandwidth is fixed, but users can choose higher-capacity DDR5 or LPDDR5X modules. The PCIe Gen 4 lanes allow for fast NVMe SSDs, which would improve load times and data-heavy tasks. Since the GPU shares system memory (the Radeon 890M has “System Shared” memory), increasing RAM capacity directly benefits graphics performance, as more available memory means less pressure on the shared pool. A sensible next upgrade for this platform would be maximizing RAM capacity and speed, plus adding a fast Gen 4 NVMe drive. A discrete GPU is theoretically possible via the 16 PCIe lanes, but the chassis constraints of a laptop (buildClass is “laptop”) make that impractical in most cases.

Usage Scenarios

High-refresh gaming: The Radeon 890M’s average benchmark score of 9210 places it in the 45th percentile of all GPUs, which is mid-pack. The PassMark G3D score of 8076 and DirectX 12 score of 37 indicate that this GPU is not built for high-refresh 1080p gaming in demanding titles. The data suggests that high-refresh gaming (144 Hz or above) is only viable in esports or older titles at low settings, and even then, frame rates will be modest. The CPU’s strong single-thread score of 4029 helps with frame pacing, but the GPU will be the limiting factor.

Streaming: The CPU’s 8 cores and 16 threads, combined with a multithread score of 25889, provide enough headroom for software encoding at moderate bitrates in 720p or 1080p. The data compression score of 327891 indicates solid throughput for encoding tasks, but the GPU’s lack of dedicated tensor cores (the FACT PACK lists tensorCores as null) means no hardware acceleration for AI-based encoding. Streaming is possible but will tax the CPU, and the overall experience will be better suited to 720p60 rather than 1080p60.

Video editing: The CPU’s single-thread score of 4029 and multithread score of 25889 support timeline scrubbing and export times that are competitive with desktop parts like the Ryzen 9 7900X (avg score 53288, delta -0.7%). The integer and floating point math scores (87940 and 55799) indicate solid processing for codecs and effects. However, the GPU’s 45th percentile standing means GPU-accelerated effects and rendering will be slow. Light 1080p editing is feasible; 4K editing will struggle without a discrete GPU.

3D rendering: The CPU’s multithread score of 25889 is decent for CPU-based rendering, but it is only 0.1% ahead of the Ryzen 5 9500F and 0.7% behind the Ryzen 9 7900X, so expect rendering times similar to those parts. The GPU’s FP32 performance of 5.939 TFLOPS is modest, and the PassMark GPU compute score of 4157 indicates limited GPU rendering acceleration. This platform is not suited for serious 3D rendering workloads; it will handle occasional renders but with long wait times.

Software development: The CPU’s 8 cores and 16 threads, plus a single-thread score of 4029, make it excellent for compilation, code analysis, and running virtual machines. The data encryption score of 16055 and extended instructions score of 24193 support cryptographic operations and SIMD-heavy code. The 91st percentile CPU ranking means it outperforms most laptops in developer tasks, and the 28 W TDP allows for sustained load without thermal throttling in a well-designed chassis.

Student and office work: The CPU’s single-thread performance and 89.6 GB/s memory bandwidth handle spreadsheets, document editing, and web browsing with ease. The GPU’s G2D score of 979 suggests adequate 2D acceleration for office applications. The low TDP (28 W CPU, 15 W GPU) means long battery life, though the FACT PACK does not specify battery capacity. This is a capable machine for everyday academic and professional work.

Balance and Bottleneck

The data shows a clear imbalance: the CPU is a top-tier performer (91st percentile), while the GPU is mid-pack (45th percentile). This means the GPU will bottleneck in any graphics-intensive workload, including gaming and GPU-accelerated rendering. The CPU’s average score of 52901 is within 0.7% of the Ryzen 9 7900X, a desktop part, but the GPU’s average score of 9210 is within 1% of the GTX 960 (avg score 9273, delta -0.7%) and GTX 465 (avg score 9294, delta -0.9%), both ancient discrete GPUs. In gaming, the CPU can feed frames faster than the GPU can render them, so frame rates will be GPU-limited. In CPU-heavy tasks like compilation or data compression, the system performs well, but any task that offloads to the GPU will see a significant drop in performance relative to the CPU’s capability. The PassMark DirectX 11 score of 73 and DirectX 12 score of 37 highlight the GPU’s limitations in modern API workloads. The pixel rate of 92.80 GPixel/s and texture rate of 185.6 GTexel/s are low for modern gaming, confirming that the GPU is the limiting factor in graphics-bound scenarios. For productivity, the CPU carries the load, and the GPU is only relevant for display output and light acceleration.

Gaming Performance

No measured FPS data exists for this exact CPU-GPU combination—the FACT PACK contains no measuredFps rows. Therefore, all frame rate figures here are estimates based on the benchmark scores. The Radeon 890M’s average score of 9210 (45th percentile) suggests that at 1080p with ultra settings, most modern games will run below 30 FPS. The PassMark DirectX 12 score of 37 is particularly low, indicating poor performance in DX12 titles. At 720p with low settings, older or esports titles may achieve 30-60 FPS, but the data does not support high-refresh gaming. The CPU’s strong single-thread score (4029) helps with CPU-bound scenarios, but the GPU’s 5.939 TFLOPS FP32 performance caps overall graphical throughput. For 1080p ultra, expect playable frame rates only in very light games (e.g., 2D titles or older indie games). The GPU’s memory is system-shared, so bandwidth depends on the installed RAM speed, but the 89.6 GB/s CPU memory bandwidth is the ceiling. In practice, this platform is best suited for 720p gaming at medium settings, with 1080p limited to low settings in less demanding titles.

Who Should Build It

This platform targets users who prioritize CPU performance over GPU capability. The 91st percentile CPU ranking makes it ideal for software developers who compile code, run containers, or use virtual machines—the 8 cores and 16 threads with a 28 W TDP provide excellent multi-threaded throughput without excessive power draw. Students and office workers benefit from the strong single-thread score (4029) for everyday tasks and the ECC memory support for data integrity in research or accounting work. Content creators who work primarily in CPU-bound tasks like video encoding (using software codecs) or data compression will find the multithread score of 25889 and data compression score of 327891 useful, but they must accept GPU limitations for effects and rendering. Small business workstations that need reliable computing for spreadsheets, databases, and light server duties will appreciate the ECC memory and 89.6 GB/s bandwidth. Gamers should look elsewhere unless they are content with 720p low-settings play, as the GPU’s 45th percentile ranking is insufficient for modern titles at mainstream resolutions. The buildClass is “laptop,” so this is a mobile solution for professionals who need carry-on performance without a discrete GPU.

Benchmark Performance

The CPU’s average benchmark score is 52901, placing it in the 91st percentile of all CPUs. Its nearest rival is the AMD Ryzen 5 9500F with an avg score of 52873, a delta of 0.1%, meaning the P164 is marginally faster. The Intel Xeon 634 scores 52974 (delta -0.1%), so the P164 is slightly slower. The AMD EPYC 7313P scores 53206 (delta -0.6%) and the AMD Ryzen 9 7900X scores 53288 (delta -0.7%), so the P164 trails these by less than 1%. This places the P164 in the same performance class as these desktop parts, though it does so at a 28 W TDP. The GPU’s average benchmark score is 9210, placing it in the 45th percentile of all GPUs. Its nearest rivals include the AMD Radeon Vega 8 (avg 9221, delta -0.1%), NVIDIA GeForce GTX 960 (avg 9273, delta -0.7%), NVIDIA GeForce GTX 465 (avg 9294, delta -0.9%), and NVIDIA GeForce GTX 850M (avg 9302, delta -1%). The GPU is slightly slower than all of these, meaning it performs like a mid-range part from nearly a decade ago. The combined percentile is 68, reflecting the CPU’s high standing offset by the GPU’s mediocrity. The PassMark GPU compute score of 4157 and Geekbench OpenCL score of 37254 indicate limited compute capability, while the 3DMark Steel Nomad DX12 score of 590 is low for modern gaming. The overall picture is a system that excels at CPU-bound workloads but falls short in any GPU-accelerated task.

FAQ

Q: What is the CPU’s performance relative to desktop parts?

A: The CPU’s average benchmark score of 52901 puts it in the 91st percentile, and it is within 0.7% of the AMD Ryzen 9 7900X (avg score 53288, delta -0.7%) and 0.1% ahead of the AMD Ryzen 5 9500F (avg score 52873, delta 0.1%).

Q: Does this platform support ECC memory?

A: Yes, the CPU supports ECC memory, which is beneficial for data integrity in workstation or server-like applications.

Q: What is the GPU’s performance compared to discrete graphics cards?

A: The Radeon 890M’s average score of 9210 places it in the 45th percentile, and it is slightly slower than the NVIDIA GeForce GTX 960 (avg score 9273, delta -0.7%) and GTX 465 (avg score 9294, delta -0.9%).

Q: Can this system handle 4K video editing?

A: The CPU’s multithread score of 25889 supports CPU-heavy editing, but the GPU’s low scores (e.g., PassMark DirectX 12 score of 37) mean GPU-accelerated effects will be slow; 4K editing is likely impractical.

Q: What is the memory bandwidth?

A: The platform offers 89.6 GB/s of memory bandwidth via a dual-channel DDR5 or LPDDR5X bus, which is shared with the GPU since it uses system memory.

Q: Is there a discrete GPU upgrade path?

A: The CPU provides PCIe Gen 4 with 16 lanes, which could theoretically support a discrete GPU, but the “laptop” buildClass makes this impractical in most chassis designs.

Q: What is the overall performance percentile?

A: The combined percentile is 68, reflecting the CPU’s 91st percentile ranking and the GPU’s 45th percentile ranking.

GPU Analysis

The AMD Radeon 890M is an integrated GPU based on the Strix Point chip with RDNA 3.5 architecture, built on a 4 nm TSMC process with 34,000 million transistors and a die size of 233 mm². It has 1024 shading units, 64 texture mapping units, and 32 raster operation units, along with 16 ray tracing cores. The transistor density is 145.9M per mm², and the GPU clocks run from a base of 400 MHz to a boost of 2900 MHz. Memory is system-shared, meaning the size, type, bus width, and bandwidth are all dependent on the CPU’s installed RAM—the FACT PACK lists memory bandwidth as “System Dependent,” but the CPU’s 89.6 GB/s dual-channel bandwidth is the practical ceiling. The GPU’s pixel rate is 92.80 GPixel/s, and its texture rate is 185.6 GTexel/s, with FP32 performance of 5.939 TFLOPS and FP16 also at 5.939 TFLOPS (1:1). The TDP is 15 W, and it uses PCIe 4.0 x8 to connect to the system.

Benchmark results show a GPU that is functional but not powerful. The PassMark G3D score of 8076 and Geekbench Vulkan score of 40808 indicate decent compute for an iGPU, but the 3DMark Steel Nomad DX12 score of 590 is very low, reflecting poor performance in modern DX12 titles. The PassMark DirectX 11 score of 73 and DirectX 12 score of 37 are particularly weak, while the DirectX 10 score of 33 and DirectX 9 score of 97 suggest older APIs run relatively better. The PassMark G2D score of 979 covers 2D tasks, and the GPU compute score of 4157 is modest. The average benchmark score of 9210 places it in the 45th percentile, meaning it outperforms about 45% of all GPUs, but it is slower than the Radeon Vega 8 (avg 9221, delta -0.1%) and GTX 960 (avg 9273, delta -0.7%). The RT cores (16) exist but are unlikely to deliver playable ray-traced frame rates given the overall compute power. The GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so feature support is modern, but raw performance lags. For rendering, the FP32 throughput of 5.939 TFLOPS can handle light 3D work or GPU-accelerated filters in photo editors, but heavy rendering tasks will be slow. The lack of dedicated tensor cores (listed as null) means no AI acceleration beyond what the CPU provides. Overall, this GPU is suitable for office work, light media consumption, and esports at low settings, but it is not a gaming or rendering powerhouse.